In developing countries, acute respiratory infections (ARIs) are responsible for two million deaths per year. Most victims are children who are less than 5 years old. Pneumonia kills 5000 children per day. The statistics for cardiovascular diseases (CVDs) are even more alarming. According to a 2009 report from the World Health Organization (WHO), CVDs kill 17 million people per year. In many resource-poor parts of the world such as India and China, many people are unable to access cardiologists, pulmonologists, and other specialists. Hence, low skilled health professionals are responsible for screening people for ARIs and CVDs in these areas. For example, in the rural areas of the Philippines, there is only one doctor for every 10,000 people. By contrast, the United States has one doctor for every 500 Americans. Due to advances in technology, it is now possible to use a smartphone for audio recording, signal processing, and machine learning. In my thesis, I have developed an Android application named Smart Auscultation. Auscultation is a process in which physicians listen to heart and lung sounds to diagnose disorders. Cardiologists spend years mastering this skill. The Smart Auscultation application is capable of recording and classifying heart sounds, and can be used by public or clinical health workers. This application can detect abnormal heart sounds with up to 92-98% accuracy. In addition, the application can record, but not yet classify, lung sounds. This application will be able to help save thousands of lives by allowing anyone to identify abnormal heart and lung sounds.
Mobile applications are growing in popularity and pose new problems in the area of software testing. In particular, mobile applications heavily depend upon user interactions and a dynamically changing environment of system events. In this thesis, we focus on user-driven events and use Q-learning, a reinforcement machine learning algorithm, to generate tests for Android applications under test (AUT). We implement a framework that automates the generation of GUI test cases by using our Q-learning approach and compare it to a uniform random (UR) implementation. A novel feature of our approach is that we generate user-driven event sequences through the GUI, without the source code or the model of the AUT. Hence, considerable amount of cost and time are saved by avoiding the need for model generation for generating the tests. Our results show that the systematic path exploration used by Q-learning results in higher average code coverage in comparison to the uniform random approach.
In recent years, brain computer interfaces (BCIs) have gained popularity in non-medical domains such as the gaming, entertainment, personal health, and marketing industries. A growing number of companies offer various inexpensive consumer grade BCIs and some of these companies have recently introduced the concept of BCI "App stores" in order to facilitate the expansion of BCI applications and provide software development kits (SDKs) for other developers to create new applications for their devices. The BCI applications access to users' unique brainwave signals, which consequently allows them to make inferences about users' thoughts and mental processes. Since there are no specific standards that govern the development of BCI applications, its users are at the risk of privacy breaches. In this work, we perform first comprehensive analysis of BCI App stores including software development kits (SDKs), application programming interfaces (APIs), and BCI applications w.r.t privacy issues. The goal is to understand the way brainwave signals are handled by BCI applications and what threats to the privacy of users exist. Our findings show that most applications have unrestricted access to users' brainwave signals and can easily extract private information about their users without them even noticing. We discuss potential privacy threats posed by current practices used in BCI App stores and then describe some countermeasures that could be used to mitigate the privacy threats. Also, develop a prototype which gives the BCI app users a choice to restrict their brain signal dynamically.
This work describes an approach to determine whether people participate in the events they tweet about. Specifically, we determine whether people are participants in events with respect to the tweet timestamp. We target all events expressed by verbs in tweets, including past, present and events that may occur in future. We define event participant as people directly involved in an event regardless of whether they are the agent, recipient or play another role. We present an annotation effort, guidelines and quality analysis with 1,096 event mentions. We discuss the label distributions and event behavior in the annotated corpus. We also explain several features used and a standard supervised machine learning approach to automatically determine if and when the author is a participant of the event in the tweet. We discuss trends in the results obtained and devise important conclusions.
This work presents an investigation of a piezoelectric sensor based energy harvesting system, which collects energy from the surrounding environment. Increasing costs and scarcity of fossil fuels is a great concern today for supplying power to electronic devices. Furthermore, generating electricity by ordinary methods is a complicated process. Disposal of chemical batteries and cables is polluting the nature every day. Due to these reasons, research on energy harvesting from renewable resources has become mandatory in order to achieve improved methods and strategies of generating and storing electricity. Many low power devices being used in everyday life can be powered by harvesting energy from natural energy resources. Power overhead and power energy efficiency is of prime concern in electronic circuits. In this work, an energy harvester is modeled and simulated in Simscape™ for the functional analysis and comparison of achieved outcomes with previous work. Results demonstrate that the harvester produces power in the 0 μW to 100 μW range, which is an adequate amount to provide supply to low power devices. Power efficiency calculations also demonstrate that the implemented harvester is capable of generating and storing power for low power pervasive applications.
In recent years, social media platforms such as Twitter and Facebook have emerged as effective tools for broadcasting messages worldwide during disaster events. With millions of messages posted through these services during such events, it has become imperative to identify valuable information that can help the emergency responders to develop effective relief efforts and aid victims. Many studies implied that the role of social media during disasters is invaluable and can be incorporated into emergency decision-making process. However, due to the "big data" nature of social media, it is very labor-intensive to employ human resources to sift through social media posts and categorize/classify them as useful information. Hence, there is a growing need for machine intelligence to automate the process of extracting useful information from the social media data during disaster events. This dissertation addresses the following questions: In a social media stream of messages, what is the useful information to be extracted that can help emergency response organizations to become more situationally aware during and following a disaster? What are the features (or patterns) that can contribute to automatically identifying messages that are useful during disasters? We explored a wide variety of features in conjunction with supervised learning algorithms to automatically identify messages that are useful during disaster events. The feature design includes sentiment features to extract the geo-mapped sentiment expressed in tweets, as well as tweet-content and user detail features to predict the likelihood of the information contained in a tweet to be quickly spread in the network. Further experimentation is carried out to see how these features help in identifying the informative tweets and filter out those tweets that are conversational in nature.
This thesis describes an approach for object recognition using the chordiogram shape-based descriptor. Global shape representations are highly susceptible to clutter generated due to the background or other irrelevant objects in real-world images. To overcome the problem, we aim to extract precise object shape using superpixel segmentation, perceptual grouping, and connected components. The employed shape descriptor chordiogram is based on geometric relationships of chords generated from the pairs of boundary points of an object. The chordiogram descriptor applies holistic properties of the shape and also proven suitable for object detection and digit recognition mechanisms. Additionally, it is translation invariant and robust to shape deformations. In spite of such excellent properties, chordiogram is not scale-invariant. To this end, we propose scale invariant chordiogram descriptors and intend to achieve a similar performance before and after applying scale invariance. Our experiments show that we achieve similar performance with and without scale invariance for silhouettes and real world object images. We also show experiments at different scales to confirm that we obtain scale invariance for chordiogram.
The use of electroencephalogram (EEG), an electrophysiological monitoring method for recording the brain activity, for authentication has attracted the interest of researchers for over a decade. In addition to exhibiting qualities of biometric-based authentication, they are revocable, impossible to mimic, and resistant to coercion attacks. However, EEG signals carry a wealth of information about an individual and can reveal private information about the user. This brings significant privacy issues to EEG-based authentication systems as they have access to raw EEG signals. This thesis proposes a privacy-preserving EEG-based authentication system that preserves the privacy of the user by not revealing the raw EEG signals while allowing the system to authenticate the user accurately. In that, perceptual hashing is utilized and instead of raw EEG signals, their perceptually hashed values are used in the authentication process. In addition to describing the authentication process, algorithms to compute the perceptual hash are developed based on two feature extraction techniques. Experimental results show that an authentication system using perceptual hashing can achieve performance comparable to a system that has access to raw EEG signals if enough EEG channels are used in the process. This thesis also presents a security analysis to show that perceptual hashing can prevent information leakage.
Managing large-scale dynamical systems (e.g., transportation systems, complex information systems, and power networks, etc.) in real-time is very challenging considering their complicated system dynamics, intricate network interactions, large scale, and especially the existence of various uncertainties. To address this issue, intelligent techniques which can quickly design decision-making strategies that are robust to uncertainties are needed. This dissertation aims to conquer these challenges by exploring a data-driven decision-making framework, which leverages big-data techniques and scalable uncertainty evaluation approaches to quickly solve optimal control problems. In particular, following techniques have been developed along this direction: 1) system modeling approaches to simplify the system analysis and design procedures for multiple applications; 2) effective simulation and analytical based approaches to efficiently evaluate system performance and design control strategies under uncertainty; and 3) big-data techniques that allow some computations of control strategies to be completed offline. These techniques and tools for analysis, design and control contribute to a wide range of applications including air traffic flow management, complex information systems, and airborne networks.
The history of mitigation programs to contain vector-borne diseases is a story of successes and failures. Due to the complex interplay among multiple factors that determine disease dynamics, the general principles for timely and specific intervention for incidence reduction or eradication of life-threatening diseases has yet to be determined. This research discusses computational methods developed to assist in the understanding of complex relationships affecting vector-borne disease dynamics. A computational framework to assist public health practitioners with exploring the dynamics of vector-borne diseases, such as malaria and dengue in homogenous and heterogeneous populations, has been conceived, designed, and implemented. The framework integrates a stochastic computational model of interactions to simulate horizontal disease transmission. The intent of the computational modeling has been the integration of stochasticity during simulation of the disease progression while reducing the number of necessary interactions to simulate a disease outbreak. While there are improvements in the computational time reducing the number of interactions needed for simulating disease dynamics, the realization of interactions can remain computationally expensive. Using multi-threading technology to improve performance upon the original computational model, multi-threading experimental results have been tested and reported. In addition, to the contact model, the modeling of biological processes specific to the corresponding pathogen-carrier vector to increase the specificity of the vector-borne disease has been integrated. Last, automation for requesting, retrieving, parsing, and storing specific weather data and geospatial information from federal agencies to study the differences between homogenous and heterogeneous populations has been implemented.
Network security is a complex field that is handled by security professionals who need certain expertise and experience to configure security systems. With the ever increasing size of the networks, managing them is going to be a daunting task. What kind of solution can be used to generate effective security configurations by both security professionals and nonprofessionals alike? In this thesis, a web tool is developed to simplify the process of configuring security systems by translating direct human language input into meaningful, working security rules. These human language inputs yield the security rules that the individual wants to implement in their network. The human language input can be as simple as, "Block Facebook to my son's PC". This tool will translate these inputs into specific security rules and install the translated rules into security equipment such as virtualized Cisco FWSM network firewall, Netfilter host-based firewall, and Snort Network Intrusion Detection. This tool is implemented and tested in both a traditional network and a cloud environment. One thousand input policies were collected from various users such as staff from UNT departments' and health science, including individuals with network security background as well as students with a non-computer science background to analyze the tool's performance. The tool is tested for its accuracy (91%) in generating a security rule. It is also tested for accuracy of the translated rule (86%) compared to a standard rule written by security professionals. Nevertheless, the network security tool built has shown promise to both experienced and inexperienced people in network security field by simplifying the provisioning process to result in accurate and effective network security rules.
Students often use the web as a source of help for problems that they encounter on programming assignments.In this work, we seek to understand how students use the web to search for help on their assignments.We used a mixed methods approach with 344 students who complete a survey and 41 students who participate in a focus group meetings and helped in recording data about their search habits.The survey reveals data about student reported search habits while the focus group uses a web browser plug-in to record actual search patterns.We examine the results collectively and as broken down by class year.Survey results show that at least 2/3 of the students from each class year rely on search engines to locate resources for help with their programming bugs in at least half of their assignments;search habits vary by class year;and the value of different types of resources such as tutorials and forums varies by class year.Focus group results exposes the high frequency web sites used by the students in solving their programming assignments.
Did you stand at a door with a bunch of keys and tried to find the right one to unlock the door? Did you hold a flower and wonder the name of it? A need of object recognition could rise anytime and any where in our daily lives. With the development of mobile devices object recognition applications become possible to provide immediate assistance. However, performing complex tasks in even the most advanced mobile platforms still faces great challenges due to the limited computing resources and computing power. In this thesis, we present an object recognition system that resides and executes within a mobile device, which can efficiently extract image features and perform learning and classification. To account for the computing constraint, a novel feature extraction method that minimizes the data size and maintains data consistency is proposed. This system leverages principal component analysis method and is able to update the trained classifier when new examples become available . Our system relieves users from creating a lot of examples and makes it user friendly. The experimental results demonstrate that a learning method trained with a very small number of examples can achieve recognition accuracy above 90% in various acquisition conditions. In addition, the system is able to perform learning efficiently.
There are several types of disorders that affect our colon’s ability to function properly such as colorectal cancer, ulcerative colitis, diverticulitis, irritable bowel syndrome and colonic polyps. Automatic detection of these diseases would inform the endoscopist of possible sub-optimal inspection during the colonoscopy procedure as well as save time during post-procedure evaluation. But existing systems only detects few of those disorders like colonic polyps. In this dissertation, we address the automatic detection of another important disorder called ulcerative colitis. We propose a novel texture feature extraction technique to detect the severity of ulcerative colitis in block, image, and video levels. We also enhance the current informative frame filtering methods by detecting water and bubble frames using our proposed technique. Our feature extraction algorithm based on accumulation of pixel value difference provides better accuracy at faster speed than the existing methods making it highly suitable for real-time systems. We also propose a hybrid approach in which our feature method is combined with existing feature method(s) to provide even better accuracy. We extend the block and image level detection method to video level severity score calculation and shot segmentation. Also, the proposed novel feature extraction method can detect water and bubble frames in colonoscopy videos with very high accuracy in significantly less processing time even when clustering is used to reduce the training size by 10 times.
In recent years, the collection of genomic data has skyrocketed and databases of genomic data are growing at a faster rate than ever before. Although many computational methods have been developed to interpret these data, they tend to struggle to process the ever increasing file sizes that are being produced and fail to take advantage of the advances in multi-core processors by using parallel processing. In some instances, loss of accuracy has been a necessary trade off to allow faster computation of the data. This thesis discusses one such algorithm that has been developed and how changes were made to allow larger input file sizes and reduce the time required to achieve a result without sacrificing accuracy. An information entropy based algorithm was used as a basis to demonstrate these techniques. The algorithm dissects the distinctive patterns underlying genomic data efficiently requiring no a priori knowledge, and thus is applicable in a variety of biological research applications. This research describes how parallel processing and object-oriented programming techniques were used to process larger files in less time and achieve a more accurate result from the algorithm. Through object oriented techniques, the maximum allowable input file size was significantly increased from 200 mb to 2000 mb. Using parallel processing techniques allowed the program to finish processing data in less than half the time of the sequential version. The accuracy of the algorithm was improved by reducing data loss throughout the algorithm. Finally, adding user-friendly options enabled the program to use requests more effectively and further customize the logic used within the algorithm.
Shadows in indoor scenarios are usually characterized with multiple light sources that produce complex shadow patterns of a single object. Without removing shadow, the foreground object tends to be erroneously segmented. The inconsistent hue and intensity of shadows make automatic removal a challenging task. In this thesis, a dynamic thresholding and transfer learning-based method for removing shadows is proposed. The method suppresses light shadows with a dynamically computed threshold and removes dark shadows using an online learning strategy that is built upon a base classifier trained with manually annotated examples and refined with the automatically identified examples in the new videos. Experimental results demonstrate that despite variation of lighting conditions in videos our proposed method is able to adapt to the videos and remove shadows effectively. The sensitivity of shadow detection changes slightly with different confidence levels used in example selection for classifier retraining and high confidence level usually yields better performance with less retraining iterations.
Publicly available datasets in health science are often large and observational, in contrast to experimental datasets where a small number of data are collected in controlled experiments. Variables' causal relationships in the observational dataset are yet to be determined. However, there is a significant interest in health science to discover and analyze causal relationships from health data since identified causal relationships will greatly facilitate medical professionals to prevent diseases or to mitigate the negative effects of the disease. Recent advances in Computer Science, particularly in Bayesian networks, has initiated a renewed interest for causality research. Causal relationships can be possibly discovered through learning the network structures from data. However, the number of candidate graphs grows in a more than exponential rate with the increase of variables. Exact learning for obtaining the optimal structure is thus computationally infeasible in practice. As a result, heuristic approaches are imperative to alleviate the difficulty of computations. This research provides effective and efficient learning tools for local causal discoveries and novel methods of learning causal structures with a combination of background knowledge. Specifically in the direction of constraint based structural learning, polynomial-time algorithms for constructing causal structures are designed with first-order conditional independence. Algorithms of efficiently discovering non-causal factors are developed and proved. In addition, when the background knowledge is partially known, methods of graph decomposition are provided so as to reduce the number of conditioned variables. Experiments on both synthetic data and real epidemiological data indicate the provided methods are applicable to large-scale datasets and scalable for causal analysis in health data. Followed by the research methods and experiments, this dissertation gives thoughtful discussions on the reliability of causal discoveries computational health science research, complexity, and implications in health science research.
POD (Point of Dispensing)-based emergency response plans involving mass prophylaxis may seem feasible when considering the choice of dispensing points within a region, overall population density, and estimated traffic demands. However, the plan may fail to serve particular vulnerable sub-populations, resulting in access disparities during emergency response. Federal authorities emphasize on the need to identify sub-populations that cannot avail regular services during an emergency due to their special needs to ensure effective response. Vulnerable individuals require the targeted allocation of appropriate resources to serve their special needs. Devising schemes to address the needs of vulnerable sub-populations is essential for the effectiveness of response plans. This research focuses on data-driven computational methods to quantify and address vulnerabilities in response plans that require the allocation of targeted resources. Data-driven methods to identify and quantify vulnerabilities in response plans are developed as part of this research. Addressing vulnerabilities requires the targeted allocation of appropriate resources to PODs. The problem of resource allocation to PODs during public health emergencies is introduced and the variants of the resource allocation problem such as the spatial allocation, spatio-temporal allocation and optimal resource subset variants are formulated. Generating optimal resource allocation and scheduling solutions can be computationally hard problems. The application of metaheuristic techniques to find near-optimal solutions to the resource allocation problem in response plans is investigated. A vulnerability analysis and resource allocation framework that facilitates the demographic analysis of population data in the context of response plans, and the optimal allocation of resources with respect to the analysis are described.
Optical character recognition (OCR) software has advanced greatly in recent years. Machine-printed text can be scanned and converted to searchable text with word accuracy rates around 98%. Reasonably neat hand-printed text can be recognized with about 85% word accuracy. However, cursive handwriting still remains a challenge, with state-of-the-art performance still around 75%. Algorithms based on hidden Markov models have been only moderately successful, while recurrent neural networks have delivered the best results to date. This thesis explored the feasibility of using a special type of feedforward neural network to convert freeform cursive handwriting to searchable text. The hidden nodes in this network were grouped into clusters, with each cluster being trained to recognize a unique character bigram. The network was trained on writing samples that were pre-segmented and annotated. Post-processing was facilitated in part by using the network to identify overlapping bigrams that were then linked together to form words and sentences. With dictionary assisted post-processing, the network achieved word accuracy of 66.5% on a small, proprietary corpus. The contributions in this thesis are threefold: 1) the novel clustered architecture of the feed-forward neural network, 2) the development of an expanded set of observers combining image masks, modifiers, and feature characterizations, and 3) the use of overlapping bigrams as the textual working unit to assist in context analysis and reconstruction.
Trusted Computing capability has become ubiquitous these days, and it is being widely deployed into consumer devices as well as enterprise platforms. As the number of threats is increasing at an exponential rate, it is becoming a daunting task to secure the systems against them. In this context, the software integrity measurement at runtime with the support of trusted platforms can be a better security strategy. Trusted Computing devices like TPM secure the evidence of a breach or an attack. These devices remain tamper proof if the hardware platform is physically secured. This type of trusted security is crucial for forensic analysis in the aftermath of a breach. The advantages of trusted platforms can be further leveraged if they can be used wisely. RADIUM (Race-free on-demand Integrity Measurement Architecture) is one such architecture, which is built on the strength of TPM. RADIUM provides an asynchronous root of trust to overcome the TOC condition of DRTM. Even though the underlying architecture is trusted, attacks can still compromise applications during runtime by exploiting their vulnerabilities. I propose an application-level integrity measurement solution that fits into RADIUM, to expand the trusted computing capability to the application layer. This is based on the concept of program invariants that can be used to learn the correct behavior of an application. I used Daikon, a tool to obtain dynamic likely invariants, and developed a method of observing these properties at runtime to verify the integrity. The integrity measurement component was implemented as a Python module on top of Volatility, a virtual machine introspection tool. My approach is a first step towards integrity attestation, using hypervisor-based introspection on RADIUM and a proof of concept of application-level measurement capability.
Computer security attacks take place due to the presence of vulnerabilities and bugs in software applications. Bugs and vulnerabilities are the result of weak software architecture and lack of standard software development practices. Despite the fact that software companies are investing millions of dollars in the research and development of software designs security risks are still at large. In some cases software applications are found to carry vulnerabilities for many years before being identified. A recent such example is the popular Heart Bleed Bug in the Open SSL/TSL. In today’s world, where new software application are continuously being developed for a varied community of users; it’s highly unlikely to have software applications running without flaws. Attackers on computer system securities exploit these vulnerabilities and bugs and cause threat to privacy without leaving any trace. The most critical vulnerabilities are those which are related to the integrity of the software applications. Because integrity is directly linked to the credibility of software application and data it contains. Here I am giving solution of maintaining web applications integrity running on RADIUM by using daikon. Daikon generates invariants, these invariants are used to maintain the integrity of the web application and also check the correct behavior of web application at run time on RADIUM architecture in case of any attack or malware. I used data invariants and program flow invariants in my solution to maintain the integrity of web-application against such attack or malware. I check the behavior of my proposed invariants at run-time using Lib-VMI/Volatility memory introspection tool. This is a novel approach and proof of concept toward maintaining web application integrity on RADIUM.
The basis of today’s security systems is the trust and confidence that the system will behave as expected and are in a known good trusted state. The trust is built from hardware and software elements that generates a chain of trust that originates from a trusted known entity. Leveraging hardware, software and a mandatory access control policy technology is needed to create a trusted measurement environment. Employing a control layer (hypervisor or microkernel) with the ability to enforce a fine grained access control policy with hyper call granularity across multiple guest virtual domains can ensure that any malicious environment to be contained. In my research, I propose the use of radium's Asynchronous Root of Trust Measurement (ARTM) capability incorporated with a secure mandatory access control policy engine that would mitigate the limitations of the current hardware TPM solutions. By employing ARTM we can leverage asynchronous use of boot, launch, and use with the hypervisor proving its state and the integrity of the secure policy. My solution is using Radium (Race free on demand integrity architecture) architecture that will allow a more detailed measurement of applications at run time with greater semantic knowledge of the measured environments. Radium incorporation of a secure access control policy engine will give it the ability to limit or empower a virtual domain system. It can also enable the creation of a service oriented model of guest virtual domains that have the ability to perform certain operations such as introspecting other virtual domain systems to determine the integrity or system state and report it to a remote entity.
One of the challenges fairly common in motivational interviewing is patient resistance to health behavior change. Hence, automated dialog systems aimed at counseling patients need to be capable of detecting resistance and appropriately altering dialog. This thesis focusses primarily on the development of such a system for automatic identification of patient resistance to behavioral change. This enables the dialogue system to direct the discourse towards a more agreeable ground and helping the patient overcome the obstacles in his or her way to change. This thesis also proposes a dialogue system framework for health behavior change via natural language analysis and generation. The proposed framework facilitates automated motivational interviewing from clinical psychology and involves three broad stages: rapport building and health topic identification, assessment of the patient’s opinion about making a change, and developing a plan. Using this framework patients can be encouraged to reflect on the options available and choose the best for a healthier life.
Email connects 85% of the world. This paper explores the pattern of information overload encountered by majority of email users and examine what steps key email providers are taking to combat the problem. Besides fighting spam, popular email providers offer very limited tools to reduce the amount of unwanted incoming email. Rather, there has been a trend to expand storage space and aid the organization of email. Storing email is very costly and harmful to the environment. Additionally, information overload can be detrimental to productivity. We propose a simple solution that results in drastic reduction of unwanted mail, also known as graymail.
We have a huge amount of video data from extensively available surveillance cameras and increasingly growing technology to record the motion of a moving object in the form of trajectory data. With proliferation of location-enabled devices and ongoing growth in smartphone penetration as well as advancements in exploiting image processing techniques, tracking moving objects is more flawlessly achievable. In this work, we explore some domain-independent qualitative and quantitative features in raw trajectory (spatio-temporal) data in videos captured by a fixed single wide-angle view camera sensor in outdoor areas. We study the efficacy of those features in classifying four basic high level actions by employing two supervised learning algorithms and show how each of the features affect the learning algorithms’ overall accuracy as a single factor or confounded with others.
Data is everywhere. The current Technological advancements in Digital, Social media and the ease at which the availability of different application services to interact with variety of systems are causing to generate tremendous volumes of data. Due to such varied services, Data format is now not restricted to only structure type like text but can generate unstructured content like social media data, videos and images etc. The generated Data is of no use unless been stored and analyzed to derive some Value. Traditional Database systems comes with limitations on the type of data format schema, access rates and storage sizes etc. Hadoop is an Apache open source distributed framework that support storing huge datasets of different formatted data reliably on its file system named Hadoop File System (HDFS) and to process the data stored on HDFS using MapReduce programming model. This thesis study is about building a Data Architecture using Hadoop and its related open source distributed frameworks to support a Data flow pipeline on a low commodity hardware. The Data flow components are, sourcing data, storage management on HDFS and data access layer. This study also discuss about a use case to utilize the architecture components. Sqoop, a framework to ingest the structured data from database onto Hadoop and Flume is used to ingest the semi-structured Twitter streaming json data on to HDFS for analysis. The data sourced using Sqoop and Flume have been analyzed using Hive for SQL like analytics and at a higher level of data access layer, Hadoop has been compared with an in memory computing system using Spark. Significant differences in query execution performances have been analyzed when working with Hadoop and Spark frameworks. This integration helps for ingesting huge Volumes of streaming json Variety data to derive better Value based analytics using Hive and ...
Segmentation, recognition and 3D reconstruction of objects have been cutting-edge research topics, which have many applications ranging from environmental and medical to geographical applications as well as intelligent transportation. In this dissertation, I focus on the study of segmentation, recognition and 3D reconstruction of objects using LiDAR data/MRI. Three main works are that (I). Feature extraction algorithm based on sparse LiDAR data. A novel method has been proposed for feature extraction from sparse LiDAR data. The algorithm and the related principles have been described. Also, I have tested and discussed the choices and roles of parameters. By using correlation of neighboring points directly, statistic distribution of normal vectors at each point has been effectively used to determine the category of the selected point. (II). Segmentation and 3D reconstruction of objects based on LiDAR/MRI. The proposed method includes that the 3D LiDAR data are layered, that different categories are segmented, and that 3D canopy surfaces of individual tree crowns and clusters of trees are reconstructed from LiDAR point data based on a region active contour model. The proposed method allows for delineations of 3D forest canopy naturally from the contours of raw LiDAR point clouds. The proposed model is suitable not only for a series of ideal cone shapes, but also for other kinds of 3D shapes as well as other kinds dataset such as MRI. (III). Novel algorithms for recognition of objects based on LiDAR/MRI. Aimed to the sparse LiDAR data, the feature extraction algorithm has been proposed and applied to classify the building and trees. More importantly, the novel algorithms based on level set methods have been provided and employed to recognize not only the buildings and trees, the different trees (e.g. Oak trees and Douglas firs), but also the subthalamus nuclei (STNs). By using the novel algorithms based ...
The extensive use of global teams to develop software has prompted researchers to investigate various factors that can enhance a team’s performance. While a significant body of research exists on global software teams, previous research has not fully explored the interrelationships and collective impact of various factors on team performance. This study explored a model that added the characteristics of a team’s culture, ability, communication frequencies, response rates, and linguistic categories to a central framework of team performance. Data was collected from two student software development projects that occurred between teams located in the United States, Panama, and Turkey. The data was obtained through online surveys and recorded postings of team activities that occurred throughout the global software development projects. Partial least squares path modeling (PLS-PM) was chosen as the analytic technique to test the model and identify the most influential factors. Individual factors associated with response rates and linguistic characteristics proved to significantly affect a team’s activity related to grade on the project, group cohesion, and the number of messages received and sent. Moreover, an examination of possible latent homogeneous segments in the model supported the existence of differences among groups based on leadership style. Teams with assigned leaders tended to have stronger relationships between linguistic characteristics and team performance factors, while teams with emergent leaders had stronger. Relationships between response rates and team performance factors. The contributions in this dissertation are three fold. 1) Novel analysis techniques using PLS-PM and clustering, 2) Use of new, quantifiable variables in analyzing team activity, 3) Identification of plausible causal indicators for team performance and analysis of the same.
As video capturing devices become more ubiquitous from surveillance cameras to smart phones, the demand of automated video analysis is increasing as never before. One obstacle in this process is to efficiently locate where a human operator’s attention should be, and another is to determine the specific types of activities or actions without ambiguity. It is the special interest of this dissertation to locate spatial and temporal regions of interest in videos and to develop a better action representation for video-based activity analysis. This dissertation follows the scheme of “locating then recognizing” activities of interest in videos, i.e., locations of potentially interesting activities are estimated before performing in-depth analysis. Theoretical properties of regions of interest in videos are first exploited, based on which a unifying framework is proposed to locate both spatial and temporal regions of interest with the same settings of parameters. The approach estimates the distribution of motion based on 3D structure tensors, and locates regions of interest according to persistent occurrences of low probability. Two contributions are further made to better represent the actions. The first is to construct a unifying model of spatio-temporal relationships between reusable mid-level actions which bridge low-level pixels and high-level activities. Dense trajectories are clustered to construct mid-level actionlets, and the temporal relationships between actionlets are modeled as Action Graphs based on Allen interval predicates. The second is an effort for a novel and efficient representation of action graphs based on a sparse coding framework. Action graphs are first represented using Laplacian matrices and then decomposed as a linear combination of primitive dictionary items following sparse coding scheme. The optimization is eventually formulated and solved as a determinant maximization problem, and 1-nearest neighbor is used for action classification. The experiments have shown better results than existing approaches for regions-of-interest detection and action ...
High precision data from household travel survey (HTS) is extremely important for the transportation research, traffic models and policy formulation. Traditional methods of data collection were imprecise because they relied on people’s memories of trip information, such as date and location, and the remainder data had to be obtained by certain supplemental tools. The traditional methods suffered from intensive labor, large time consumption, and unsatisfactory data precision. Recent research trends to employ smartphone apps to collect HTS data. In this study, there are two goals to be addressed. First, a smartphone app is developed to realize a smartphone-based method only for data collection. Second, the researcher evaluates whether this method can supply or replace the traditional tools of HTS. Based on this premise, the smartphone app, TravelSurvey, is specially developed and used for this study. TravelSurvey is currently compatible with iPhone 4 or higher and iPhone Operating System (iOS) 6 or higher, except iPhone 6 or iPhone 6 plus and iOS 8. To evaluate the feasibility, eight individuals are recruited to participate in a pilot HTS. Afterwards, seven of them are involved in a semi-structured interview. The interview is designed to collect interviewees’ feedback directly, so the interview mainly concerns the users’ experience of TravelSurvey. Generally, the feedback is positive. In this study, the pilot HTS data is successfully uploaded to the server by the participants, and the interviewees prefer this smartphone-based method. Therefore, as a new tool, the smartphone-based method feasibly supports a typical HTS for data collection.
The purpose of this project is to explore the GPU for general purpose computing. The GPU is a massively parallel computing device that has a high-throughput, exhibits high arithmetic intensity, has a large market presence, and with the increasing computation power being added to it each year through innovations, the GPU is a perfect candidate to complement the CPU in performing computations. The GPU follows the single instruction multiple data (SIMD) model for applying operations on its data. This model allows the GPU to be very useful for assisting the CPU in performing computations on data that is highly parallel in nature. The compute unified device architecture (CUDA) is a parallel computing and programming platform for NVIDIA GPUs. The main focus of this project is to show the power, speed, and performance of a CUDA-enabled GPU for digital video watermark insertion in the H.264 video compression domain. Digital video watermarking in general is a highly computationally intensive process that is strongly dependent on the video compression format in place. The H.264/MPEG-4 AVC video compression format has high compression efficiency at the expense of having high computational complexity and leaving little room for an imperceptible watermark to be inserted. Employing a human visual model to limit distortion and degradation of visual quality introduced by the watermark is a good choice for designing a video watermarking algorithm though this does introduce more computational complexity to the algorithm. Research is being conducted into how the CPU-GPU execution of the digital watermark application can boost the speed of the applications several times compared to running the application on a standalone CPU using NVIDIA visual profiler to optimize the application.
The increasingly popular cloud-computing paradigm provides on-demand access to computing and storage with the appearance of unlimited resources. Users are given access to a variety of data and software utilities to manage their work. Users rent virtual resources and pay for only what they use. In spite of the many benefits that cloud computing promises, the lack of dependability in shared virtualized infrastructures is a major obstacle for its wider adoption, especially for mission-critical applications. Virtualization and multi-tenancy increase system complexity and dynamicity. They introduce new sources of failure degrading the dependability of cloud computing systems. To assure cloud dependability, in my dissertation research, I develop autonomic failure identification and diagnosis techniques that are crucial for understanding emergent, cloud-wide phenomena and self-managing resource burdens for cloud availability and productivity enhancement. We study the runtime cloud performance data collected from a cloud test-bed and by using traces from production cloud systems. We define cloud signatures including those metrics that are most relevant to failure instances. We exploit profiled cloud performance data in both time and frequency domain to identify anomalous cloud behaviors and leverage cloud metric subspace analysis to automate the diagnosis of observed failures. We implement a prototype of the anomaly identification system and conduct the experiments in an on-campus cloud computing test-bed and by using the Google datacenter traces. Our experimental results show that our proposed anomaly detection mechanism can achieve 93% detection sensitivity while keeping the false positive rate as low as 6.1% and outperform other tested anomaly detection schemes. In addition, the anomaly detector adapts itself by recursively learning from these newly verified detection results to refine future detection.
Mobile phone advancements and ubiquitous internet connectivity are resulting in ever expanding possibilities in the application of smart phones. Users of mobile phones are now capable of hosting server applications from their personal devices. Whether providing services individually or in an ad hoc network setting the devices are currently not configured for defending against distributed denial of service (DDoS) attacks. These attacks, often launched from a botnet, have existed in the space of personal computing for decades but recently have begun showing up on mobile devices. Research is done first into the required steps to develop a potential botnet on the Android platform. This includes testing for the amount of malicious traffic an Android phone would be capable of generating for a DDoS attack. On the other end of the spectrum is the need of mobile devices running networked applications to develop security against DDoS attacks. For this mobile, phones are setup, with web servers running Apache to simulate users running internet connected applications for either local ad hoc networks or serving to the internet. Testing is done for the viability of using commonly available modules developed for Apache and intended for servers as well as finding baseline capabilities of mobiles to handle higher traffic volumes. Given the unique challenge of the limited resources a mobile phone can dedicate to Apache when compared to a dedicated hosting server a new method was needed. A proposed defense algorithm is developed for mitigating DDoS attacks against the mobile server that takes into account the limited resources available on the mobile device. The algorithm is tested against TCP socket flooding for effectiveness and shown to perform better than the common Apache module installations on a mobile device.
The promise of service oriented computing, and the availability of Web services promote the delivery and creation of new services based on existing services, in order to meet new demands and new markets. As Web and internet based services move into Clouds, inter-dependency of services and their complexity will increase substantially. There are standards and frameworks for specifying and composing Web Services based on functional properties. However, mechanisms to individually address non-functional properties of services and their compositions have not been well established. Furthermore, the Cloud ontology depicts service layers from a high-level, such as Application and Software, to a low-level, such as Infrastructure and Platform. Each component that resides in one layer can be useful to another layer as a service. It hints at the amount of complexity resulting from not only horizontal but also vertical integrations in building and deploying a composite service. To meet the requirements and facilitate using Web services, we first propose a WSDL extension to permit specification of non-functional or Quality of Service (QoS) properties. On top of the foundation, the QoS-aware framework is established to adapt publicly available tools for Web services, augmented by ontology management tools, along with tools for performance modeling to exemplify how the non-functional properties such as response time, throughput, or utilization of services can be addressed in the service acquisition and composition process. To facilitate Web service composition standards, in this work we extended the framework with additional qualitative information to the service descriptions using Business Process Execution Language (BPEL). Engineers can use BPEL to explore design options, and have the QoS properties analyzed for the composite service. The main issue in our research is performance evaluation in software system and engineering. We researched the Web service computation as the first half of this dissertation, and performance antipattern ...
Future generation cellular networks are expected to deliver an omnipresent broadband access network for an endlessly increasing number of subscribers. Long term Evolution (LTE) represents a significant milestone towards wireless networks known as 4G cellular networks. A key feature of LTE is the implementation of enhanced Radio Resource Management (RRM) mechanism to improve the system performance. The structure of LTE networks was simplified by diminishing the number of the nodes of the core network. Also, the design of the radio protocol architecture is quite unique. In order to achieve high data rate in LTE, 3rd Generation Partnership Project (3GPP) has selected Orthogonal Frequency Division Multiplexing (OFDM) as an appropriate scheme in terms of downlinks. However, the proper scheme for an uplink is the Single-Carrier Frequency Domain Multiple Access due to the peak-to-average-power-ratio (PAPR) constraint. LTE packet scheduling plays a primary role as part of RRM to improve the system’s data rate as well as supporting various QoS requirements of mobile services. The major function of the LTE packet scheduler is to assign Physical Resource Blocks (PRBs) to mobile User Equipment (UE). In our work, we formed a proposed packet scheduler algorithm. The proposed scheduler algorithm acts based on the number of UEs attached to the eNodeB. To evaluate the proposed scheduler algorithm, we assumed two different scenarios based on a number of UEs. When the number of UE is lower than the number of PRBs, the UEs with highest Channel Quality Indicator (CQI) will be assigned PRBs. Otherwise, the scheduler will assign PRBs based on a given proportional fairness metric. The eNodeB’s throughput is increased when the proposed algorithm was implemented.
In many real-world applications, it is common to have uneven number of examples among multiple classes. The data imbalance, however, usually complicates the learning process, especially for the minority classes, and results in deteriorated performance. Boosting methods were proposed to handle the imbalance problem. These methods need elongated training time and require diversity among the classifiers of the ensemble to achieve improved performance. Additionally, extending the boosting method to handle multi-class data sets is not straightforward. Examples of applications that suffer from imbalanced multi-class data can be found in face recognition, where tens of classes exist, and in capsule endoscopy, which suffers massive imbalance between the classes. This dissertation introduces RegBoost, a new boosting framework to address the imbalanced, multi-class problems. This method applies a weighted stratified sampling technique and incorporates a regularization term that accommodates multi-class data sets and automatically determines the error bound of each base classifier. The regularization parameter penalizes the classifier when it misclassifies instances that were correctly classified in the previous iteration. The parameter additionally reduces the bias towards majority classes. Experiments are conducted using 12 diverse data sets with moderate to high imbalance ratios. The results demonstrate superior performance of the proposed method compared to several state-of-the-art algorithms for imbalanced, multi-class classification problems. More importantly, the sensitivity improvement of the minority classes using RegBoost is accompanied with the improvement of the overall accuracy for all classes. With unpredictability regularization, a diverse group of classifiers are created and the maximum accuracy improvement reaches above 24%. Using stratified undersampling, RegBoost exhibits the best efficiency. The reduction in computational cost is significant reaching above 50%. As the volume of training data increase, the gain of efficiency with the proposed method becomes more significant.
Free and fair elections are the basis for democracy, but conducting elections is not an easy task. Different groups of people are trying to influence the outcome of the election in their favor using the range of methods, from campaigning for a particular candidate to well-financed lobbying. Often the stakes are too high, and the methods are illegal. Two main properties of any voting scheme are the privacy of a voter’s choice and the integrity of the tally. Unfortunately, they are mutually exclusive. Integrity requires making elections transparent and auditable, but at the same time, we must preserve a voter’s privacy. It is always a trade-off between these two requirements. Current voting schemes favor privacy over auditability, and thus, they are vulnerable to voting fraud. I propose two novel voting systems that can achieve both privacy and verifiability. The first protocol is based on cryptographical primitives to ensure the integrity of the final tally and privacy of the voter. The second protocol is a simple paper-based voting scheme that achieves almost the same level of security without usage of cryptography.
The spread of infectious diseases has been a public concern throughout human history. Historic recorded data has reported the severity of infectious disease epidemics in different ages. Ancient Greek physician Hippocrates was the first to analyze the correlation between diseases and their environment. Nowadays, health authorities are in charge of planning strategies that guarantee the welfare of citizens. The simulation of contagion scenarios contributes to the understanding of the epidemic behavior of diseases. Computational models facilitate the study of epidemics by integrating disease and population data to the simulation. The use of detailed demographic and geographic characteristics allows researchers to construct complex models that better resemble reality and the integration of these attributes permits us to understand the rules of interaction. The interaction of individuals with similar characteristics forms synthetic structures that depict clusters of interaction. The synthetic environments facilitate the study of the spread of infectious diseases in diverse scenarios. The characteristics of the population and the disease concurrently affect the local and global epidemic progression. Every cluster’ epidemic behavior constitutes the global epidemic for a clustered population. By understanding the correlation between structured populations and the spread of a disease, current dissertation research makes possible to identify risk groups of specific characteristics and devise containment strategies that facilitate health authorities to improve mitigation strategies.
Software applications’ performance is hindered by a variety of factors, but most notably by the well-known CPU-memory speed gap (often known as the memory wall). This results in the CPU sitting idle waiting for data to be brought from memory to processor caches. The addressing used by caches cause non-uniform accesses to various cache sets. The non-uniformity is due to several reasons, including how different objects are accessed by the code and how the data objects are located in memory. Memory allocators determine where dynamically created objects are placed, thus defining addresses and their mapping to cache locations. It is important to evaluate how different allocators behave with respect to the localities of the created objects. Most allocators use a single attribute, the size, of an object in making allocation decisions. Additional attributes such as the placement with respect to other objects, or specific cache area may lead to better use of cache memories. In this dissertation, we proposed and implemented a framework that allows for the development and evaluation of new memory allocation techniques. At the root of the framework is a memory tracing tool called Gleipnir, which provides very detailed information about every memory access, and relates it back to source level objects. Using the traces from Gleipnir, we extended a commonly used cache simulator for generating detailed cache statistics: per function, per data object, per cache line, and identify specific data objects that are conflicting with each other. The utility of the framework is demonstrated with a new memory allocator known as equivalence class allocator. The new allocator allows users to specify cache sets, in addition to object size, where the objects should be placed. We compare this new allocator with two well-known allocators, viz., Doug Lea and Pool allocators.
Emergency Medical Dispatch Protocols are guidelines that a 9-1-1 dispatcher uses to evaluate the nature of emergency, resources to send and the nature of help provided to the 9-1-1 caller. The current Dispatch Protocols are based on voice only call. But the Next Generation 9-1-1 (NG9-1-1) architecture will allow multimedia emergency calls. In this thesis I analyze and model the Emergency Medical Dispatch Protocols for NG9-1-1 architecture. I have identified various technical aspects to improve the NG9-1-1 Dispatch Protocols. The devices (smartphone) at the caller end have advanced to a point where they can be used to send and receive video, pictures and text. There are sensors embedded in them that can be used for initial diagnosis of the injured person. There is a need to improve the human computer (smartphone) interface to take advantage of technology so that callers can easily make use of various features available to them. The dispatchers at the 9-1-1 call center can make use of these new protocols to improve the quality and the response time. They will have capability of multiple media streams to interact with the caller and the first responders.The specific contributions in this thesis include developing applications that use smartphone sensors. The CPR application uses the smartphone to help administer effective CPR even if the person is not trained. The application makes the CPR process closed loop, i.e., the person who administers the CPR as well as the 9-1-1 operator receive feedback and prompt from the application about the correctness of the CPR. The breathing application analyzes the quality of breathing of the affected person and automatically sends the information to the 9-1-1 operator. In order to improve the Human Computer Interface at the caller and the operator end, I have analyzed Fitts law and extended it so that it ...
Ambiguity is inherent to human language. In particular, word sense ambiguity is prevalent in all natural languages, with a large number of the words in any given language carrying more than one meaning. Word sense disambiguation is the task of automatically assigning the most appropriate meaning to a polysemous word within a given context. Generally the problem of resolving ambiguity in literature has revolved around the famous quote “you shall know the meaning of the word by the company it keeps.” In this thesis, we investigate the role of context for resolving ambiguity through three different approaches. Instead of using a predefined monolingual sense inventory such as WordNet, we use a language-independent framework where the word senses and sense-tagged data are derived automatically from Wikipedia. Using Wikipedia as a source of sense-annotations provides the much needed solution for knowledge acquisition bottleneck. In order to evaluate the viability of Wikipedia based sense-annotations, we cast the task of disambiguating polysemous nouns as a monolingual classification task and experimented on lexical samples from four different languages (viz. English, German, Italian and Spanish). The experiments confirm that the Wikipedia based sense annotations are reliable and can be used to construct accurate monolingual sense classifiers. It is a long belief that exploiting multiple languages helps in building accurate word sense disambiguation systems. Subsequently, we developed two approaches that recast the task of disambiguating polysemous nouns as a multilingual classification task. The first approach for multilingual word sense disambiguation attempts to effectively use a machine translation system to leverage two relevant multilingual aspects of the semantics of text. First, the various senses of a target word may be translated into different words, which constitute unique, yet highly salient signal that effectively expand the target word’s feature space. Second, the translated context words themselves embed co-occurrence information ...
One in seven people in the world use online social networking for a variety of purposes -- to keep in touch with friends and family, to share special occasions, to broadcast announcements, and more. The majority of society has been bought into this new era of communication technology, which allows everyone on the internet to share information with friends. Since social networking has rapidly become a main form of communication, holes in privacy have become apparent. It has come to the point that the whole concept of sharing information requires restructuring. No longer are online social networks simply technology available for a niche market; they are in use by all of society. Thus it is important to not forget that a sense of privacy is inherent as an evolutionary by-product of social intelligence. In any context of society, privacy needs to be a part of the system in order to help users protect themselves from others. This dissertation attempts to address the lack of privacy management in online social networks by designing models which understand the social science behind how we form social groups and share information with each other. Social relationship strength was modeled using activity patterns, vocabulary usage, and behavioral patterns. In addition, automatic configuration for default privacy settings was proposed to help prevent new users from leaking personal information. This dissertation aims to mobilize a new era of social networking that understands social aspects of human network, and uses that knowledge to honor users' privacy.
Service-oriented architecture enables web services to operate in a loosely-coupled setting and provides an environment for dynamic discovery and use of services over a network using standards such as WSDL, SOAP, and UDDI. Web service has both functional and non-functional characteristics. This thesis work proposes to add QoS descriptions (non-functional properties) to WSDL and compose various services to form a business process. This composition of web services also considers QoS properties along with functional properties and the composed services can again be published as a new Web Service and can be part of any other composition using Composed WSDL.
In recent years there has been growing interest in limitless realism in computer graphics applications. Among those, my foremost concentration falls into the complex physical simulations and modeling with diverse applications for the gaming industry. Different simulations have been virtually successful by replicating the details of physical process. As a result, some were strong enough to lure the user into believable virtual worlds that could destroy any sense of attendance. In this research, I focus on fire simulations and its deformation process towards various virtual objects. In most game engines model loading takes place at the beginning of the game or when the game is transitioning between levels. Game models are stored in large data structures. Since changing or adjusting a large data structure while the game is proceeding may adversely affect the performance of the game. Therefore, developers may choose to avoid procedural simulations to save resources and avoid interruptions on performance. I introduce a process to implement a real-time model deformation while maintaining performance. It is a challenging task to achieve high quality simulation while utilizing minimum resources to represent multiple events in timely manner. Especially in video games, this overwhelming criterion would be robust enough to sustain the engaging player's willing suspension of disbelief. I have implemented and tested my method on a relatively modest GPU using CUDA. My experiments conclude this method gives a believable visual effect while using small fraction of CPU and GPU resources.
How do the content and writing style of people who drink alcohol beverages stand out from non-drinkers? How much information can we learn about a person's alcohol consumption behavior by reading text that they have authored? This thesis attempts to extend the methods deployed in authorship attribution and authorship profiling research into the domain of automatically identifying the human action of drinking alcohol beverages. I examine how a psycholinguistics dictionary (the Linguistics Inquiry and Word Count lexicon, developed by James Pennebaker), together with Kenneth Burke's concept of words as symbols of human action, and James Wertsch's concept of mediated action provide a framework for analyzing meaningful data patterns from the content of blogs written by consumers of alcohol beverages. The contributions of this thesis to the research field are twofold. First, I show that it is possible to automatically identify blog posts that have content related to the consumption of alcohol beverages. And second, I provide a framework and tools to model human behavior through text analysis of blog data.
Computational modeling is of fundamental significance in mapping possible disease spread, and designing strategies for its mitigation. Conventional contact networks implement the simulation of interactions as random occurrences, presenting public health bodies with a difficult trade off between a realistic model granularity and robust design of intervention strategies. Recently, researchers have been investigating the use of agent-based models (ABMs) to embrace the complexity of real world interactions. At the same time, theoretical approaches provide epidemiologists with general optimization models in which demographics are intrinsically simplified. The emerging study of affiliation networks and co-affiliation networks provide an alternative to such trade off. Co-affiliation networks maintain the realism innate to ABMs while reducing the complexity of contact networks into distinctively smaller k-partite graphs, were each partition represent a dimension of the social model. This dissertation studies the optimization of intervention strategies for infectious diseases, mainly distributed in school systems. First, concepts of synthetic populations and affiliation networks are extended to propose a modified algorithm for the synthetic reconstruction of populations. Second, the definition of multi-coaffiliation networks is presented as the main social model in which risk is quantified and evaluated, thereby obtaining vulnerability indications for each school in the system. Finally, maximization of the mitigation coverage and minimization of the overall cost of intervention strategies are proposed and compared, based on centrality measures.
In this research, multi-perspective image registration using LiDAR and visual images was considered. 2D-3D image registration is a difficult task because it requires the extraction of different semantic features from each modality. This problem is solved in three parts. The first step involves detection and extraction of common features from each of the data sets. The second step consists of associating the common features between two different modalities. Traditional methods use lines or orthogonal corners as common features. The third step consists of building the projection matrix. Many existing methods use global positing system (GPS) or inertial navigation system (INS) for an initial estimate of the camera pose. However, the approach discussed herein does not use GPS, INS, or any such devices for initial estimate; hence the model can be used in places like the lunar surface or Mars where GPS or INS are not available. A variation of the method is also described, which does not require strong features from both images but rather uses intensity gradients in the image. This can be useful when one image does not have strong features (such as lines) or there are too many extraneous features.
This thesis explores the classification of emotions in song lyrics, using automatic approaches applied to a novel corpus of 100 popular songs. I use crowd sourcing via Amazon Mechanical Turk to collect line-level emotions annotations for this collection of song lyrics. I then build classifiers that rely on textual features to automatically identify the presence of one or more of the following six Ekman emotions: anger, disgust, fear, joy, sadness and surprise. I compare different classification systems and evaluate the performance of the automatic systems against the manual annotations. I also introduce a system that uses data collected from the social network Twitter. I use the Twitter API to collect a large corpus of tweets manually labeled by their authors for one of the six emotions of interest. I then compare the classification of emotions obtained when training on data automatically collected from Twitter versus data obtained through crowd sourced annotations.
Globally distributed software teams are widespread throughout industry. But finding reliable methods that can properly assess a team's activities is a real challenge. Methods such as surveys and manual coding of activities are too time consuming and are often unreliable. Recent advances in information retrieval and linguistics, however, suggest that automated and/or semi-automated text classification algorithms could be an effective way of finding differences in the communication patterns among individuals and groups. Communication among group members is frequent and generates a significant amount of data. Thus having a web-based tool that can automatically analyze the communication patterns among global software teams could lead to a better understanding of group performance. The goal of this thesis, therefore, is to compare automatic and semi-automatic measures of communication and evaluate their effectiveness in classifying different types of group activities that occur within a global software development project. In order to achieve this goal, we developed a web-based component that can be used to help clean and classify communication activities. The component was then used to compare different automated text classification techniques on various group activities to determine their effectiveness in correctly classifying data from a global software development team project.
Multi-modal image fusion is an active research area with many civilian and military applications. Fusion is defined as strategic combination of information collected by various sensors from different locations or different types in order to obtain a better understanding of an observed scene or situation. Fusion of multi-modal images cannot be completed unless these two modalities are spatially aligned. In this research, I consider two important problems. Multi-modal, multi-perspective image registration and decision level fusion of multi-modal images. In particular, LiDAR and visual imagery. Multi-modal image registration is a difficult task due to the different semantic interpretation of features extracted from each modality. This problem is decoupled into three sub-problems. The first step is identification and extraction of common features. The second step is the determination of corresponding points. The third step consists of determining the registration transformation parameters. Traditional registration methods use low level features such as lines and corners. Using these features require an extensive optimization search in order to determine the corresponding points. Many methods use global positioning systems (GPS), and a calibrated camera in order to obtain an initial estimate of the camera parameters. The advantages of our work over the previous works are the following. First, I used high level-features, which significantly reduce the search space for the optimization process. Second, the determination of corresponding points is modeled as an assignment problem between a small numbers of objects. On the other side, fusing LiDAR and visual images is beneficial, due to the different and rich characteristics of both modalities. LiDAR data contain 3D information, while images contain visual information. Developing a fusion technique that uses the characteristics of both modalities is very important. I establish a decision-level fusion technique using manifold models.
This dialog allows you to filter your current search.
Each of the Years listed note their name and the number of records that will be limited down to if you choose that option.
The list can be sorted by name or the count.
Having trouble finding an option within the list of Years? Start typing and we'll update the list to show only those items that match your needs.
This dialog allows you to filter your current search.
Each of the Department listed note their name and the number of records that will be limited down to if you choose that option.
The list can be sorted by name or the count.