Twitter is one of the major social media platforms highlighting public opinion. With over 330 million users across the globe, Twitter provides insights into global sentiments on many topics. One can estimate global sentiments towards certain events relating to COVID-19 by analyzing the most common phrases and their related sentiment scores from Twitter API data. This project has compiled the most used trigrams in tweets relating to COVID-19 to calculate sentiment scores for the period from March 22 to August 7, 2020. Another goal of the project is to optimize data collection from Twitter API. Twitter limits access to tweet …
continued below
Situated at the intersection of people, technology, and information, the College of Information's faculty, staff and students invest in innovative research, collaborative partnerships, and student-centered education to serve a global information society. The college offers programs of study in information science, learning technologies, and linguistics.
Twitter is one of the major social media platforms highlighting public opinion. With over 330 million users across the globe, Twitter provides insights into global sentiments on many topics. One can estimate global sentiments towards certain events relating to COVID-19 by analyzing the most common phrases and their related sentiment scores from Twitter API data. This project has compiled the most used trigrams in tweets relating to COVID-19 to calculate sentiment scores for the period from March 22 to August 7, 2020. Another goal of the project is to optimize data collection from Twitter API. Twitter limits access to tweet contents to 900 requests per 15 minutes for unpaid API users. For student data scientists, paying for increased API usage is financially infeasible. So, to deal with the rate limit, the project has written functions using the Tweepy python library to collect Twitter API data. The Pandas library has also been used to sample 139000 tweets from over 300 million. The IEEE Dataset provided sentiment scores for the full population. So, to check the integrity of my sample, I performed a Pearson correlation test between the full dataset and sample data, and got 0.84, showing the sample is representative of the population.
Physical Description
1 p.
Notes
The International Conference on Knowledge Management (ICKM) provides researchers and practitioners from all over the world a forum for discussion and exchange of ideas concerning theoretical and practical aspects of Knowledge Management. ICKM 2020 held December 3-5 as a virtual meeting using Cisco WebEx (technical support from NC Central University, Durham, NC) and hosted by the Knowledge and Information Professional Association (KIPA). The conference theme is “Knowledge Commons in the City of Medicine,” covering the institutional analysis of data and the development of knowledge sharing in any subject area.
16th International Conference on Knowledge Management (ICKM-2020), December 3-5, 2020. Virtual meeting using Cisco WebEx (technical support from NC Central University), Durham, NC, United States,
This poster is part of the following collection of related materials.
International Conference on Knowledge Management (ICKM)
Serving as digital proceedings, this collection includes papers, posters, and slides from invited talks as well as practitioner and sponsor presentations for the annual International Conference on Knowledge Management (ICKM).
