Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Aleah Archibald - North Carolina A&T State University
Social media is a significant tool for individuals to express their opinions and emotions. One of the top used social media is Twitter, which is used by over 300 million peoples monthly in 2017. Researchers are devoted to understanding the characteristics of twitter users and their messages for cyber security and policing. This project aims to investigate the characteristics of tweet messages using text mining techniques and keyword strategies. Specifically, our objectives are the identification of keyword strategy association between daily life and drug substances and the time series pattern of frequencies of keyword usage. Three analytical steps achieve this study: data extraction, data cleanup, and data analysis via API and text mining tools in R software. Analytical results will be presented in word cloud, association analysis, network analysis, and temporal analysis. Our results of 48,000 tweets show that each keyword is associated with either drug substances and/or daily life, which was identifiable via association and network analyses. The temporal pattern shows that twitter users tend to tweet selected keywords the most frequently between 7 p.m. and 10 p.m. This tweet analysis would help aid to cyber policing in knowing various features of keywords and temporal pattern.
Not SubmittedFunder Acknowledgement(s): National Science Foundation
Faculty Advisor: Dr. Seongtae Kim, skim@ncat.edu
Role: Analyzing tweet messages from twitter the refer to drug substances