Developing Machine Learning Model for Disambiguate Pattern Recognition on Social Media
Farahnaz Golrooy Motlagh, Huthaifa A. Al Issa · 2020 International Conference on Computation, Automation and Knowledge Management (ICCAKM) · 2020
Social media data have been shown to be a useful resource for the recognition of novel drug use practices and trends. In a random sample collected using Twitter Application Programming Interface (API), “dabs” related tweets produced high volume data than other keywords related to marijuana concentrate in all the states that have legalized for recreational use or for medical use. The key objectives of this research are: 1-Describing a development model of a Machine Learning (ML) classifier to identify “dabs“ that are related to marijuana concentrates and evaluate its performance. 2-Examining dabs-related data between March 2016 and June 2017 after application of the ML classifier model. 3-Analyzing the trends on the number of relevant tweets, number of tweets with retweets, and unique users.