Mining Web Access Logs Using Graphs for Understanding User Behavior

Recreational cannabis was legalized on Oct 17th, 2018. Internationally, some countries have already legalized or are going to legalize the recreational/medical cannabis products. Unfortunately, there are not much data available for stakeholders to make decisions in a data-driven way. However, the government, industry and community are hungry to obtain statistical data to understand the impact of the legalization and forecast the trends.
Lift & Co built a cannabis data platform to collect consumer product reviews, starting with medical cannabis products since 2014. Today, Lift & Co. have 50,000 active users and 80,000 product reviews. We would like to fully utilize the large volume of data crowdsourced from the community in our platform to find insights on the user cannabis consumption behaviors. The results of this project will benefit the licensed-producers, retailers and Canadian government agencies for more reasonable data-driven decision making. The underlying technical problem that will assist in solving the above-mentioned problem is that of Collaborative Filtering.
User behavior is a tough nut to crack. Users are complicated, unexpected, and impatient. As mobile technology improves, users become more skilled and more knowledgeable about their devices and what they want to gain from them. TBC

Faculty Supervisor:

Apurva Narayan

Student:

Partner:

Lift & Co Corp

Discipline:

Computer science

Sector:

Information and cultural industries

University:

The University of British Columbia - Okanagan

Program: