Positive Response Recommendation in Social Media Based on Human Need Analysis

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Defense

Title: "Positive Response Recommendation in Social Media Based on Human 
        Need Analysis"

by

LIN Shuya

Abstract:

A long-standing area of interest in positive psychology is to understand 
how to bring about sustainable happiness. In view of the prevalence of 
"Kua kua qun" (praise groups) phenomenon in recent years, this project 
aims to explore how daily social media interactions might help to boost 
happiness and improve the life quality of users. Based on state-of-art 
human need analysis of affective events, we constructed a model to analyze 
user tweets and reveal underlying needs. Then, leveraging the LDA modeling 
technique and the established HappyDB corpus, our pipeline recommends 
keywords and similar stories that could help followers to construct 
content-rich positive responses that would delight the post owner. 
Potential applications of our work include enhancing user experience in 
online forums and assisting with self-help anxiety and depression 
management.


Date            : 4 May 2020 (Monday)

Time            : 17:00 - 17:40

Zoom Meeting    : https://hkust.zoom.us/j/174667391

Advisor         : Dr. MA Xiaojuan

2nd Reader      : Dr. WONG Raymond Chi-Wing