Sean Chen

Master Thesis 

Explainable Recommendation System for Solving Review Loss

Aug 2019 - Jun 2020

• Proposed a review-base recommender system named HANN-Plus, a hierarchical attention neural network to model user’s preference and product’s preference
• HANN-Plus not only can provides rating prediction, but also can be used to generate the representation for making user aware of why such products are recommended.
• Extensive experiments on real-world datasets of Amazon illustrate that HANN-Plus outperforms the state-of-the-art rating prediction methods

Torrance Tests of Creative Thinking (TTCT) 

Sep 2018 ~ Jun 2019

• Built an android application for TTCT (Thinking App), a test of creativity, originally involved simple tests of divergent thinking and other problem-solving skills
• Host MariaDB server to query and update user data daily
• Implemented machine learning for scoring of TTCT: fastText word embedding to initialize user’s data, k nearest neighbors to calculate the similarity between user’s data and ConceptNet from Academia Sinica