Considering user profile and product information in biased product review classification: A Dual Memory Network Model

Utilized `dual memory network` to classify biased product reviews, one for `user profile` and one for `product information`; Achived `best sentiment analysis accuracy, MAE and RMSE` in Yelp and IMDB datasets Supervised by [Prof. Qin Lu](

Let Computer Vision Be The Eye: "Sense+" For the Visually Impaired

A PWA based on `Computer Vision` technologies powered by `object detection model` built using TensorFlow and Microsoft Cognitive Services; Support `natural language commands and responses`; Collaborated with [Hong Kong Blind Union]( Links: 🔗 [**Website**]( | 🔗 [**PolyU Oustanding Works by Students Archive**](

Automatic Anti-Phishing Content Generation for Hong Kong Police Force

*Work in progress...* Supervised by [Prof. Jiannong Cao](