| Author: | Wong, Wai Cheung |
| Title: | A study of cold-user problem : from cold start definition to deep learning-based recommendation solution |
| Advisors: | Li, Maggie (COMP) |
| Degree: | M.Sc. |
| Year: | 2021 |
| Subject: | Recommender systems (Information filtering) Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Computing |
| Pages: | ii, 52 pages : color illustrations |
| Language: | English |
| Abstract: | The need for deep learning-based recommendation system has considerably increased with the advent of huge databases, powerful computing, and deep learning algorithms, where an extremely large selection of potential social entertainment, products and services are available. Recommendation system aims to predict the next best action and provide user with personalized recommendations. Collaborative filtering (CF) technique is a mainstream approach in the industry, which utilizes user preferences based on past behaviors, for instance, their ratings. This approach has been facing a challenge of cold start problem (user cold-start and item cold-start) that is a hot topic in the academical field on recommendation system. This dissertation focuses on addressing cold-user problem in collaborative filtering recommendation system from analyzing the definition of cold start to the implementation of a Neural Collaborative Filtering (NCF) model for Top-k recommendation. To handle with cold-user problem in CF, we use opining mining technique to explore explicit aspect from textual data and leverage on dropout neural network approach for model development. Comprehensive experiments on two real-world datasets demonstrating improvements made by our deep learning-based model compare to existing baseline model. Empirical evidence our deep learning-based solutions can enhance the performance on handling cold-user recommendation. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 5863.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 2.88 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/11377