Recommendation systems based on extreme multi-label classification
This project aims to implement a recommender system using extreme multi-label classification algorithms. In the era of big data, traditional recommender systems are unable to keep up with the scale and size of data available. Extreme multi-label classification can tag a given target with multiple la...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/149716 |
| _version_ | 1826109540625547264 |
|---|---|
| author | Chua, Song Ann |
| author2 | Lihui CHEN |
| author_facet | Lihui CHEN Chua, Song Ann |
| author_sort | Chua, Song Ann |
| collection | NTU |
| description | This project aims to implement a recommender system using extreme multi-label classification algorithms. In the era of big data, traditional recommender systems are unable to keep up with the scale and size of data available. Extreme multi-label classification can tag a given target with multiple labels that are most relevant to it from an extremely large dataset of labels. This report summarises the design implementation and empirical studies of extreme multi-label classification algorithms for recommendation systems on the MovieLens 1M benchmark dataset. This project studied 2 tree-based extreme multi-label classification algorithms, FastXML and AttentionXML, and implemented them using Python for a movie recommender system. This was to investigate the reformulation of the recommender problem as a multi-label classification task. The dataset was prepared such that each item that can be recommended by the system was treated as a unique label that can be tagged to a user by the classifier. The 2 algorithms were compared based on accuracy as well as computational resources required. The accuracy of AttentionXML was 46.6%, 5% larger than that of FastXML’s accuracy of 41.4%. However, FastXML had a smaller computational requirement than AttentionXML. The memory footprints of AttentionXML’s models were smaller than FastXML’s models. This is because AttentionXML used more computational resources to train a deep model for each layer of its tree, while FastXML used more memory to train a larger tree ensemble to make up for the lower accuracy per tree. |
| first_indexed | 2024-10-01T02:19:54Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/149716 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:19:54Z |
| publishDate | 2021 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1497162023-07-07T18:24:06Z Recommendation systems based on extreme multi-label classification Chua, Song Ann Lihui CHEN School of Electrical and Electronic Engineering ELHCHEN@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Electrical and electronic engineering This project aims to implement a recommender system using extreme multi-label classification algorithms. In the era of big data, traditional recommender systems are unable to keep up with the scale and size of data available. Extreme multi-label classification can tag a given target with multiple labels that are most relevant to it from an extremely large dataset of labels. This report summarises the design implementation and empirical studies of extreme multi-label classification algorithms for recommendation systems on the MovieLens 1M benchmark dataset. This project studied 2 tree-based extreme multi-label classification algorithms, FastXML and AttentionXML, and implemented them using Python for a movie recommender system. This was to investigate the reformulation of the recommender problem as a multi-label classification task. The dataset was prepared such that each item that can be recommended by the system was treated as a unique label that can be tagged to a user by the classifier. The 2 algorithms were compared based on accuracy as well as computational resources required. The accuracy of AttentionXML was 46.6%, 5% larger than that of FastXML’s accuracy of 41.4%. However, FastXML had a smaller computational requirement than AttentionXML. The memory footprints of AttentionXML’s models were smaller than FastXML’s models. This is because AttentionXML used more computational resources to train a deep model for each layer of its tree, while FastXML used more memory to train a larger tree ensemble to make up for the lower accuracy per tree. Bachelor of Engineering (Electrical and Electronic Engineering) 2021-06-07T02:41:43Z 2021-06-07T02:41:43Z 2021 Final Year Project (FYP) Chua, S. A. (2021). Recommendation systems based on extreme multi-label classification. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/149716 https://hdl.handle.net/10356/149716 en A3044-201 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Electrical and electronic engineering Chua, Song Ann Recommendation systems based on extreme multi-label classification |
| title | Recommendation systems based on extreme multi-label classification |
| title_full | Recommendation systems based on extreme multi-label classification |
| title_fullStr | Recommendation systems based on extreme multi-label classification |
| title_full_unstemmed | Recommendation systems based on extreme multi-label classification |
| title_short | Recommendation systems based on extreme multi-label classification |
| title_sort | recommendation systems based on extreme multi label classification |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Electrical and electronic engineering |
| url | https://hdl.handle.net/10356/149716 |
| work_keys_str_mv | AT chuasongann recommendationsystemsbasedonextrememultilabelclassification |