A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems
To promote online businesses and sales, e-commerce industry focuses to fulfill users' demands by giving them top set of recommendations which are ranked through different ranking measures.Deep learning based auto-encoder models have further improved the performance of recommender systems. A sta...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8830343/ |
| _version_ | 1819174422128361472 |
|---|---|
| author | Zeshan Aslam Khan Syed Zubair Kashif Imran Rehan Ahmad Sharjeel Abid Butt Naveed Ishtiaq Chaudhary |
| author_facet | Zeshan Aslam Khan Syed Zubair Kashif Imran Rehan Ahmad Sharjeel Abid Butt Naveed Ishtiaq Chaudhary |
| author_sort | Zeshan Aslam Khan |
| collection | DOAJ |
| description | To promote online businesses and sales, e-commerce industry focuses to fulfill users' demands by giving them top set of recommendations which are ranked through different ranking measures.Deep learning based auto-encoder models have further improved the performance of recommender systems. A state-of-the-art collaborative denoising auto-encoder (CDAE) models user-item interactions as a corrupted version of users rating inputs. However, this architecture still lacks users' ratings-trend information which is an important parameter to recommend top-N items to users. In this paper, building upon CDAE characteristics, we propose a novel users rating-trend based collaborative denoising auto-encoder (UT-CDAE) which determines user-item correlations by evaluating rating-trend(High or Low) of a user towards a set of items. This inclusion of a user's rating-trend provides additional regularization flexibility which helps to predict improved top-N recommendations. The correctness of the suggested method is verified through different ranking evaluation metrics i.e., (mean reciprocal rank, mean average precision and normalized discounted gain), for various input corruption values, learning rates and regularization parameters.Experiments on standard ML-100K and ML-1M datasets show that suggested model has improved performance overstate-of-the-art denoising auto-encodermodels. |
| first_indexed | 2024-12-22T20:38:43Z |
| format | Article |
| id | doaj.art-389db11467fa41f8be48680f9ec49863 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T20:38:43Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-389db11467fa41f8be48680f9ec498632022-12-21T18:13:23ZengIEEEIEEE Access2169-35362019-01-01714128714131010.1109/ACCESS.2019.29406038830343A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender SystemsZeshan Aslam Khan0https://orcid.org/0000-0003-3902-5814Syed Zubair1Kashif Imran2Rehan Ahmad3Sharjeel Abid Butt4Naveed Ishtiaq Chaudhary5Electrical Engineering Department, International Islamic University, Islamabad, PakistanElectrical Engineering Department, International Islamic University, Islamabad, PakistanDepartment of Electrical Power Engineering, U.S.-Pakistan Center for Advanced Studies in Energy, National University of Sciences and Technology (NUST), Islamabad, PakistanElectrical Engineering Department, International Islamic University, Islamabad, PakistanElectrical Engineering Department, International Islamic University, Islamabad, PakistanElectrical Engineering Department, International Islamic University, Islamabad, PakistanTo promote online businesses and sales, e-commerce industry focuses to fulfill users' demands by giving them top set of recommendations which are ranked through different ranking measures.Deep learning based auto-encoder models have further improved the performance of recommender systems. A state-of-the-art collaborative denoising auto-encoder (CDAE) models user-item interactions as a corrupted version of users rating inputs. However, this architecture still lacks users' ratings-trend information which is an important parameter to recommend top-N items to users. In this paper, building upon CDAE characteristics, we propose a novel users rating-trend based collaborative denoising auto-encoder (UT-CDAE) which determines user-item correlations by evaluating rating-trend(High or Low) of a user towards a set of items. This inclusion of a user's rating-trend provides additional regularization flexibility which helps to predict improved top-N recommendations. The correctness of the suggested method is verified through different ranking evaluation metrics i.e., (mean reciprocal rank, mean average precision and normalized discounted gain), for various input corruption values, learning rates and regularization parameters.Experiments on standard ML-100K and ML-1M datasets show that suggested model has improved performance overstate-of-the-art denoising auto-encodermodels.https://ieeexplore.ieee.org/document/8830343/Auto-encoderscollaborative filteringdenoisinge-commercerecommender systemstop-N recommendations |
| spellingShingle | Zeshan Aslam Khan Syed Zubair Kashif Imran Rehan Ahmad Sharjeel Abid Butt Naveed Ishtiaq Chaudhary A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems IEEE Access Auto-encoders collaborative filtering denoising e-commerce recommender systems top-N recommendations |
| title | A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems |
| title_full | A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems |
| title_fullStr | A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems |
| title_full_unstemmed | A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems |
| title_short | A New Users Rating-Trend Based Collaborative Denoising Auto-Encoder for Top-N Recommender Systems |
| title_sort | new users rating trend based collaborative denoising auto encoder for top n recommender systems |
| topic | Auto-encoders collaborative filtering denoising e-commerce recommender systems top-N recommendations |
| url | https://ieeexplore.ieee.org/document/8830343/ |
| work_keys_str_mv | AT zeshanaslamkhan anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT syedzubair anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT kashifimran anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT rehanahmad anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT sharjeelabidbutt anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT naveedishtiaqchaudhary anewusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT zeshanaslamkhan newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT syedzubair newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT kashifimran newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT rehanahmad newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT sharjeelabidbutt newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems AT naveedishtiaqchaudhary newusersratingtrendbasedcollaborativedenoisingautoencoderfortopnrecommendersystems |