Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization
Recommendation systems based on convolutional neural network (CNN) have attracted great attention due to their effectiveness in processing unstructured data such as images or audio. However, a huge amount of raw data produced by data crawling and digital transformation is structured, which makes it...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Future Internet |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-5903/14/1/20 |
| _version_ | 1797493943268540416 |
|---|---|
| author | Tan Nghia Duong Nguyen Nam Doan Truong Giang Do Manh Hoang Tran Duc Minh Nguyen Quang Hieu Dang |
| author_facet | Tan Nghia Duong Nguyen Nam Doan Truong Giang Do Manh Hoang Tran Duc Minh Nguyen Quang Hieu Dang |
| author_sort | Tan Nghia Duong |
| collection | DOAJ |
| description | Recommendation systems based on convolutional neural network (CNN) have attracted great attention due to their effectiveness in processing unstructured data such as images or audio. However, a huge amount of raw data produced by data crawling and digital transformation is structured, which makes it difficult to utilize the advantages of CNN. This paper introduces a novel autoencoder, named Half Convolutional Autoencoder, which adopts convolutional layers to discover the high-order correlation between structured features in the form of Tag Genome, the side information associated with each movie in the MovieLens 20 M dataset, in order to generate a robust feature vector. Subsequently, these new movie representations, along with the introduction of users’ characteristics generated via Tag Genome and their past transactions, are applied into well-known matrix factorization models to resolve the initialization problem and enhance the predicting results. This method not only outperforms traditional matrix factorization techniques by at least 5.35% in terms of accuracy but also stabilizes the training process and guarantees faster convergence. |
| first_indexed | 2024-03-10T01:27:14Z |
| format | Article |
| id | doaj.art-dd082a9504b24f57aab039d8cd755f17 |
| institution | Directory Open Access Journal |
| issn | 1999-5903 |
| language | English |
| last_indexed | 2024-03-10T01:27:14Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Future Internet |
| spelling | doaj.art-dd082a9504b24f57aab039d8cd755f172023-11-23T13:49:31ZengMDPI AGFuture Internet1999-59032022-01-011412010.3390/fi14010020Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix FactorizationTan Nghia Duong0Nguyen Nam Doan1Truong Giang Do2Manh Hoang Tran3Duc Minh Nguyen4Quang Hieu Dang5School of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamSchool of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamSchool of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamSchool of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamSchool of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamSchool of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, VietnamRecommendation systems based on convolutional neural network (CNN) have attracted great attention due to their effectiveness in processing unstructured data such as images or audio. However, a huge amount of raw data produced by data crawling and digital transformation is structured, which makes it difficult to utilize the advantages of CNN. This paper introduces a novel autoencoder, named Half Convolutional Autoencoder, which adopts convolutional layers to discover the high-order correlation between structured features in the form of Tag Genome, the side information associated with each movie in the MovieLens 20 M dataset, in order to generate a robust feature vector. Subsequently, these new movie representations, along with the introduction of users’ characteristics generated via Tag Genome and their past transactions, are applied into well-known matrix factorization models to resolve the initialization problem and enhance the predicting results. This method not only outperforms traditional matrix factorization techniques by at least 5.35% in terms of accuracy but also stabilizes the training process and guarantees faster convergence.https://www.mdpi.com/1999-5903/14/1/20autoencodercollaborative filteringconvolutional neural networkmatrix factorizationrecommendation system |
| spellingShingle | Tan Nghia Duong Nguyen Nam Doan Truong Giang Do Manh Hoang Tran Duc Minh Nguyen Quang Hieu Dang Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization Future Internet autoencoder collaborative filtering convolutional neural network matrix factorization recommendation system |
| title | Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization |
| title_full | Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization |
| title_fullStr | Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization |
| title_full_unstemmed | Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization |
| title_short | Utilizing Half Convolutional Autoencoder to Generate User and Item Vectors for Initialization in Matrix Factorization |
| title_sort | utilizing half convolutional autoencoder to generate user and item vectors for initialization in matrix factorization |
| topic | autoencoder collaborative filtering convolutional neural network matrix factorization recommendation system |
| url | https://www.mdpi.com/1999-5903/14/1/20 |
| work_keys_str_mv | AT tannghiaduong utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization AT nguyennamdoan utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization AT truonggiangdo utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization AT manhhoangtran utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization AT ducminhnguyen utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization AT quanghieudang utilizinghalfconvolutionalautoencodertogenerateuseranditemvectorsforinitializationinmatrixfactorization |