Truong Giang Do
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Truong Giang Do has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Tan Nghia Duong
Hanoi University of Science and Technology (HUST)
2 shared publications - Duc Minh Nguyen
School of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, Vietnam
1 shared publication - Hoang Tran Manh
Hanoi University of Science and Technology (HUST)
1 shared publication - Manh Hoang Tran
Affiliation School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, Hanoi, Vietnam
1 shared publication - Nguyen Nam Doan
School of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, Vietnam
1 shared publication - Quang Hieu Dang
School of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 100000, Vietnam
1 shared publication
Latest Publications
- Open Access
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 181-188
A Generalized Autorec Framework Applying Content-based Information for Resolving Data Sparsity Problem
Recent years have witnessed several neural network-based collaborative filtering systems that yield immense success in providing users with personalized suggestions. These systems, however, severely suffer from the data sparsity problem. Similar to other neural network-based collaborative filtering systems, the sparse rating information also affects Autorec at the input layer. This paper represents a generalized hybrid Autorec framework that can accept a variety of content-based representations as input to create a more efficient system. Subsequently, to fully employ the capability of the Autoencoder architecture in compressing information, rating vectors compressed via the conventional Autorec, together with content-based information, creates a robust hybrid input for our framework to resolve the sparsity problem. Empirical experiments demonstrate the state-of-the-art performance of our system compared to other hybrid recommendation systems.
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 181-188
Open AccessA Generalized Autorec Framework Applying Content-based Information for Resolving Data Sparsity Problem
Recent years have witnessed several neural network-based collaborative filtering systems that yield immense success in providing users with personalized suggestions. These systems, however, severely suffer from the data sparsity problem. Similar to other neural network-based collaborative filtering systems, the sparse rating information also affects Autorec at the input layer. This paper represents a generalized hybrid Autorec framework that can accept a variety of content-based representations as input to create a more efficient system. Subsequently, to fully employ the capability of the Autoencoder architecture in compressing information, rating vectors compressed via the conventional Autorec, together with content-based information, creates a robust hybrid input for our framework to resolve the sparsity problem. Empirical experiments demonstrate the state-of-the-art performance of our system compared to other hybrid recommendation systems.
- Open Access
- 4 January 2022
- Published by MDPI AG in Future Internet
- Vol. 14 (1) , 20
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 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.
- 4 January 2022
- Published by MDPI AG in Future Internet
- Vol. 14 (1) , 20
Open AccessUtilizing 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 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.