Minsung Hong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Minsung Hong has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Jason J. Jung
Department of Computer Engineering, Chung-Ang University, Seoul, South Korea
11 shared publications - Chulmo Koo
Kyung Hee University
2 shared publications - Namho Chung
Kyung Hee University, Seoul, Republic of Korea
2 shared publications - O-Joun Lee
Department of Artificial Intelligence, The Catholic University of Korea, Bucheon, Republic of Korea
2 shared publications - Bong-Soo Sohn
Computer Science Department, Chung-Ang University, Seoul, South Korea
1 shared publication Chulmo Koo
Smart Tourism Education Platform, Kyung-Hee University, Seoul, South Korea
1 shared publication
Latest Publications
- Open Access
- 12 July 2023
- Published by Springer Nature in Research Square
Multi Sequential Embedding-based Cross-Domain Recommendation
In the real world, users interact with various items in a sequential and multi-behavior manner. Cross-domain recommendations (CDRs) usually transfer knowledge across different domains to improve their performance in a target domain. We propose a novel Multi Sequential Embedding for Cross-Domain Recommendation named MSECDR. Exploiting Doc2Vec, it reflects the sequentiality of user-item interactions in multiple domains on user and item modeling. In particular, this approach enables us to learn three types of item relationships (named intra-/inter- and other interaction sequences) simultaneously for representation learning. In addition, a deep neural network structure is introduced to combine the linearity and non-linearity of user-item interactions.Extensive experiments with six domain datasets from two real-world services (Amazon and TripAdvisor) demonstrate that MSECDR outperforms the state-of-the-art for single and cross-domain recommendations. Furthermore, experimental results show that MSECDR can effectively and efficiently adapt to CDRs fusing more domains.
- 12 July 2023
- Published by Springer Nature in Research Square
Open AccessMulti Sequential Embedding-based Cross-Domain Recommendation
In the real world, users interact with various items in a sequential and multi-behavior manner. Cross-domain recommendations (CDRs) usually transfer knowledge across different domains to improve their performance in a target domain. We propose a novel Multi Sequential Embedding for Cross-Domain Recommendation named MSECDR. Exploiting Doc2Vec, it reflects the sequentiality of user-item interactions in multiple domains on user and item modeling. In particular, this approach enables us to learn three types of item relationships (named intra-/inter- and other interaction sequences) simultaneously for representation learning. In addition, a deep neural network structure is introduced to combine the linearity and non-linearity of user-item interactions.Extensive experiments with six domain datasets from two real-world services (Amazon and TripAdvisor) demonstrate that MSECDR outperforms the state-of-the-art for single and cross-domain recommendations. Furthermore, experimental results show that MSECDR can effectively and efficiently adapt to CDRs fusing more domains.
- 19 January 2023
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 16 (2) , 147-162
TPEDTR: temporal preference embedding-based deep tourism recommendation with card transaction data
- 19 January 2023
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 16 (2) , 147-162
TPEDTR: temporal preference embedding-based deep tourism recommendation with card transaction data
- Open Access
- 1 January 2023
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 20 (3) , 911-931
Tourism recommendation based on word embedding from card transaction data
In the tourism industry, millions of card transactions generate a massive volume of big data. The card transactions eventually reflect customers? consumption behaviors and patterns. Additionally, recommender systems that incorporate users? personal preferences and consumption is an important subject of smart tourism. However, challenges exist such as handling the absence of rating data and considering spatial factor that significantly affects recommendation performance. This paper applies well-known Doc2Vec techniques to the tourism recommendation. We use them on non-textual features, card transaction dataset, to recommend tourism business services to target user groups who visit a specific location while addressing the challenges above. For the experiments, a card transaction dataset among eight years from Shinhan, which is one of the major card companies in the Republic of Korea, is used. The results demonstrate that the use of vector space representations trained by the Doc2Vec techniques considering spatial information is promising for tourism recommendations.
- 1 January 2023
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 20 (3) , 911-931
Open AccessTourism recommendation based on word embedding from card transaction data
In the tourism industry, millions of card transactions generate a massive volume of big data. The card transactions eventually reflect customers? consumption behaviors and patterns. Additionally, recommender systems that incorporate users? personal preferences and consumption is an important subject of smart tourism. However, challenges exist such as handling the absence of rating data and considering spatial factor that significantly affects recommendation performance. This paper applies well-known Doc2Vec techniques to the tourism recommendation. We use them on non-textual features, card transaction dataset, to recommend tourism business services to target user groups who visit a specific location while addressing the challenges above. For the experiments, a card transaction dataset among eight years from Shinhan, which is one of the major card companies in the Republic of Korea, is used. The results demonstrate that the use of vector space representations trained by the Doc2Vec techniques considering spatial information is promising for tourism recommendations.
- 14 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
DSER: Deep-Sequential Embedding for single domain Recommendation
- 14 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
DSER: Deep-Sequential Embedding for single domain Recommendation
- 10 March 2022
- Published by Springer Nature in Applied Intelligence
- Vol. 52 (13) , 15006-15025
Sentiment aware tensor model for multi-criteria recommendation
- 10 March 2022
- Published by Springer Nature in Applied Intelligence
- Vol. 52 (13) , 15006-15025
Sentiment aware tensor model for multi-criteria recommendation
- Open Access
- 1 January 2021
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 18 (1) , 285-301
Hypothetical tensor-based multi-criteria recommender system for new users with partial preferences
Multi-Criteria Recommender Systems (MCRSs) have been developed to improve the accuracy of single-criterion rating-based recommender systems that could not express and reflect users’ fine-grained rating behaviors. In most MCRSs, new users are asked to express their preferences on multi-criteria of items, to ad15 dress the cold-start problem. However, some of the users’ preferences collected are usually not complete due to users’ cognitive limitation and/or unfamiliarity on item domains, which is called ‘partial preferences’. The fundamental challenge and then negatively affects to accurately recommend items according to users’ preferences through MCRSs. In this paper, we propose a Hypothetical Tensor Model (HTM) to leverage auxiliary data complemented through three intuitive rules dealing with user’s unfamiliarity. First, we find four patterns of partial preferences that are caused by users’ unfamiliarity. And then the rules are defined by considering relationships between multi-criteria. Lastly, complemented preferences are modeled by a tensor to maintain an inherent structure of and correlations between the multi-criteria. Experiments on a TripAdvisor dataset showed that HTM improves MSE performances from 40 to 47% by comparing with other baseline methods. In particular, effective nesses of each rule regarding multi-criteria on HTM are clearly revealed.
- 1 January 2021
- Published by National Library of Serbia in Computer Science and Information Systems
- Vol. 18 (1) , 285-301
Open AccessHypothetical tensor-based multi-criteria recommender system for new users with partial preferences
Multi-Criteria Recommender Systems (MCRSs) have been developed to improve the accuracy of single-criterion rating-based recommender systems that could not express and reflect users’ fine-grained rating behaviors. In most MCRSs, new users are asked to express their preferences on multi-criteria of items, to ad15 dress the cold-start problem. However, some of the users’ preferences collected are usually not complete due to users’ cognitive limitation and/or unfamiliarity on item domains, which is called ‘partial preferences’. The fundamental challenge and then negatively affects to accurately recommend items according to users’ preferences through MCRSs. In this paper, we propose a Hypothetical Tensor Model (HTM) to leverage auxiliary data complemented through three intuitive rules dealing with user’s unfamiliarity. First, we find four patterns of partial preferences that are caused by users’ unfamiliarity. And then the rules are defined by considering relationships between multi-criteria. Lastly, complemented preferences are modeled by a tensor to maintain an inherent structure of and correlations between the multi-criteria. Experiments on a TripAdvisor dataset showed that HTM improves MSE performances from 40 to 47% by comparing with other baseline methods. In particular, effective nesses of each rule regarding multi-criteria on HTM are clearly revealed.
- Open Access
- 18 June 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (12) , 4183
Cognitive Similarity-Based Collaborative Filtering Recommendation System
This paper provides a new approach that improves collaborative filtering results in recommendation systems. In particular, we aim to ensure the reliability of the data set collected which is to collect the cognition about the item similarity from the users. Hence, in this work, we collect the cognitive similarity of the user about similar movies. Besides, we introduce a three-layered architecture that consists of the network between the items (item layer), the network between the cognitive similarity of users (cognition layer) and the network between users occurring in their cognitive similarity (user layer). For instance, the similarity in the cognitive network can be extracted from a similarity measure on the item network. In order to evaluate our method, we conducted experiments in the movie domain. In addition, for better performance evaluation, we use the F-measure that is a combination of two criteria and . Compared with the Pearson Correlation, our method more accurate and achieves improvement over the baseline 11.1% in the best case. The result shows that our method achieved consistent improvement of 1.8% to 3.2% for various neighborhood sizes in MAE calculation, and from 2.0% to 4.1% in RMSE calculation. This indicates that our method improves recommendation performance.- 18 June 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (12) , 4183
Open AccessCognitive Similarity-Based Collaborative Filtering Recommendation System
This paper provides a new approach that improves collaborative filtering results in recommendation systems. In particular, we aim to ensure the reliability of the data set collected which is to collect the cognition about the item similarity from the users. Hence, in this work, we collect the cognitive similarity of the user about similar movies. Besides, we introduce a three-layered architecture that consists of the network between the items (item layer), the network between the cognitive similarity of users (cognition layer) and the network between users occurring in their cognitive similarity (user layer). For instance, the similarity in the cognitive network can be extracted from a similarity measure on the item network. In order to evaluate our method, we conducted experiments in the movie domain. In addition, for better performance evaluation, we use the F-measure that is a combination of two criteria and . Compared with the Pearson Correlation, our method more accurate and achieves improvement over the baseline 11.1% in the best case. The result shows that our method achieved consistent improvement of 1.8% to 3.2% for various neighborhood sizes in MAE calculation, and from 2.0% to 4.1% in RMSE calculation. This indicates that our method improves recommendation performance. - Open Access
Implicit Stochastic Gradient Descent Method for Cross-Domain Recommendation System
The previous recommendation system applied the matrix factorization collaborative filtering (MFCF) technique to only single domains. Due to data sparsity, this approach has a limitation in overcoming the cold-start problem. Thus, in this study, we focus on discovering latent features from domains to understand the relationships between domains (called domain coherence). This approach uses potential knowledge of the source domain to improve the quality of the target domain recommendation. In this paper, we consider applying MFCF to multiple domains. Mainly, by adopting the implicit stochastic gradient descent algorithm to optimize the objective function for prediction, multiple matrices from different domains are consolidated inside the cross-domain recommendation system (CDRS). Additionally, we design a conceptual framework for CDRS, which applies to different industrial scenarios for recommenders across domains. Moreover, an experiment is devised to validate the proposed method. By using a real-world dataset gathered from Amazon Food and MovieLens, experimental results show that the proposed method improves 15.2% and 19.7% in terms of computation time and MSE over other methods on a utility matrix. Notably, a much lower convergence value of the loss function has been obtained from the experiment. Furthermore, a critical analysis of the obtained results shows that there is a dynamic balance between prediction accuracy and computational complexity.
Open AccessImplicit Stochastic Gradient Descent Method for Cross-Domain Recommendation System
The previous recommendation system applied the matrix factorization collaborative filtering (MFCF) technique to only single domains. Due to data sparsity, this approach has a limitation in overcoming the cold-start problem. Thus, in this study, we focus on discovering latent features from domains to understand the relationships between domains (called domain coherence). This approach uses potential knowledge of the source domain to improve the quality of the target domain recommendation. In this paper, we consider applying MFCF to multiple domains. Mainly, by adopting the implicit stochastic gradient descent algorithm to optimize the objective function for prediction, multiple matrices from different domains are consolidated inside the cross-domain recommendation system (CDRS). Additionally, we design a conceptual framework for CDRS, which applies to different industrial scenarios for recommenders across domains. Moreover, an experiment is devised to validate the proposed method. By using a real-world dataset gathered from Amazon Food and MovieLens, experimental results show that the proposed method improves 15.2% and 19.7% in terms of computation time and MSE over other methods on a utility matrix. Notably, a much lower convergence value of the loss function has been obtained from the experiment. Furthermore, a critical analysis of the obtained results shows that there is a dynamic balance between prediction accuracy and computational complexity.
- 23 April 2019
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 33 (3)
Victim detection platform in IoT paradigm
Summary: The advent of Internet of Things (IoT) paradigm is having a great ripple effect in the field of disaster management as well as many other areas. Victim detection has been studied as rescuer‐ or victim‐oriented approach, which utilises the IoT advanced technologies, but they have weaknesses according to disaster types. In this paper, we propose a Victim Detection Platform (VDP) architecture combining both approaches using various advanced technologies such as IoT, drone and edge/cloud computing to offer higher quality services, which ultimately result in swifter, better responses and more lives saved. First, we have reviewed related literature regarding three crucial issues (ie, multi‐modal evaluation for reliable data, edge‐based real‐time response, and privacy‐preserved Big Data analysis) to satisfy the service. In order to achieve these important aspects, the VDP architecture is described along with roles/relations of technologies and concepts leveraged to and data sources considered in the platform. To explore the realisation possibility of the proposed approach in real disasters, a validation scenario considering the critical issues and details in a victim detection task is illustrated, and then appropriate techniques, which will be utilised in the proposed VDP, and its justifications are discussed. Finally, we debate how the techniques are harmonised in the VDP and what the challenges are.
- 23 April 2019
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 33 (3)
Victim detection platform in IoT paradigm
Summary: The advent of Internet of Things (IoT) paradigm is having a great ripple effect in the field of disaster management as well as many other areas. Victim detection has been studied as rescuer‐ or victim‐oriented approach, which utilises the IoT advanced technologies, but they have weaknesses according to disaster types. In this paper, we propose a Victim Detection Platform (VDP) architecture combining both approaches using various advanced technologies such as IoT, drone and edge/cloud computing to offer higher quality services, which ultimately result in swifter, better responses and more lives saved. First, we have reviewed related literature regarding three crucial issues (ie, multi‐modal evaluation for reliable data, edge‐based real‐time response, and privacy‐preserved Big Data analysis) to satisfy the service. In order to achieve these important aspects, the VDP architecture is described along with roles/relations of technologies and concepts leveraged to and data sources considered in the platform. To explore the realisation possibility of the proposed approach in real disasters, a validation scenario considering the critical issues and details in a victim detection task is illustrated, and then appropriate techniques, which will be utilised in the proposed VDP, and its justifications are discussed. Finally, we debate how the techniques are harmonised in the VDP and what the challenges are.
- 6 July 2018
- Published by Springer Nature in Mobile Networks and Applications
- Vol. 23 (4) , 1111-1122
Towards Social Big Data-Based Affective Group Recommendation
- 6 July 2018
- Published by Springer Nature in Mobile Networks and Applications
- Vol. 23 (4) , 1111-1122
Towards Social Big Data-Based Affective Group Recommendation