• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.87-92
• /
• 2021

Bias, variance, error and learning are important factors for performance in modeling a big data based recommendation system. The recommendation model in this system must reduce complexity while maintaining the explanatory diagram. In addition, the sparsity of the dataset and the prediction of the system are more likely to be inversely proportional to each other. Therefore, a product recommendation model has been proposed through learning the similarity between products by using a factorization method of the sparsity of the dataset. In this paper, the generalization ability of the model is improved by applying the max-norm regularization as an optimization method for the loss function of this model. The solution is to apply a stochastic projection gradient descent method that projects a gradient. The sparser data became, it was confirmed that the propsed regularization method was relatively effective compared to the existing method through lots of experiment.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.323-338
• /
• 2023

TV advertising with deep analysis of watching pattern of audiences is important to set-top box audience targeting. Applying session-based recommendation model(SBR) to internet commercial, or recommendation based on searching history of user showed its effectiveness in previous studies, but applying SBR to the TV advertising was difficult in South Korea due to data unavailabilities. Also, traditional SBR has limitations for dealing with user preferences, especially in data with user identification information. To tackle with these problems, we first obtain set-top box data from three major broadcasting companies in South Korea(SKB, KT, LGU+) through collaboration with Korea Broadcast Advertising Corporation(KOBACO), and this data contains of watching sequence of 4,847 anonymized users for 6 month respectively. Second, we develop personalized session-based recommendation model to deal with hierarchical data of user-session-item. Experiments conducted on set-top box audience dataset and two other public dataset for validation. In result, our proposed model outperformed baseline model in some criteria.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.05a
• /
• pp.13-17
• /
• 2003

본 논문에서는 낮은 대역폭, 높은 지연, 그리고 잦은 네트워크 단절로 인한 모바일 컴퓨팅 환경의 문제점들을 해결하기 위한 효과적인 캐시 적재 기법으로서 협업 추천 기반의 데이터 마이닝 전략을 제안하였다. 캐시 적재가 모바일 클라이언트의 이러한 문제점들을 해결하기 위한 효율적인 방법이 된다는 기존의 연구는 많이 진행되어 왔다. 하지만 모바일 컴퓨터의 요구에 대한 이력 정보만을 이용한 기존의 연구는 모바일 클라이언트가 필요로 하는 모든 정보 요구를 만족하지 못하였다. 특히 저장 공간의 제약을 갖는 모바일 컴퓨터의 한계 때문에 더욱 큰 어려움을 갖게 되었다. 본 연구에서는 모바일 클라이언트의 이력 정보에 대하여 데이터 마이닝 기법을 적용한 캐시 적재 기법을 제안하여 적은 캐시 용량만으로도 모바일 클라이언트의 요구를 만족할 수 있는 아이템들을 효과적으로 서비스할 수 있도록 하였다. CSIM Simulator를 이용하여 모의 데이터를 생성하여, 제안 모형의 성능 평가를 위한 실험을 수행하였다. Cache hit ratio를 이용한 객관적인 성능 평가를 통하여 제안된 모형이 모바일 클라이언트의 캐시 적재 기법으로서 우수한 성능을 보임이 확인되었다.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.19-38
• /
• 2017

Collaborative filtering (CF) algorithm has been popularly used for implementing recommender systems. Until now, there have been many prior studies to improve the accuracy of CF. Among them, some recent studies adopt 'hybrid recommendation approach', which enhances the performance of conventional CF by using additional information. In this research, we propose a new hybrid recommender system which fuses CF and the results from the social network analysis on trust and distrust relationship networks among users to enhance prediction accuracy. The proposed algorithm of our study is based on memory-based CF. But, when calculating the similarity between users in CF, our proposed algorithm considers not only the correlation of the users' numeric rating patterns, but also the users' in-degree centrality values derived from trust and distrust relationship networks. In specific, it is designed to amplify the similarity between a target user and his or her neighbor when the neighbor has higher in-degree centrality in the trust relationship network. Also, it attenuates the similarity between a target user and his or her neighbor when the neighbor has higher in-degree centrality in the distrust relationship network. Our proposed algorithm considers four (4) types of user relationships - direct trust, indirect trust, direct distrust, and indirect distrust - in total. And, it uses four adjusting coefficients, which adjusts the level of amplification / attenuation for in-degree centrality values derived from direct / indirect trust and distrust relationship networks. To determine optimal adjusting coefficients, genetic algorithms (GA) has been adopted. Under this background, we named our proposed algorithm as SNACF-GA (Social Network Analysis - based CF using GA). To validate the performance of the SNACF-GA, we used a real-world data set which is called 'Extended Epinions dataset' provided by 'trustlet.org'. It is the data set contains user responses (rating scores and reviews) after purchasing specific items (e.g. car, movie, music, book) as well as trust / distrust relationship information indicating whom to trust or distrust between users. The experimental system was basically developed using Microsoft Visual Basic for Applications (VBA), but we also used UCINET 6 for calculating the in-degree centrality of trust / distrust relationship networks. In addition, we used Palisade Software's Evolver, which is a commercial software implements genetic algorithm. To examine the effectiveness of our proposed system more precisely, we adopted two comparison models. The first comparison model is conventional CF. It only uses users' explicit numeric ratings when calculating the similarities between users. That is, it does not consider trust / distrust relationship between users at all. The second comparison model is SNACF (Social Network Analysis - based CF). SNACF differs from the proposed algorithm SNACF-GA in that it considers only direct trust / distrust relationships. It also does not use GA optimization. The performances of the proposed algorithm and comparison models were evaluated by using average MAE (mean absolute error). Experimental result showed that the optimal adjusting coefficients for direct trust, indirect trust, direct distrust, indirect distrust were 0, 1.4287, 1.5, 0.4615 each. This implies that distrust relationships between users are more important than trust ones in recommender systems. From the perspective of recommendation accuracy, SNACF-GA (Avg. MAE = 0.111943), the proposed algorithm which reflects both direct and indirect trust / distrust relationships information, was found to greatly outperform a conventional CF (Avg. MAE = 0.112638). Also, the algorithm showed better recommendation accuracy than the SNACF (Avg. MAE = 0.112209). To confirm whether these differences are statistically significant or not, we applied paired samples t-test. The results from the paired samples t-test presented that the difference between SNACF-GA and conventional CF was statistical significant at the 1% significance level, and the difference between SNACF-GA and SNACF was statistical significant at the 5%. Our study found that the trust/distrust relationship can be important information for improving performance of recommendation algorithms. Especially, distrust relationship information was found to have a greater impact on the performance improvement of CF. This implies that we need to have more attention on distrust (negative) relationships rather than trust (positive) ones when tracking and managing social relationships between users.

• Journal of Korea Multimedia Society
• /
• v.12 no.2
• /
• pp.219-230
• /
• 2009

With the development of broadcasting technology from analogue to interactive digital, the number of TV channels and TV contents provided to audiences is increasing in a rapid speed. In this multi-channel world, it is difficult to adapt to the increase of the TV channel numbers and their contents merely using remote controller to search channels. For these reasons, the EPG system, one of the essential services providing convenience to audiences, is proposed in this paper. Collaborative filtering method with multi-step filtering is used in EPG to recommend contents according to the preference of audience groups with similar preference. To implement our designed TV contents recommendation EPG, we prefer DiTV and use JavaXlet programming based on MHP. The European DVB-MHP specification will be also our domestic standard in DiTV. Finally, the result is verified by OpenMHP emulator.

• The Journal of the Korea Contents Association
• /
• v.7 no.12
• /
• pp.292-302
• /
• 2007

Thousands of studies on predicting information and products that are suitable for customers' preference have been actively proceeding. In massive information, unnecessary information should be removed to satisfy customers' needs. This Information filtering has been proceeding with several methods such as content-based and collaborative filtering etc. These conventional filtering methods have scarcity and scalability problems. Thus, this paper proposes a recommendation system using BPN to solve them. Data obtained by survey questionnaire are used as training data of neural network. The recommendation system using neural network is expected to recommend suitable products because it creates optimal network. Finally, the prototype for recommendation system based on neural network is proposed to collect data and recommend appropriate methods through survey questionnaire. As a result, this research improved the problems of conventional information filtering.

• Information Systems Review
• /
• v.23 no.3
• /
• pp.77-96
• /
• 2021

This research investigates the pivotal role of AI-enabled technologies in vitalizing the local economy. Collaborating with a leading search engine company, we examine the direct and indirect of an AI-based location search service on the success of sampled 7,035 local restaurants in Gangnam area in Seoul. We find that increased use of AI-enabled search and recommendation services significantly improved the selections of previously less-discovered or less-popular restaurants by users, and it also enhanced the stores' overall conversion rates. The main research findings have contributions to extant literature in theorizing the value of AI applications in local economy and have managerial implications for search businesses and local stores by recommending strategic use of AI applications in their businesses that are effective in highly competitive markets.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.05a
• /
• pp.218-225
• /
• 2007

Recommendation system is one way of implementing personalized service. The collaborative filtering is one of the major techniques that have been employed for recommendation systems. It has proven its effectiveness in the recommendation systems for such domain as motion picture or music. However, it has some limitations, i.e., sparsity and scalability. In this research, as one way of overcoming such limitations, we proposed the stepwise collaborative filtering method. To show the practicality of our proposed method, we designed and implemented a movie recommendation system which we shall call Step_CF, and its performance was evaluated using MovieLens data. The performance of Step_CF was better than that of Basic_CF that was implemented using the original collaborative filtering method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.23-28
• /
• 2012

Recently, due to the increasing demand for mobile devices, mobile searching market is rapidly growing. However, there is the limit of screen size, when searching for mobile devices, various results should be shown at a glance. The reason is that results are important given that up to 43 percent of people tend to check only first page. In this paper, a set of keywords for searching will be used to find out the users' interests. Users were divided into groups after going through Collaboration filtering. Therefore, the result of this experiment, reduced time for searching and improved quality of searching were confirmed.