• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.777-785
• /
• 2009

Collaborative filtering is a technique used to predict whether a particular user will like a particular item. User-based or item-based collaborative techniques have been used extensively in many commercial recommender systems. In this paper, a hybrid collaborative filtering method that combines user-based and item-based methods using a low-dimensional linear model is proposed. The proposed method solves the problems of sparsity and a large database by using NMF among the low-dimensional linear models. In collaborative filtering systems the methods using the NMF are useful in expressing users as semantic relations. However, they are model-based methods and the process of computation is complex, so they can not recommend items dynamically. In order to complement the shortcomings, the proposed method clusters users into groups by using NMF and selects features of groups by using TF-IDF. Mutual information is then used to compute similarities between items. The proposed method clusters users into groups and extracts features of groups on offline and determines the most suitable group for an active user using the features of groups on online. Finally, the proposed method reduces the time required to classify an active user into a group and outperforms previous methods by combining user-based and item-based collaborative filtering methods.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.316-325
• /
• 2003

A user preference prediction method using an exiting collaborative filtering technique has used the nearest-neighborhood method based on the user preference about items and has sought the user's similarity from the Pearson correlation coefficient. Therefore, it does not reflect any contents about items and also solve the problem of the sparsity. This study suggests the preference prediction system using the similarity weight granted Bayesian estimated value and the associative user clustering to complement problems of an exiting collaborative preference prediction method. This method suggested in this paper groups the user according to the Genre by using Association Rule Hypergraph Partitioning Algorithm and the new user is classified into one of these Genres by Naive Bayes classifier to slove the problem of sparsity in the collaborative filtering system. Besides, for get the similarity between users belonged to the classified genre and new users, this study allows the different estimated value to item which user vote through Naive Bayes learning. If the preference with estimated value is applied to the exiting Pearson correlation coefficient, it is able to promote the precision of the prediction by reducing the error of the prediction because of missing value. To estimate the performance of suggested method, the suggested method is compared with existing collaborative filtering techniques. As a result, the proposed method is efficient for improving the accuracy of prediction through solving problems of existing collaborative filtering techniques.

• Management & Information Systems Review
• /
• v.35 no.3
• /
• pp.115-130
• /
• 2016

This study aims to identify which social information have significant influence on the improvement of recommendation trust and how these effects can be different according to the product involvement level. Based on the relevant literature reviews, this study posits four characteristics of recommendation trust, which are closeness, similarity, sincerity, and reputation, and established a research model for the relationship between social information and recommendation trust. And we found a moderating effect of product involvement on the relationship between social information and recommendation trust. 205 trust relationships(links) from 55 respondents of Google Docs. survey data have been collected and tested using multiple regression and hierarchical regression analysis. The results of our hypotheses testing are summarized as follows. Firstly, four social information characteristics of closeness, similarity, sincerity, and reputation have a significantly positive effect on recommendation trust. Secondly, a moderating effect of product involvement between recommendation trust and antecedents (e.g., closeness and reputation) of social information is significant. From the results, we provide theoretical and managerial implications, and suggestions for further research.

• The Journal of the Korea Contents Association
• /
• v.12 no.12
• /
• pp.489-498
• /
• 2012

This paper consists of two parts: In the first part, we describe our work to build hierarchical knowledge base of digital library patron's research interests and learning topics in various scholarly areas through analyzing well classified Electronic Theses and Dissertations (ETDs) of NDLTD Union catalog. Journal articles from ACM Transactions and conference web sites of computing areas also are added in the analysis to specialize computing fields. This hierarchical knowledge base would be a useful tool for many social computing and information service applications, such as personalization, recommender system, text mining, technology opportunity mining, information visualization, and so on. In the second part, we compare four grouping algorithms to select best one for our data mining researches by testing each one with the hierarchical knowledge base we described in the first part. From these two studies, we intent to show traditional verification methods for social community miming researches, based on interviewing and answering questionnaires, which are expensive, slow, and privacy threatening, can be replaced with systematic, consistent, fast, and privacy protecting methods by using our suggested hierarchical knowledge base.

• Information Systems Review
• /
• v.14 no.3
• /
• pp.115-129
• /
• 2012

A matchmaking system is a type of recommender systems that provides a set of dating partners suitable for the user by online. Many matchmaking systems, which are widely used these days, require users to specify their preferences with regards to ideal dating partners based on criteria such as age, job and salary. However, some users are not aware of their exact preferences, or are reluctant to reveal this information even if they do know. Also, users' selection standards are not fixed and can change according to circumstances. This paper suggests a new matchmaking system called Decision Tree based Matchmaking System (DTMS) that automatically adjusts the stated standards of a user by analyzing the characteristics of the people the user chose to contact. AMMS provides recommendations for new users on the basis of their explicit preferences. However, as a user's behavioral records are accumulated, it begins to analyze their hidden implicit preferences using a decision tree technique. Subsequently, DTMS reflects these implicit preferences in proportion to their predictive accuracy. The DTMS is regularly updated when a user's data size increases by a set amount. This paper suggests an architecture for the DTMS and presents the results of the implementation of a prototype.

• The KIPS Transactions:PartB
• /
• v.10B no.3
• /
• pp.287-296
• /
• 2003

There has been much research focused on collaborative filtering technique in Recommender System. However, these studies have shown the First-Rater Problem and the Sparsity Problem. The main purpose of this Paper is to solve these Problems. In this Paper, we suggest the user's predicting preference method using Bayesian estimated value and the associative user clustering for the recalculation of preference. In addition to this method, to complement a shortcoming, which doesn't regard the attribution of item, we use Representative Attribute-Neighborhood method that is used for the prediction when we find the similar neighborhood through extracting the representative attribution, which most affect the preference. We improved the efficiency by using the associative user's clustering analysis in order to calculate the preference of specific item within the cluster item vector to the collaborative filtering algorithm. Besides, for the problem of the Sparsity and First-Rater, through using Association Rule Hypergraph Partitioning algorithm associative users are clustered according to the genre. New users are classified into one of these genres by Naive Bayes classifier. In addition, in order to get the similarity value between users belonged to the classified genre and new users, and this paper allows the different estimated value to item which user evaluated through Naive Bayes learning. As applying the preference granted the estimated value to Pearson correlation coefficient, it can make the higher accuracy because the errors that cause the missing value come less. We evaluate our method on a large collaborative filtering database of user rating and it significantly outperforms previous proposed method.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.5
• /
• pp.1235-1241
• /
• 2016

Personalized App recommendation system is recently famous since the number of various apps that can be used in smart phones that increases exponentially. However, the site users using google play site with malwares have experienced severe damages of privacy exposure and extortion as well as a simple damage of satisfaction descent at the same time. In addition, Sybil attack (Sybil) manipulating the score (rating) of each app with falmay also present because of the social networks development. Up until now, the sybil detection studies and malicious apps studies have been conducted independently. But it is important to determine finally the existence of intelligent attack with Sybil and malware simultaneously when we consider the intelligent attack types in real-time. Therefore, in this paper we experimentally evaluate the relationship between malware and sybils based on real cralwed dataset of goodlplay. Through the extensive evaluations, the correlation between malware and sybils is low for malware providers to hide themselves from Anti-Virus (AV).

• The KIPS Transactions:PartD
• /
• v.13D no.5 s.108
• /
• pp.661-670
• /
• 2006

With the development of e-Commerce and the proliferation of easily accessible information, information filtering systems such as recommender and SDI systems have become popular to prune large information spaces so that users are directed toward those items that best meet their needs and preferences. Until now, many information filtering methods have been proposed to support filtering systems. XML is emerging as a new standard for information. Recently, filtering systems need new approaches in dealing with XML documents. So, in this paper our system suggests a method to create multiform user profiles with XML's ability to represent structure. This system consists of two parts; one is an administrator profile definition part that an administrator defines to analyze users purchase pattern before a transaction such as purchase happens directly. an other is a user profile creation part module which is applied by the defined profile. Administrator profiles are made from DTD information and it is supposed to point the specific part of a document conforming to the DTD. Proposed system builds user's profile more accurately to get adaptability for user's behavior of buying and provide useful product information without inefficient searching based on such user's profile.

• Knowledge Management Research
• /
• v.22 no.4
• /
• pp.27-40
• /
• 2021

Click-Through Rate (CTR) prediction is a key function that determines the ranking of candidate items in the recommendation system and recommends high-ranking items to reduce customer information overload and achieve profit maximization through sales promotion. The fields of natural language processing and image classification are achieving remarkable growth through the use of deep neural networks. Recently, a transformer model based on an attention mechanism, differentiated from the mainstream models in the fields of natural language processing and image classification, has been proposed to achieve state-of-the-art in this field. In this study, we present a method for improving the performance of a transformer model for CTR prediction. In order to analyze the effect of discrete and categorical CTR data characteristics different from natural language and image data on performance, experiments on embedding regularization and transformer normalization are performed. According to the experimental results, it was confirmed that the prediction performance of the transformer was significantly improved when the L2 generalization was applied in the embedding process for CTR data input processing and when batch normalization was applied instead of layer normalization, which is the default regularization method, to the transformer model.

• Journal of Korean Library and Information Science Society
• /
• v.53 no.1
• /
• pp.231-263
• /
• 2022

The understanding and capability to utilize artificial intelligence (AI) incorporated technology has become a required basic skillset for the people living in today's information age, and various members of the university have also increasingly become aware of the need for AI education. Amidst such shifting societal demands, both domestic and international university libraries have recognized the users' need for educational content centered on AI, but a user-centered service that aims to provide personalized recommendations of digital AI educational content is yet to become available. It is critical while the demand for AI education amongst university students is progressively growing that university libraries acquire a clear understanding of user intention towards an AI educational content recommender system and the potential factors contributing to its success. This study intended to ascertain the factors affecting acceptance of such system, using the Extended Technology Acceptance Model with added variables - innovativeness, self-efficacy, social influence, system quality and task-technology fit - in addition to perceived usefulness, perceived ease of use, and intention to use. Quantitative research was conducted via online research surveys for university students, and quantitative research was conducted through written interviews of university librarians. Results show that all groups, regardless of gender, year, or major, have the intention to use the AI-related Educational Content Recommendation System, with the task suitability factor being the most dominant variant to affect use intention. University librarians have also expressed agreement about the necessity of the recommendation system, and presented budget and content quality issues as realistic restrictions of the aforementioned system.