• 한국컴퓨터정보학회논문지
• /
• 제24권9호
• /
• pp.77-82
• /
• 2019

Collaborative filtering is a popular technique for recommender systems and used in many practical commercial systems. Its basic principle is select similar neighbors of a current user and from their past preference information on items the system makes recommendations for the current user. One of the major problems inherent in this type of system is data sparsity of ratings. This is mainly caused from the underlying similarity measures which produce neighbors based on the ratings records. This paper handles this problem and suggests a new similarity measure. The proposed method takes users rating patterns into account for computing similarity, without just relying on the commonly rated items as in previous measures. Performance experiments of various existing measures are conducted and their performance is compared in terms of major performance metrics. As a result, the proposed measure reveals better or comparable achievements in all the metrics considered.

• Asia pacific journal of information systems
• /
• 제27권2호
• /
• pp.126-138
• /
• 2017

The proliferation of items increased the difficulty of customers in finding the specific items they want to purchase. To solve this problem, companies adopted recommender systems, such as collaborative filtering systems, to provide personalization services. However, companies use only meaningful and essential data given the explosive growth of data. Some customers are concerned that their private information may be exposed because CF systems necessarily deal with personal information. Based on these concerns, we analyze the effects of the amount of information on recommendation performance. We assume that a customer could choose to provide overall information or partial information. Experimental results indicate that customers who provided overall information generally demonstrated high performance, but differences exist according to the characteristics of products. Our study can provide companies with insights concerning the efficient utilization of data.

• 한국컴퓨터정보학회논문지
• /
• 제25권9호
• /
• pp.101-108
• /
• 2020

추천 시스템은 전자 상거래 시에 고객들의 상품 선택의 편의를 제공하므로 반드시 구비되어야 할 기능이다. 협력 필터링은 다른 사용자들이 선호하였던 상품이나 현 사용자가 과거 선호하였던 상품들을 위주로 추천 리스트를 제공하는 기법으로서, 가장 널리 활용되는 대표적 기법이다. 최근 딥러닝 인공지능 기술을 활용하여 추천 시스템의 성능 향상을 달성하는 연구가 활발히 진행되고 있다. 본 연구에서는 사용자가 부여한 평가등급만을 이용하여 딥러닝 기술의 일종인 제한 볼츠만 기계 학습을 통해 협력 필터링 기반의 추천 시스템을 개발한다. 또한 학습의 효율성과 성능을 위하여 학습 파라미터 변경 알고리즘을 제시한다. 제안 시스템의 성능 평가를 위하여 실험 분석을 통해 기존의 다양한 전통적 협력 필터링 기법들과 비교 분석을 실시하였으며, 제안 알고리즘은 기본적인 제한 볼츠만 기계 모델보다 우수한 성능을 가져오는 것으로 확인되었다.

• 한국경영과학회지
• /
• 제31권2호
• /
• pp.27-39
• /
• 2006

The Collaborative Filtering (CF) is one of the popular techniques for personalization in e-commerce storefronts. For CF-based recommendation, every customer needs to provide subjective evaluation ratings for some products based on his/her preference. Also, if an e-commerce site recommends a new product, some customers should rate it. However, there is no in-depth investigation on the impacts on recommendation performance of two number of ratings, i.e. the number of ratings of an individual customer and the number of ratings of an item, even though these are important factors to determine performance of CF methods. In this study, using publicly available EachMovie data set, we empirically investigate the relationships between the two number of ratings and the performance of CF. For the purpose, three analyses were executed. The first and second analyses were performed to investigate the relationship between the number of ratings of a particular customer and the recommendation performance of CF. In the third analysis, we investigate the relationship between the number of ratings on a particular item and the recommendation performance of CF. From these experiments, we can find that there are thresholds in terms of the number of ratings below which the recommendation performances increase monotonically. That is, the number of ratings of a customer and the number of ratings on an item are critical to the recommendation performance of CF when the number of ratings is less than the thresholds, but the value of the ratings decreases after the numbers of ratings pass the thresholds. The results of the experiments provide insight to making operational decisions concerning collaborative filtering in practice.

• 한국IT서비스학회지
• /
• 제18권4호
• /
• pp.83-95
• /
• 2019

It is becoming more and more difficult to cope with new knowledge and technology required by society by the efforts of one person or organization according to the development of science and technology. As a method to overcome this, collaborative research is becoming important. This tendency is increasing in the government R&D projects as well, and the 'A' test research institute, which is the subject of this paper, is also increasing a collaborative research. The purpose of this study is to analyze the network characteristics among the participating researchers in the government R&D project conducted by the institution A, and to ascertain how the network characters of the researchers actually affect the financial performance of the team. The results of the analysis show that 'closeness centrality' and 'degree of centrality' contribute positively to the financial performance of the team. On the other hand, 'betweenness centrality' and 'eigenvector centrality' have a negative effect on the financial performance of the team because they are not directly related to financial performance.

• 한국통신학회논문지
• /
• 제27권10C호
• /
• pp.1015-1026
• /
• 2002

최근 협력 시스템에 대한 연구가 증가되면서 협력 작업에 대한 연구와 더불어 협력 시스템 보안에 대한 연구가 중요시되고 있다. 협력 시스템에서 인증 및 암호화의 경우 기존의 정책을 사용하여 시스템의 견고성을 유지할 수 있다. 그러나 접근 제어 정책에서 기존의 정책을 그대로 사용하게 되는 경우 분산 환경, 개방된 네트워크, 다양한 주체와 객체의 존재로 인하여 협력 시스템은 신분, 직무, 그룹, 보안등급, 무결성 등급, 허가권을 포함해야 하는 다양한 접근상황을 고려해야 된다. 이 경우, 낮은 보안 등급의 주체가 높은 보안 등급의 객체로 접근을 허용 하거나, 높은 보안 등급의 주체가 낮은 보안 등급의 객체로 접근을 막아야 하는 복합적 상황을 해결하지 못한다. 더욱이 모든 접근상황을 제어하기 위하여 객체에 여러 접근제어 요소를 포함하여 접근 제어를 알고리즘화 할 경우 불필요한 상황을 모두 고려해야 하기 때문에 시스템의 성능 저하를 야기 시킨다. 이 같은 문제를 해결하기 위하여, 본 논문에서는 협력 시스템의 특징에 맞는 새로운 접근제어 프레임워크를 제안한다. 제안된 접근제어의 특징은 주체 및 객체에 다수의 접근 요소를 정의하여 기존 정책 보다 협력시스템과 같은 복합적인 상황에서 용이하게 적용되도록 하였다. 그리고 객체의 종류를 접근될 요소의 특징에 따라 세 가지로 구분하였고, 구분된 각 객체의 특징에 따라 알고리즘이 구현됨으로 빠르고 원활한 협력 작업이 수행되도록 하였다. 또한, 접근 요소 및 정책 변경이 용이하도록 확장성을 고려하였다. 모의실험 결과 다수의 접근 요소를 사용하였지만 시스템 성능은 접근 제어 정책을 적용하지 않았을 때와 큰 차이를 보이지 않았으며 복합적인 상황의 접근제어에서도 확실한 접근 제어가 가능했다.

• Asia pacific journal of information systems
• /
• 제18권4호
• /
• pp.83-103
• /
• 2008

In the new web based learning environment which has recently emerged, a variety of new learning objectives and teaching methods suited to this learning environment have been adopted. Recently, web based project-based learning methods have received a great deal of attention from those wishing to improve learning performance. The objective of this study is to identify the impact of characteristics of communication media and instruction behavior on collaborative interaction and project performance through web based group projects. The characteristics of communication media were divided into richness, flexibility, and ease of use, and the characteristics of instruction behavior were divided into support and expression, which are independent variables. Collaborative interaction as a mediate variable, was divided into information sharing and negotiation. Project performance was the dependent variable. To verify the proposed research model empirically, an experiment was conducted in which learners participated in on-line and off-line courses with group projects. The group project was conducted virtual product development(VPD), and designed a web-site about the VPD. At the end of the project, a survey was conducted. Of the 270 students, 239 responded. The students were assigned to groups of 3 or 4 members, and represented different genders and levels of computer competence. The reliability, validity, and correlation of research variables were analyzed using SPSS 14.0, and the measurement model and the structural goodness-of-fit of the research model were verified through SEM analysis using Lisrel 8.54. We found important results as follows; First, richness and ease of use has positive impacts on each of sharing information and negotiation. This suggests that richness and ease of use are useful in sharing information which is related to the task and agreeing in opinions among group members. However, flexibility has not positive impacts on sharing information and negotiation. This implies that there is no great difference in performance of PC and information literacy of user. Second, support and expression of instructor have positive impacts on sharing information and negotiation. This indicates that instructors play an important role in encouraging learners to participate in the project and communicating with them, sharing information related to the project, making a resonable decision and finally leading them to improve a project performance. Third, collaborative interaction has a positive impact on project performance. This result shows that if the ability to share information and negotiate among students was improved then a project performance would be improved as well. Recently, in the state of revitalized web based learning, it is opportune that web-based group project is practically conducted, and the impact of characteristics of communication media and characteristics of instruction behavior on sharing information, negotiating among group members and improving a project performance is verified. On the basis of these results, we propose that forms of learning, such as web based project, could be one of solution which is to enforce interaction among learners, and ultimately improve learning performance. Moreover web-based group project is able to make up for a weakness which makes it difficult to make interpersonal relations or friendship among learners in computer mediated communication or web based learning.

• 한국컴퓨터정보학회논문지
• /
• 제23권7호
• /
• pp.11-17
• /
• 2018

In this paper, we propose how to improve the lack of diversity in collaborative filtering, using tag scores contained in items rather than ratings of items. Collaborative filtering has excellent performance among recommendation system, but it is evaluated as lacking diversity. In order to solve this problem, this paper proposes a method for supplementing diversity lacking in collaborative filtering by using tags. By using tags that can be used universally without using the characteristics of specific articles in a recommendation system, The proposed method can be used.

• Management Science and Financial Engineering
• /
• 제12권2호
• /
• pp.71-85
• /
• 2006

Collaborative filtering, among other recommender systems, has been known as the most successful recommendation technique. However, it requires the user-item rating data, which may not be easily available. As an alternative, some collaborative filtering algorithms have been developed recently by utilizing the market basket data in the form of the binary user-item matrix. Viewing the recommendation scheme as a two-class classification problem, we proposed a new collaborative filtering scheme using a regularized discriminant analysis applied to the binary user-item data. The proposed discriminant model was built in terms of the major principal components and was used for predicting the probability of purchasing a particular item by an active user. The proposed scheme was illustrated with two modified real data sets and its performance was compared with the existing user-based approach in terms of the recommendation precision.

• Journal of information and communication convergence engineering
• /
• 제17권2호
• /
• pp.135-141
• /
• 2019

Collaborative filtering algorithms often encounter data sparsity issues. To overcome this issue, auxiliary information of relevant items is analyzed and an item attribute matrix is derived. In this study, we combine the user-item attribute preference with the traditional similarity calculation method to develop an improved similarity calculation approach and use weights to control the importance of these two elements. A collaborative filtering algorithm based on user-item attribute preference is proposed. The experimental results show that the performance of the recommender system is the most optimal when the weight of traditional similarity is equal to that of user-item attribute preference similarity. Although the rating-matrix is sparse, better recommendation results can be obtained by adding a suitable proportion of user-item attribute preference similarity. Moreover, the mean absolute error of the proposed approach is less than that of two traditional collaborative filtering algorithms.