소셜 네트워크는 사용자들의 공통된 관심사, 경험, 그리고 일상 생활들을 함께 공유하기 위해 소셜 네트워크 상 사람들을 서로 연결시켜주는 거대한 커뮤니케이션 플랫폼이다. 소셜 네트워크상의 사용자들은 포스팅, 댓글, 인스턴스 메시지, 게임, 소셜 이벤트 외에도 다양한 애플리케이션을 통해 다른 사용자들과 소통하고 개인 정보 관리하는데 많은 시간을 소비한다. 소셜 네트워크 상의 풍부한 사용자 정보는 추천시스템이 추천 성능을 향상시키기 위해 필요한 큰 잠재력이 되었다. 대부분의 사용자들은 어떤 상품을 구매하기 전 가까운 관계이거나 같은 성향을 가진 사람들의 의견을 반영하여 의사 결정을 하게 된다. 그러므로 소셜 네트워크에서의 사용자 관계는 추천시스템을 위한 사용자 선호도 예측을 효율적으로 높이는데 중요한 요소라 할 수 있다. 일부 연구자들은 소셜 네트워크에서의 사용자와 다른 사용자들 사이의 상호작용 즉, 소셜 관계(social relationship)와 같은 소셜 데이터가 추천시스템에서 추천의 질에 어떠한 영향을 미치는가를 연구하고 있다. 추천시스템은 아마존, 이베이, Last.fm과 같은 큰 규모의 전자상거래 사이트 또한 채택하여 사용되는 시스템으로, 추천시스템을 위한 방법으로는 협업적 여과 방법과 내용 기반 여과 방법이 있다. 협업적 여과 방법은 사용자들의 선호도 학습에 의해 사용자가 아직 평가하지 않은 아이템 중 선호할 수 있는 아이템을 정확하게 제안하기 위한 추천시스템 방법 중 하나이다. 협업적 여과는 사용자들의 데이터에 초점을 맞춘 방법으로 유사한 배경과 선호도를 가지는 사용자들로부터 정보를 수집하여 사용자들의 선호도 예측을 자동으로 발생시킨다. 특히 협업적 여과는 근접한 이웃 사용자들에 의해서 목적 사용자가 선호할 수 있는 아이템을 제시하는 것으로 유사한 이웃 사용자를 찾는 것이 중요하다. 좋은 이웃 사용자 발견은 사용자와 아이템을 고려하는 방법이 일반적이다. 각 사용자는 아이템 즉, 영화, 상품, 책 등에 자신의 선호도를 나타내기 위하여 평가 값을 입력하고, 시스템은 이를 바탕으로 사용자-평가 행렬을 구축한다. 이 사용자-평가 행렬은 목적 사용자와 유사하게 아이템을 평가한 사용자 그룹을 찾기 위한 것으로, 목적 사용자가 아직 평가하지 않은 아이템에 대하여 사용자-평가 매트릭스를 통해 그 평가 값을 예측한다. 현재 이 협업적 여과 방법은 전자상거래와 정보 검색에서 적용되어 개인화 시스템에 효율적으로 사용되고 있다. 하지만 초기 사용자 문제, 데이터 희박성 문제와 확장성 그리고 예측 정확도 향상 등 해결해야 할 과제가 여전히 남아 있다. 이러한 문제들을 해소하기 위해 많은 연구자들은 하이브리드, 신뢰기반, 소셜 네트워크 기반 협업적 여과와 같은 다양한 방법을 제안하였다. 본 논문에서는 전통적인 협업적 여과 방식의 예측 정확도와 추천 성능을 향상시키기 위해 소셜 네트워크에 존재하는 소셜 관계를 이용한 협업적 여과 시스템을 제안한다. 소셜 관계는 소셜 네트워크 서비스 중 하나인 페이스북 사용자들이 남긴 포스팅과 사용자의 소셜 네트워크 친구와 의견 교류 중 남긴 코멘트와 같은 사용자 행동을 기반으로 정의된다. 소셜 관계를 구축하기 위해 소셜 네트워크 사용자의 포스팅과 댓글을 추출하고, 추출된 텍스트에 불용어 및 특수 기호 제거와 스테밍 등 전처리를 수행하였다. 특징 벡터는 TF-IDF
소셜 네트워크는 사용자들의 공통된 관심사, 경험, 그리고 일상 생활들을 함께 공유하기 위해 소셜 네트워크 상 사람들을 서로 연결시켜주는 거대한 커뮤니케이션 플랫폼이다. 소셜 네트워크상의 사용자들은 포스팅, 댓글, 인스턴스 메시지, 게임, 소셜 이벤트 외에도 다양한 애플리케이션을 통해 다른 사용자들과 소통하고 개인 정보 관리하는데 많은 시간을 소비한다. 소셜 네트워크 상의 풍부한 사용자 정보는 추천시스템이 추천 성능을 향상시키기 위해 필요한 큰 잠재력이 되었다. 대부분의 사용자들은 어떤 상품을 구매하기 전 가까운 관계이거나 같은 성향을 가진 사람들의 의견을 반영하여 의사 결정을 하게 된다. 그러므로 소셜 네트워크에서의 사용자 관계는 추천시스템을 위한 사용자 선호도 예측을 효율적으로 높이는데 중요한 요소라 할 수 있다. 일부 연구자들은 소셜 네트워크에서의 사용자와 다른 사용자들 사이의 상호작용 즉, 소셜 관계(social relationship)와 같은 소셜 데이터가 추천시스템에서 추천의 질에 어떠한 영향을 미치는가를 연구하고 있다. 추천시스템은 아마존, 이베이, Last.fm과 같은 큰 규모의 전자상거래 사이트 또한 채택하여 사용되는 시스템으로, 추천시스템을 위한 방법으로는 협업적 여과 방법과 내용 기반 여과 방법이 있다. 협업적 여과 방법은 사용자들의 선호도 학습에 의해 사용자가 아직 평가하지 않은 아이템 중 선호할 수 있는 아이템을 정확하게 제안하기 위한 추천시스템 방법 중 하나이다. 협업적 여과는 사용자들의 데이터에 초점을 맞춘 방법으로 유사한 배경과 선호도를 가지는 사용자들로부터 정보를 수집하여 사용자들의 선호도 예측을 자동으로 발생시킨다. 특히 협업적 여과는 근접한 이웃 사용자들에 의해서 목적 사용자가 선호할 수 있는 아이템을 제시하는 것으로 유사한 이웃 사용자를 찾는 것이 중요하다. 좋은 이웃 사용자 발견은 사용자와 아이템을 고려하는 방법이 일반적이다. 각 사용자는 아이템 즉, 영화, 상품, 책 등에 자신의 선호도를 나타내기 위하여 평가 값을 입력하고, 시스템은 이를 바탕으로 사용자-평가 행렬을 구축한다. 이 사용자-평가 행렬은 목적 사용자와 유사하게 아이템을 평가한 사용자 그룹을 찾기 위한 것으로, 목적 사용자가 아직 평가하지 않은 아이템에 대하여 사용자-평가 매트릭스를 통해 그 평가 값을 예측한다. 현재 이 협업적 여과 방법은 전자상거래와 정보 검색에서 적용되어 개인화 시스템에 효율적으로 사용되고 있다. 하지만 초기 사용자 문제, 데이터 희박성 문제와 확장성 그리고 예측 정확도 향상 등 해결해야 할 과제가 여전히 남아 있다. 이러한 문제들을 해소하기 위해 많은 연구자들은 하이브리드, 신뢰기반, 소셜 네트워크 기반 협업적 여과와 같은 다양한 방법을 제안하였다. 본 논문에서는 전통적인 협업적 여과 방식의 예측 정확도와 추천 성능을 향상시키기 위해 소셜 네트워크에 존재하는 소셜 관계를 이용한 협업적 여과 시스템을 제안한다. 소셜 관계는 소셜 네트워크 서비스 중 하나인 페이스북 사용자들이 남긴 포스팅과 사용자의 소셜 네트워크 친구와 의견 교류 중 남긴 코멘트와 같은 사용자 행동을 기반으로 정의된다. 소셜 관계를 구축하기 위해 소셜 네트워크 사용자의 포스팅과 댓글을 추출하고, 추출된 텍스트에 불용어 및 특수 기호 제거와 스테밍 등 전처리를 수행하였다. 특징 벡터는 TF-IDF
Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people v...
Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people via posting, commenting, messaging, games, social events, and applications. Due to the growth of user's distributed information in social network, there is a great potential to utilize the social data to enhance the quality of recommender system. There are some researches focusing on social network analysis that investigate how social network can be used in recommendation domain. Among these researches, we are interested in taking advantages of the interaction between a user and others in social network that can be determined and known as social relationship. Furthermore, mostly user's decisions before purchasing some products depend on suggestion of people who have either the same preferences or closer relationship. For this reason, we believe that user's relationship in social network can provide an effective way to increase the quality in prediction user's interests of recommender system. Therefore, social relationship between users encountered from social network is a common factor to improve the way of predicting user's preferences in the conventional approach. Recommender system is dramatically increasing in popularity and currently being used by many e-commerce sites such as Amazon.com, Last.fm, eBay.com, etc. Collaborative filtering (CF) method is one of the essential and powerful techniques in recommender system for suggesting the appropriate items to user by learning user's preferences. CF method focuses on user data and generates automatic prediction about user's interests by gathering information from users who share similar background and preferences. Specifically, the intension of CF method is to find users who have similar preferences and to suggest target user items that were mostly preferred by those nearest neighbor users. There are two basic units that need to be considered by CF method, the user and the item. Each user needs to provide his rating value on items i.e. movies, products, books, etc to indicate their interests on those items. In addition, CF uses the user-rating matrix to find a group of users who have similar rating with target user. Then, it predicts unknown rating value for items that target user has not rated. Currently, CF has been successfully implemented in both information filtering and e-commerce applications. However, it remains some important challenges such as cold start, data sparsity, and scalability reflected on quality and accuracy of prediction. In order to overcome these challenges, many researchers have proposed various kinds of CF method such as hybrid CF, trust-based CF, social network-based CF, etc. In the purpose of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which integrates traditional CF technique with social relationship between users discovered from user's behavior in social network i.e. Facebook. We identify user's relationship from behavior of user such as posts and comments interacted with friends in Facebook. We believe that social relationship implicitly inferred from user's behavior can be likely applied to compensate the limitation of conventional approach. Therefore, we extract posts and comments of each user by using Facebook Graph API and calculate feature score among each term to obtain feature vector for computing similarity of user. Then, we combine the result with similarity value computed using traditional CF technique. Finally, our system provides a list of recommended items according to neighbor users who have the biggest total similarity value to the target user. In order to verify and evaluate our proposed method we have performed an experiment on data collected from our Movies Rating S
Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people via posting, commenting, messaging, games, social events, and applications. Due to the growth of user's distributed information in social network, there is a great potential to utilize the social data to enhance the quality of recommender system. There are some researches focusing on social network analysis that investigate how social network can be used in recommendation domain. Among these researches, we are interested in taking advantages of the interaction between a user and others in social network that can be determined and known as social relationship. Furthermore, mostly user's decisions before purchasing some products depend on suggestion of people who have either the same preferences or closer relationship. For this reason, we believe that user's relationship in social network can provide an effective way to increase the quality in prediction user's interests of recommender system. Therefore, social relationship between users encountered from social network is a common factor to improve the way of predicting user's preferences in the conventional approach. Recommender system is dramatically increasing in popularity and currently being used by many e-commerce sites such as Amazon.com, Last.fm, eBay.com, etc. Collaborative filtering (CF) method is one of the essential and powerful techniques in recommender system for suggesting the appropriate items to user by learning user's preferences. CF method focuses on user data and generates automatic prediction about user's interests by gathering information from users who share similar background and preferences. Specifically, the intension of CF method is to find users who have similar preferences and to suggest target user items that were mostly preferred by those nearest neighbor users. There are two basic units that need to be considered by CF method, the user and the item. Each user needs to provide his rating value on items i.e. movies, products, books, etc to indicate their interests on those items. In addition, CF uses the user-rating matrix to find a group of users who have similar rating with target user. Then, it predicts unknown rating value for items that target user has not rated. Currently, CF has been successfully implemented in both information filtering and e-commerce applications. However, it remains some important challenges such as cold start, data sparsity, and scalability reflected on quality and accuracy of prediction. In order to overcome these challenges, many researchers have proposed various kinds of CF method such as hybrid CF, trust-based CF, social network-based CF, etc. In the purpose of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which integrates traditional CF technique with social relationship between users discovered from user's behavior in social network i.e. Facebook. We identify user's relationship from behavior of user such as posts and comments interacted with friends in Facebook. We believe that social relationship implicitly inferred from user's behavior can be likely applied to compensate the limitation of conventional approach. Therefore, we extract posts and comments of each user by using Facebook Graph API and calculate feature score among each term to obtain feature vector for computing similarity of user. Then, we combine the result with similarity value computed using traditional CF technique. Finally, our system provides a list of recommended items according to neighbor users who have the biggest total similarity value to the target user. In order to verify and evaluate our proposed method we have performed an experiment on data collected from our Movies Rating S
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
제안 방법
After the experiment about prediction accuracy, in order to evaluate performance of our method we did another experiment of precision, recall, and F1-measure that are well-known for evaluating top-N recommendation. Therefore, we apply precision, recall, and F1-measure metrics with top-N ranging from 5 to 40 to observe how well SRBCF can give a good recommendation result.
Furthermore, we setup an experiment about prediction coverage to measure the domain of items over which the system can make recommendation. For confidently confirming that our proposed method is better in suggesting items to user, we also did the experiment compared with two benchmark algorithms such as Traditional Collaborative Filtering (TCF) and Friendship Fusing Collaborative Filtering (FFCF).
In the aim of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which incorporates social relationship between users discovered from user’s behavior in social network i.e. Facebook1) with traditional CF technique.
In this paper, we propose an effective collaborative filtering method that integrates social relationship of each user discovered from user’s behavior in Facebook with traditional CF in order to improve performance and prediction accuracy of recommender system.
In this section, we will explain in detail about our proposed method which is divided into two parts such as system architecture and our method named as social relationship-based collaborative filtering (SRBCF).
(2009) learns the role of two types of social relationship, membership and friendship, in order to fuse with standard CF to more accurately predict user’s interests and produce better recommendation. Moreover, they presented two different platforms to integrate explicit social relationship into standard CF method, the weightedsimilarity fusion and the graph fusion via random walk, to identify the best performance platform. Weighted-similarity fusion is conducted by leveraging the two social relationships to strengthen user similarity calculation process by combining similarity from friendship and/or membership with similarity from user-rating matrix.
In addition, they use adopting Cosine Correlation in order to calculate the friendship similarity. Next, they calculate the final similarity value between two users by combining similarity calculated from user-rating matrix and similarity computed from user-user matrix in a weighted approach. On the other hand, first they need to obtain user-user similarity based on membership data when fusing the membership.
Second, we did the experiment about MAE value to evaluate the prediction accuracy of our method. Since the size of neighborhood has a significant effect on the prediction accuracy, we calculate the MAE with number of neighbors ranging from 5 to 20.
Pre- diction accuracy evaluation is conducted to dem- onstrate how much our method gives the correctness of recommendation to user in terms of MAE. Then, the evaluation on performance is made to illustrate the effectiveness of SRBCF in terms of precision, recall, and F1-measure. Furthermore, we setup an experiment about prediction coverage to measure the domain of items over which the system can make recommendation.
대상 데이터
After we retrieved movies from IMDB, we have implemented a Movies Rating System in order to allow users to express their favorite movies by rating the movies. The dataset contains 51 users and 2,507 movies. The rating value ranges from 1 to 5 for user to state their favorite movies and only Facebook users are allowed to use the system.
이론/모형
Mean Absolute Error: There are several methods that have been widely used in evaluating the accuracy of recommender system such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Normalized Mean Absolute Error (NMAE). In our approach evaluation, we use Mean Absolute Error metric, a statistic accuracy measure- ment adopted to evaluate recommendation accuracy, to determine the correctness of our proposed method. MAE evaluates the prediction accuracy through computing the value distinction between predicted rating value generated by system and actual rating value provided by user.
Precision, recall, and F1-measure : The “Precision-Recall” method is used to evaluate the recommendation performance.
성능/효과
[Figure 6], [Figure 7], and [Figure 8] are depicted the result of precision, recall, and F1-measure respectively. According to the results, we claim that SRBCF can produce recommendation with better performance than TCF and FFCF while these two benchmark methods can produce recommendation nearly the same performance. Furthermore, precision and recall can be linked to probabilities that directly affect the user.
First, CF collects taste information from many users then using information gleaned from neighbor users for prediction and finally recommendation of user’s interests were automatically generated.
Furthermore, it also shows that user’s rating matrix contributes 30% of the weight while user’s behavior contributes 70% in calculating user-user similarity.
참고문헌 (20)
Ahmed, S., J. W. Kim, and S. G. Kang, "Enhanced Recommendation Algorithm using Semantic Collaborative Filtering : E-commerce Portal," Journal of Intelligence and Information Systems, Vol.17, No.3(2011), 79-98.
Bhuiyan, T., Y. Xu, and A. Josang, SimTrust: A New Method of Trust Network Generation, in Proceeding of 2010 8th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing (EUC), Hong Kong, China, 2010.
Chen, W. and S. Fong, Social Network Collaborative Filtering Framework and Online Trust Factors : A Case Study on Facebook, in Proceeding of Fifth International Conference on Digital Information Management (ICDIM), Thunder Bay, ON, Canada, 2010.
Chen, W., R. Khoury, and S. Fong, "Web 2.0 Recommendation service by multi-collaborative filtering trust network algorithm," Information Systems Frontiers, September, 2012.
Choi, K. H., G. W. Kim, D. H. Yoo, and Y. M. Suh, "New Collaborative Filtering Based on Similarity Integration and Temporal Information," Journal of Intelligence and Information Systems, Vol.17, No.3(2011), 147-168.
Cleverdon, C. and M. Keen, "Factors Determining the Performance of Indexing Systems," ASLIB Cranfield Research Project, Cranfield, 1966.
De Meo, P., E. Ferrara, and G. Fiumara, A. Provetti, Improving Recommendation Quality by Merging Collaborative Filtering and Social Relationships, in Proceeding of International Conference on Intelligent Systems Design and Applications (ISDA), Cordoba, Spain, 2011.
Gao, Y. L., B. Xu, and H. M. Cai, Information Recommendation Method Research Based on Trust Network and Collaborative Filtering, in Proceeding of IEEE 8th International Conference on e-Business Engineering (ICEBE), Beijing, China, 2011.
Ge, M., C. Delgado-Battenfeld, D. Jannach, Beyond Accuracy : Evaluating Recommender Systems by Coverage and Serendipity, in RecSys 2010 Proceeding of the fourth ACM conference on Recommender Systems, Barcelona, Spain, 2010.
Groh, G. and C. Ehmig, Recommendation in Taste Related Domains : Collaborative Filtering vs. Social Filtering, in Proceeding of the 2007 international ACM conference on supporting group work, Sanibel Island, Florida, USA, 2007.
Hameed, M. A., O. A. Jadaan, and S. Ramachandram, "Collaborative Filtering Based Recommendation System : A survey," International Journal on Computer Science and Engineering (IJCSE), Vol.4(2012).
Herlocker, J. L., J. A. Konstan, L. G. Terveen, and J. T. Riedl, "Evaluating collaborative filtering recommender systems," Journal of ACM Transactions on Information Systems (TOIS), Vol.22(2004), 5-53.
Kazienko, P. and K. Musial, "Recommendation Framework for Online Social Networks," Advance in Web Intelligence and Data Mining Studies in Computational Intelligence, Vol.23 (2006), 111-120.
Liu, F. and H. J. Lee, "Use of social network information to enhance collaborative filtering performance," Expert Systems with Applications, Vol.37(2010), 4772-4778.
Mu, X. W., Y. Chen, and T. Y. Li, User-Based Collaborative Filtering Based on Improved Similarity Algorithm, in Proceeding of 3rd IEEE International Conference on Computer Science and Information Technology (ICCSIT), Chengdu, China, 2010.
Park, J. H. and Y. H. Cho, "Social Network Analysis for the Effective Adoption of Recommender Systems," Journal of Intelligence and Information Systems, Vol.17, No.4(2011), 305-316.
Shi, X. Y., H. W. Ye, and S. J. Gong, A Personalized Recommender Integrating Item-based and User-based Collaborative Filtering, in Proceeding of ISBIM 08 International Seminar on Business and Information Management, Wuhan, China, 2008.
Sirawit, S. and B. Kijsirikul, A Step Towards High Quality One-class Collaborative Filtering using Online Social Relationships, in Proceeding of International conference on Advanced Computer Science and Information System (ICACSIS), Jakarta, Indonesia, 2011.
Wang, Q., X. H. Yuan, and M. Sun, Collaborative Filtering Recommendation Algorithm based on Hybrid User Model, in Proceeding of 2010 7th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), Yantai, Shandong, China, 2010.
Yuan, Q., Sh. Zhao, L. Chen, Y. Liu, Sh. Ding, X. Zhang, and W. Zheng, Augmenting Collaborative Recommender by Fusing Explicit Social Relationships, in Proceeding of ACM RecSys '09 Workshop on Recommender Systems and The Social Web, New York, USA, 2009.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.