인터넷에서 매 순간 발생하는 데이터의 홍수는 사용자가 필요로 하는 유용한 정보를 검색하는데 어려움을 초래한다. 그래서 많은 사용자들이 자신이 원하는 정보를 쉽게 찾기 위한 기법을 고안하고 이를 지원하는 도구를 개발하게 되었다. 이런 유용한 도구 중 하나인 추천시스템은 기존의 사용자 정보를 분석하여 사용자가 원하는 제품이나 정보를 추천하는 것이다. 본 논문에서는 추천시스템을 활용하여 원하는 정보를 제안하는데 인구통계학적인 기법을 사용한다. 인구통계학 기반 추천시스템은 나이, 성별과 같은 인구통계학적인 특성을 사용하여 유용한 정보를 추출한다. 본 연구는 영화 선택 시 중요한 요소인 사용자 평점을 분석하고 이를 활용할 수 있는 방법을 제시하였다. 이를 위해 Internet Movie Database(IMDB) 웹 사이트에 있는 영화의 사용자 평점을 인구통계학적 요인으로 분석하였다. 본 논문에서는 인구통계학적 분석을 위해 사용자를 성별과 연령대로 분류하였고, 각 영화를 22개 장르로 나눈 IMDB 기준에 따라 사용자 평점을 분석하였다. 각 장르별 영화에 대해 사용자 그룹의 평균 평점을 F-테스트와 T-테스트를 수행하여 그 장르 영화 평점과 동일한 결과를 나타내는 대표 그룹을 찾아내었다. 인구통계학적 분석 결과인 대표 그룹은 새로운 영화가 개봉될 때 대표 그룹에 대한 프로모션과 추천을 통해 영화 홍보를 할 수 있는 대상을 찾아내는데 유용하다.
인터넷에서 매 순간 발생하는 데이터의 홍수는 사용자가 필요로 하는 유용한 정보를 검색하는데 어려움을 초래한다. 그래서 많은 사용자들이 자신이 원하는 정보를 쉽게 찾기 위한 기법을 고안하고 이를 지원하는 도구를 개발하게 되었다. 이런 유용한 도구 중 하나인 추천시스템은 기존의 사용자 정보를 분석하여 사용자가 원하는 제품이나 정보를 추천하는 것이다. 본 논문에서는 추천시스템을 활용하여 원하는 정보를 제안하는데 인구통계학적인 기법을 사용한다. 인구통계학 기반 추천시스템은 나이, 성별과 같은 인구통계학적인 특성을 사용하여 유용한 정보를 추출한다. 본 연구는 영화 선택 시 중요한 요소인 사용자 평점을 분석하고 이를 활용할 수 있는 방법을 제시하였다. 이를 위해 Internet Movie Database(IMDB) 웹 사이트에 있는 영화의 사용자 평점을 인구통계학적 요인으로 분석하였다. 본 논문에서는 인구통계학적 분석을 위해 사용자를 성별과 연령대로 분류하였고, 각 영화를 22개 장르로 나눈 IMDB 기준에 따라 사용자 평점을 분석하였다. 각 장르별 영화에 대해 사용자 그룹의 평균 평점을 F-테스트와 T-테스트를 수행하여 그 장르 영화 평점과 동일한 결과를 나타내는 대표 그룹을 찾아내었다. 인구통계학적 분석 결과인 대표 그룹은 새로운 영화가 개봉될 때 대표 그룹에 대한 프로모션과 추천을 통해 영화 홍보를 할 수 있는 대상을 찾아내는데 유용하다.
Nowadays, overflowing data produced every second from the internet make people to be difficult to search for the useful information. That's why people have invented and developed unique tools that they get some relevant information. In this paper, the recommender system, one of the effective tools, ...
Nowadays, overflowing data produced every second from the internet make people to be difficult to search for the useful information. That's why people have invented and developed unique tools that they get some relevant information. In this paper, the recommender system, one of the effective tools, is used and it helps us to get the useful information that we want by using demographic information to predict new items of interest. The demographic recommender system in this paper computes users' similarity using demographic information, age and gender. So we performed demographic analysis on movie ratings on Internet Movie Database (IMDB) web site that movies are rated by thousands of people, where users submitted a movie rating after they watched a recent popular film. Meanwhile, we can understand that user's ratings, among various determinants of box office, is very essential factor in the study on recommendation of movie. This paper is aimed at analyzing movie average ratings directly given by film viewers, categorizing them into groups by sex and age, investigating the entire group and finding the representative group by examining it with F-test and T-test. This result is used to promote and recommend for the target group only. Therefore, this study is considerably significant as presenting utilization for movie business as well as showing how to analyze demographic information on movie ratings on the web.
Nowadays, overflowing data produced every second from the internet make people to be difficult to search for the useful information. That's why people have invented and developed unique tools that they get some relevant information. In this paper, the recommender system, one of the effective tools, is used and it helps us to get the useful information that we want by using demographic information to predict new items of interest. The demographic recommender system in this paper computes users' similarity using demographic information, age and gender. So we performed demographic analysis on movie ratings on Internet Movie Database (IMDB) web site that movies are rated by thousands of people, where users submitted a movie rating after they watched a recent popular film. Meanwhile, we can understand that user's ratings, among various determinants of box office, is very essential factor in the study on recommendation of movie. This paper is aimed at analyzing movie average ratings directly given by film viewers, categorizing them into groups by sex and age, investigating the entire group and finding the representative group by examining it with F-test and T-test. This result is used to promote and recommend for the target group only. Therefore, this study is considerably significant as presenting utilization for movie business as well as showing how to analyze demographic information on movie ratings on the web.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
This work is aimed at analyzing the data about movie average ratings directly given by netizens, categorizing them into groups by sex and age, finding and investigating the entire group and the most similar group by genre, and proposing that movie promotion, marketing, and recommendation strategies should be performed on the group.
가설 설정
• Alternative Hypothesis : There is difference in population mean between two groups.
• Null Hypothesis : There is no difference in population mean between two groups.
제안 방법
Distribution of rating data and group with most similar distribution were found by creating boxplot with minimum value, second quartile value, median, fourth quartile, mean, and maximum value acquired through analysis on basic data. Then, a group with most similar value with total ratings in regards to mean, median, standard deviation, and skew value was found.
In this paper, we analyzed user movie ratings based on demographic information and found the representative group of each genre. This study searched the representative group most similar with total ratings through analysis on basic statistics by analyzing the rating data for each gender and age group and proposed to recommend a movie for that representative group after examining it with F-test and T-test. Males Aged 18～29 group was presented to be the group with most influence as it was presented to a group that represents the whole group in 13 genres.
대상 데이터
Data used in this study was extracted from the world largest Internet Movie Database. The IMDB is an internet movie database owned by Amazon.
데이터처리
In order to test whether or not a group most similar with total ratings based on basic statistics can represent total group, T-test which examines the significant difference between the mean of two groups was conducted. [Figure 4] illustrates the procedures to conduct T-test.
이론/모형
This is attribute-based recommendation [11]. In this study, demographic recommendation techniques using IMDB data are used. Recommended the group classified by the target group as age group, gender group.
참고문헌 (28)
Ahn, H., "A Hybrid Collaborative Filtering Recommender System Using a New Similarity Measure," Proceedings of the 6th WSEAS International Conference on Applied Computer Science, 2007.
Bozidar, K., Dijana, O., and Nina, B., "Temporal Recommender Systems," Proceedings of the 10th WSEAS international conference on Applied computer and applied computational science, 2011.
Choi, E., "Analysis of Future Growth in Korea Movie Industry," The Korea Contents Association, Vol. 8, No. 11, pp. 134-143, 2008.
Dellarocas, C., Zhang, X., and Awad, N., "Exploring The Value of Online Product Reviews in Forecasting Sales : The Case of Motion Picture," Journal of Interactive Marketing, Vol. 21, No. 4, pp. 23-45, 2007.
Evangelopoulos, N., "Text Mining for Customer Satisfaction Monitoring," Proceedings of 5th WSEAS Int. Conf. on Simulation, Modeling and Optimization, 2005.
Ha, S., "Helping Online Customers Decide through Web Personalization," IEEE Intelligent Systems, Vol. 17, No. 6, pp. 34-43, 2002.
John, S. and Pokorny, M., An Economic History of Film, Routledge, New York, 2005.
Kazienko, P., "Product Recommendation in E-Commerce Using Direct and Indirect Confidence for Historical User Sessions," Proceedings of 7th International Conference on Discovery Science, 2004.
Kazienko, P. and Kiewra, M., "Integration of Relational Databases and Web Site Content for Product and Page Recommendation," Proceedings of 8th International Database Engineering and Applications Symposium, IDEAS '04, 2004.
Kazienko, P. and Kiewra, M., Intelligent Technologies for Inconsistent Knowledge Processing. Advanced Knowledge International, Adelaide, South Australia, 2004.
Kazienko, P. and Kolodziejski, P., "Personalized Integration of Recommendation Methods for E-commerce," International Journal of Computer Science and Applications, Vol. 3, No. 3, pp. 12-26, 2006.
Konstan, J., Miller, B., Maltz, D., Herlocker, J., Gordon, L., and Riedl, J., "Applying Collaborative Filtering to Usenet News," Communications Of The ACM, Vol. 40, No. 3, pp. 77-87, 1997.
Konstan, J. and Riedl, J., "Recommender Systems : From Algorithms to User Experience," User Model, User-Adapt, Interact, Vol. 22, No. 1-2, pp. 101-123, 2012.
Krulwich, B., "Lifestyle Finder : Intelligent User Profiling Using Large-Scale Demographic Data," AI Magazine, Vol. 18, No. 2, p. 37, 1997.
Lee, J., "How eWOM Reduces Uncertainties in Decision-making Process Using the Concept of Entropy in Information Theory," The Journal of Society for e-Business Studies, Vol. 16, No. 4, pp. 241-256, 2011.
Lee, J. and Myoung, H., "Development of a Book Recommender System for Internet Bookstore using Case-based Reasoning," The Journal of Society for e-Business Studies, Vol. 13, No. 4, pp. 173-191, 2008.
Mooney, R. and Roy, L., "Content-based book recommending using learning for text categorization," Fifth ACM Conference on Digital Libraries, 2000.
Nagy, Z., "AJAX-Based Data Collection Method for Recommender Systems," Proceedings of the 16th WSEAS International Conference on Communications, 2012.
Park, S. and Song, H., "Online reviews and Word-of-Mouth Effect on the movie : The Thrist," Korea Regional Communication Research Association, Vol. 10, No. 4, pp. 157-191, 2010.
Pazzani, M., "A framework for collaborative, content-based and demographic filtering," Artificial Intelligence Review, Vol. 13, No.(5-6), 1999.
Pilaszy, I., Zibriczky, D., and Tikk, D., "Fast ALS-based Matrix Factorization for Explicit and Implicit Feedback Datasets," Proceedings of the fourth ACM conference on Recommender systems, 2010.
Resnick, P., Iacovou, N., Suchak, M., Bergstrom, P., and Riedl, J., "GroupLens : An Open Architecture for Collaborative Filtering of Netnews," Proceedings of the 1994 ACM conference on Computer supported cooperative work, 1994.
Rich, E., "User Modeling via Stereotypes," Cognitive Science, Vol. 3, pp. 329-354, 1979.
Sarwar, B., Karypis, G., Konstan, J., and Riedl, J., "Analysis of Recommendation Algorithms for E-Commerce," Proceedings of the 2nd ACM Conference on Electronic Commerce, 2000.
Vassiliou, C., Stamoulis, D., and Martakos, D., "A Recommender System Framework combining Neural Networks and Collaborative Filtering," Proceedings of the 5th WSEAS Int. Conf. on Instrumentation, Measurement, Circuits and Systems, 2006.
Vozalis, M. and Margaritis, K., "Using SVD and Demographic Data for the Enhancement of Generalized Collaborative Filtering," Information Sciences, Vol. 177, No. 15, pp. 3017-3037, 2007.
Wang, H. and Wu, C., "A Strategy-Oriented Operation Module for Recommender Systems in E-Commerce," Proceedings of the 9th WSEAS International Conference on Applied Informatics and Communications, 2009.
Yang, W., Wang, Z., and You, M., "An improved collaborative filtering method for recommendations generation," 2004 IEEE International Conference on Systems, Man and Cybernetics, 2004.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.