• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.185-198
• /
• 2021

With the development of Internet technology, because traditional recommendation algorithms cannot learn the in-depth characteristics of users or items, this paper proposed a recommendation algorithm based on the AMITI(attention mechanism and improved TF-IDF) to solve this problem. By introducing the two-layer attention mechanism into the CNN, the feature extraction ability of the CNN is improved, and different preference weights are assigned to item features, recommendations that are more in line with user preferences are achieved. When recommending items to target users, the scoring data and item type data are combined with TF-IDF to complete the grouping of the recommendation results. In this paper, the experimental results on the MovieLens-1M data set show that the AMITI algorithm improves the accuracy of recommendation to a certain extent and enhances the orderliness and selectivity of presentation methods.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.15 no.3
• /
• pp.111-118
• /
• 2019

Chapter 1 and 2 of this study examined the methodology of modern storytelling using the hero narrative structure. This involved analyses of Joseph Campbell's monomyth structure, which is the prototype model of modern storytelling and Christopher Vogler's story structure in which the monomyth structure is classified into 3 Acts and 12 Stages. In addition, the movie 'Captain America' was analyzed based on Vogler's narrative structure theory. According to the analyses results, the hero of 'Captain America' fully follows Christopher Vogler's hero narrative structure but in some cases, he does not follow the existing hero narrative structure. It is interpreted that this is because heroes of modern tales have different birth backgrounds from mythical heroes. There is also a difference even in the stage where a hero completes all adventures and returns home between modern tales and myths due to different social backgrounds. Therefore, these findings provide a basis for modification and supplementation of a modern hero epic. Through this analysis, the modified modern hero narrative structure is evaluated to be appropriate as a basic model for modern storytelling. Further, it is expected that those who frame a film script can complete a new hero epic while following the structure of syntagmatic systems by selecting a level among Northrop Frye's paragmatic systems based on this structure and considering story themes and heroes' personality and characteristics.

• Journal of KIISE:Software and Applications
• /
• v.36 no.4
• /
• pp.298-305
• /
• 2009

In contrast to content_based filtering systems, collaborative filtering systems not only don't contain information of items, they can not recommend items when users don't provide the information of their interests. In this paper, we propose the recommender agent using association item tree to solve the shortcomings of collaborative filtering systems. Firstly, the proposed method clusters users into groups using vector space model and K-means algorithm and selects group typical rating values. Secondly, the degree of associations between items is extracted from computing mutual information between items and an associative item tree is generated by group. Finally, the method recommends items to an active user by using a group typical rating value and an association item tree. The recommender agent recommends items by combining user information with item information. In addition, it can accurately recommend items to an active user, whose information is insufficient at first rate, by using an association item tree based on mutual information for the similarity between items. The proposed method is compared with previous methods on the data set of MovieLens recommender system.

• Cartoon and Animation Studies
• /
• s.19
• /
• pp.89-112
• /
• 2010

Today many countries consider the culture and technology industry as a high value-added industry, and make intensive investments into cultural content for its growth. In the culture and technology industry, the OSMU (One Source Multi Use) business structure, which creates content using diverse media based on one original work and thus generates a high added value, has received much attention. Recently Korea has also witnessed an increase in the cases in which the cartoon medium is converted into movies or drama. Cartoons are now put to the most use as an original source, in which the process of active secondary reproduction and recreation are taking place. In the process of recreating original works, media characteristics function as an important variable. Stories are newly reproduced by the detailed information (form) characteristics of each medium. Despite those developments, however, there are not as many active comparative studies on media-converted works as there are actual productions of them, which calls for a need to more fully activate specific comparisons of cases. By comparing and investigating original works and secondarily recreated ones through media conversion, the study made comparisons between the cartoon and movie medium and described the common and different features of the two media in terms of detailed information characteristics. In order to provide a specific case, it also analyzed the cartoon Tajja and movie Tajja through media conversion, focusing on 'conversion of narrative structure.'

• Journal of Convergence for Information Technology
• /
• v.10 no.10
• /
• pp.40-46
• /
• 2020

With the development of the Internet, users share their experiences and opinions. Since related keywords are used witho0ut considering information such as the general emotion or genre of an unstructured document such as a movie review, the sensitivity accuracy according to the appropriate emotional situation is impaired. Therefore, we propose a system that predicts emotions based on information such as the genre to which the unstructured document created by users belongs or overall emotions. First, representative keyword related to emotion sets such as Joy, Anger, Fear, and Sadness are extracted from the unstructured document, and the normalized weights of the emotional feature words and information of the unstructured document are trained in a system that combines CNN and LSTM as a training set. Finally, by testing the refined words extracted through movie information, morpheme analyzer and n-gram, emoticons, and emojis, it was shown that the accuracy of emotion prediction using emotions and F-measure were improved. The proposed prediction system can predict sentiment appropriately according to the situation by avoiding the error of judging negative due to the use of sad words in sad movies and scary words in horror movies.

• The Journal of the Korea Contents Association
• /
• v.22 no.3
• /
• pp.152-164
• /
• 2022

This thesis explores the relationship between Dominic Cooke's film (2017) and classical music. To analyze the relationship, researchers applied precedent research to the study. The relationship between the final scene of the movie King's Speech (2010) and the volume and instrumental changes of the Beethoven Symphony is analyzed by David Bashwiner, and Soohwan Ahn analyzed semantic association between the hotel conversation scene in a and Debussy's Arabesque. In addition, the study of application of Schumann's Träumerei to films was used as a methodology to find out how extra-musical information build meaningful sonority. Mozart's K.593, Haydn's Op.77 No.1, and Schubert's D.810 were used in the movie . This study analyzed the functions of Mozart, Haydn, and Schubert's music in . In order to express the relationship between the characters and their inner intentions, this film utilized the relationship between instruments, musical information and non-musical information of the pieces. Through this study, it is analyzed that the information of classical music functions and the core information of the plot of the movie combine together to improve the understanding of narrative.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.441-450
• /
• 2003

Collaborative filtering is one of the most widely used methodologies for recommendation system. Collaborative filtering is based on a data matrix of each customer's preferences and frequently, there exits missing data problem. We introduced two imputation approach (multiple imputation via Markov Chain Monte Carlo method and multiple imputation via bootstrap method) to improve the prediction performance of collaborative filtering and evaluated the performance using EachMovie data.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.285-287
• /
• 1998

본 논문에서는 다양한 멀티미디어 스트림(multimedia stream)을 재생할 수 있는DirectShow SDK를 이용하여, 비디오의 특수효과를 구현하기 위한 변환 필터(Transform filter) 제작 방법에 대해 소개한다. DirectShow는 여러 가지 포맷(MPEG, AVI, MOV, WAV)으로 부호화된 digital movie나 sound를 재생할 수 있는 run-time(.ock)과 dynamic-link library (DLLs)를 제공하여 필터의 제작을 용이하게 한다. 본 논문에서는 디지털 영상 특수효과 중에서 Clip이라는 특수효과를 예로하여 변환 필터를 제작하는 방법을 소개한다. 또한 DirectShow Documentation에서 제공된 변환방법을 수정하여 변환필터의 입력 pin에서 출력 pin으로의 불필요한 복사를 줄임으로써 수행 속도를 개선하는 특수효과 구현 방법을 제시한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.25-27
• /
• 2003

MPEG-7은 XML을 이용하여 사용자가 원하는 내용을 검색하고 브라우징할 수 있도록 하는 멀티미디어 컨텐츠의 다양한 정보를 기술하고 있다. 그러나 많은 부분으로 나누어진 세부화된 장면들을 찾는 질의가 주어졌을때 원하는 정보를 찾기 위하여 모든 문서의 내용을 검색하는 것은 상당한 시간이 요구되며, 일반적인 XML 문서의 인덱싱 기법을 적용할 경우 MPEG-7의 관점에서 부합되지 않은 점이 있다. 본 논문에서는 이를 개선하기 위하여 MPEG-7 문서의 구조적인 정보를 기반으로 하는 인덱싱 기법을 제안하고 질의를 하여 동영상내의 원하는 부분을 효율적으로 찾는 방법을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.460-462
• /
• 2003

본 논문에서는 모바일 환경에서. 사용자의 상황정보와 개인적 선호도 정보를 고려하여 사용자에게 적합한 영화를 선택. 추천하고 선택된 영화의 요약(Summary)을 서비스 할 수 있는 시스템을 제안한다. 제안된 시스템은 사용자가 이동하는 상황에 따라 변하는 위치, 시간 정보와 개인적 선호도인 영화장르 정보에 기반한 영화선택 서비스를 제공하고, 선택된 영화 콘텐츠의 요약을 MPEG-7 메타데이터로 기술하고, 이를 이용해 요약을 효과적으로 소모할 수 있게 한다. 제안된 시스템을 통해, 모바일 환경 기반 영화 선택 및 서비스 시스템(Mobile Environment-Based Movie Selection and Summary Service System)을 실현하고, 그 효용성을 입증하였다.