• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.2
• /
• pp.218-223
• /
• 2011

Various research studies are underway to explore music classification by genre. Because sound professionals define the criterion of music to categorize differently each other, those classification is not easy to come up clear result. When a new genre is appeared, there is onerousness to renew the criterion of music to categorize. Therefore, music is classified by emotional adjectives, not genre. We classified music by light and shade in precedent study. In this paper, we propose the music classification system that is based on emotional adjectives to suitable search for atmosphere, and the classification criteria is three kinds; light and shade in precedent study, intense and placid, and grandeur and trivial. Variance Considered Machines that is an improved algorithm for Support Vector Machine was used as classification algorithm, and it represented 85% classification accuracy with the result that we tried to classify 525 songs.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.373-375
• /
• 2023

본 논문에서는 멀티 레이블을 이용한 CNN 구조 활용과 NLP 학습을 이용하여 한국 영화의 장르를 예측하는 방법을 제안한다. 포스터는 영화의 전반적인 내용을 한눈에 알아볼 수 있게 하는 매체이기 때문에 다양한 요소들로 구성되어 있다. 합성곱 신경망(Convolutional neural network)을 활용해, 한국 영화 포스터가 가지는 특징들을 추출하여 영화 장르 분류를 진행하였다. 하지만, 영화의 경우 감독이 생각하는 장르와 관객이 영화를 봤을 때, 느끼는 장르가 다를 수 있다. 그렇기 때문에 장르 예측에 있어서 문제가 발생할 수 있다. 이러한 문제를 완화하기 위해 본 논문에서는 합성곱 신경망 활용뿐만 아니라, 자연어 처리(Natural Language Processing)를 같이 활용한 방법을 제안한다.

• Archives of design research
• /
• v.16 no.3
• /
• pp.351-360
• /
• 2003

The concept of genre used to be applied to classify the fine arts also can investigate various way of communication system and classification. The structuring element of genre was identified by form and contents in that field. But the classification of the web, which is new communication tool, was made by defining the purpose of the web. In this paper the genre system, which consists of form, contents, and function, is applied to classify the web so that we offer tile opportunity to identify dearer characteristics of it. In order to investigate the genre elements in the communication process the structure of the semiotic triad after Charles S. Peirce was adapted, which was labeled as representamen, object, and interpretant. The representamen substitutes for the web function, the object does for the form of the web, and the representamen does for the web contents. According to the Peirce's the representamen identify the object but on the other hand it is identified by the interpretant. Logical structure of the fact that form of the web is identified by its function, and the function is identified by the contents is proved by following the theory. Therefore, the concept of web genre is supported by the element of genre having a logical structure activating in the communication process. We suggest that in recent complicated communication circumstance the genre concept should be adapted to implement the effective web communication design.

• Journal of Digital Convergence
• /
• v.12 no.6
• /
• pp.325-332
• /
• 2014

It is generally accepted that classification accuracy is affected by the number of learning documents, but there are few studies that show how this influences automatic text classification. This study is focused on evaluating the deviation-based classification model which is developed recently for genre-based classification and comparing it to other classification algorithms with the changing number of training documents. Experiment results show that the deviation-based classification model performs with a superior accuracy of 0.8 from categorizing 7 genres with only 21 training documents. This exceeds the accuracy of Bayesian and SVM. The Deviation-based classification model obtains strong feature selection capability even with small number of training documents because it learns subject information within genre while other methods use different learning process.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.197-200
• /
• 2005

본 논문은 유해한 동영상을 실시간으로 분석하고 차단하기 위하여, 동영상의 비주얼 특징으로서 그룹 프레임(Group of Frame) 특징을 추출하여 SVM 학습모델을 활용하는 유해 동영상 분류에 관한 것이다. 지금까지 동영상 분류에 관한 연구는 주로 입력 동영상을 뉴스, 스포츠, 영화, 뮤직 비디오, 상업 비디오 등 사전에 정의한 몇 개의 장르에 자동으로 할당하는 기술이었다. 그러나 이러한 분류 기술은 미리 정의한 장르에 따른 일반적인 분류 모델을 사용하기 때문에 분류의 정확도가 높지 않다. 따라서, 유해 동영상을 실시간으로 자동 분류하기 위해서는, 신속하고 효과적인 동영상 내용분석에 적합한 유해 동영상 특화의 특징 추출과 분류 모델 연구가 필요하다. 본 논문에서는 유해 동영상에 대하여 신속하고, 정확한 분류를 위하여 유해 동영상의 대표 특징으로서 그룹프레임 특징을 정의하고, 이를 추출하여 SVM 학습 모델을 생성하고 분류에 활용하는 매우 높은 성능의 분석 방법을 제시하였다. 이는 최근 인터넷 뿐만 아니라 다양한 매체를 통하여 급속도로 번지고 있는 유해 동영상 차단 분야에 적극 활용될 수 있을 것으로 기대된다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.3 s.303
• /
• pp.115-120
• /
• 2005

In this paper, we propose a new robust content-based musical genre classification algorithm using multi-feature clustering(MFC) method. In contrast to previous works, this paper focuses on two practical issues of the system dependency problem on different input query patterns(or portions) and input query lengths which causes serious uncertainty of the system performance. In order to solve these problems, a new approach called multi-feature clustering(MFC) based on k-means clustering is proposed. To verify the performance of the proposed method, several excerpts with variable duration were extracted from every other position in a queried music file. Effectiveness of the system with MFC and without MFC is compared in terms of the classification accuracy. It is demonstrated that the use of MFC significantly improves the system stability of musical genre classification performance with higher accuracy rate.

• Annual Conference on Human and Language Technology
• /
• 2009.10a
• /
• pp.44-49
• /
• 2009

통계기계번역에서 도메인에 특화된 번역을 시도하여 성능향상을 얻는 방법이 있다. 이를 위하여 문장의 유형이나 장르에 따라 클러스터링을 수행한다. 그러나 기존의 연구 중 문장의 유형 정보와 장르에 따른 정보를 동시에 사용한 경우는 없었다. 본 논문에서는 문장 사이의 문법적 구조 유사성으로 문장을 유형별로 분류하는 새로운 기법을 제시하였고, 단어 유사도 정보로 문서의 장르를 구분하여 기존의 두 기법을 통합하였다. 이렇게 분류된 말뭉치에서 추출한 모델과 전체 말뭉치에서 추출된 모델에서 보간법(interpolation)을 사용하여 통계기계번역의 성능을 향상하였다. 문장구조의 유사성과 단어 유사도 계산을 위하여 각각 커널과 코사인 유사도를 적용하였으며, 두 유사도를 적용하여 말뭉치를 분류하는 과정은 K-Means 알고리즘과 유사한 기계학습 기법을 사용하였다. 이를 일본어-영어의 특허문서에서 실험한 결과 최선의 경우 약 2.5%의 상대적인 성능 향상을 얻었다.

• The KIPS Transactions:PartB
• /
• v.14B no.2
• /
• pp.89-98
• /
• 2007

For browsing, searching, and manipulating video documents, an indexing technique to describe video contents is required. Until now, the indexing process is mostly carried out by specialists who manually assign a few keywords to the video contents and thereby this work becomes an expensive and time consuming task. Therefore, automatic classification of video content is necessary. We propose a fully automatic and computationally efficient method for analysis and summarization of spots news video for 5 spots news video such as soccer, golf, baseball, basketball and volleyball. First of all, spots news videos are classified as anchor-person Shots, and the other shots are classified as news reports shots. Shot classification is based on image preprocessing and color features of the anchor-person shots. We then use the dominant color of the field and motion features for analysis of sports shots, Finally, sports shots are classified into five genre type. We achieved an overall average classification accuracy of 75% on sports news videos with 241 scenes. Therefore, the proposed method can be further used to search news video for individual sports news and sports highlights.

• The KIPS Transactions:PartB
• /
• v.11B no.5
• /
• pp.555-562
• /
• 2004

A genre or a style is another view of documents different from a subject or a topic. The style is also a criterion to classify the documents. There have been several studies on detecting a style of textual documents. However, only a few of them dealt with web documents. In this paper we suggest sets of features to detect styles of web documents. Web documents are different from textual documents in that Dey contain URL and HTML tags within the pages. We introduce the features specific to web documents, which are extracted from URL and HTML tags. Experimental results enable us to evaluate their characteristics and performances.

• Journal of Korea Game Society
• /
• v.10 no.3
• /
• pp.61-71
• /
• 2010

This study looks into the usage pattern in online games based on genres and socio-demographic characteristics. Compared to the prior studies that adopted survey as their main research method, this study has analyzed the actual data of game login records and adopted parametric modeling and mathematical approach. In terms of the socio-demographic characteristics, the following facts were confirmed: men > women by gender, students > white-collars > housewives > blue-collars > self-employed > jobless(etc.) by occupation, college graduates > K-12 students > high-school graduates > undergrads & grads by academic background, 3∼5 million > 1∼3 million > over 5 million > less than 1 million by income levels, and not married > married by marital status. In terms of genres, the population of the players is in the order of web board games, RPG, action/racing/shooting, and sports. The RPG game is confirmed to have a higher level of MCR (Max Concurrent User Ratio) than any other genres. On the other hand, the hypothesis on the difference in Repeated Use Ratio according to genres is rejected. This study has also confirmed that interactions exist between gender and age; genre and gender; genre and age among online game users, and conducted post-hoc analysis about those interactions.