• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.27-40
• /
• 2023

With the increase in the spread of smart devices and the impact of COVID-19, the consumption of media contents through smart devices has significantly increased. Along with this trend, the amount of media contents viewed through OTT platforms is increasing, that makes contents recommendations on these platforms more important. Previous contents-based recommendation researches have mostly utilized metadata that describes the characteristics of the contents, with a shortage of researches that utilize the contents' own descriptive metadata. In this paper, various text data including titles and synopses that describe the contents were used to recommend similar contents. KLUE-RoBERTa-large, a Korean language model with excellent performance, was used to train the model on the text data. A dataset of over 20,000 contents metadata including titles, synopses, composite genres, directors, actors, and hash tags information was used as training data. To enter the various text features into the language model, the features were concatenated using special tokens that indicate each feature. The test set was designed to promote the relative and objective nature of the model's similarity classification ability by using the three contents comparison method and applying multiple inspections to label the test set. Genres classification and hash tag classification prediction tasks were used to fine-tune the embeddings for the contents meta text data. As a result, the hash tag classification model showed an accuracy of over 90% based on the similarity test set, which was more than 9% better than the baseline language model. Through hash tag classification training, it was found that the language model's ability to classify similar contents was improved, which demonstrated the value of using a language model for the contents-based filtering.

• International Journal of Contents
• /
• v.13 no.2
• /
• pp.66-74
• /
• 2017

Patent classification is becoming more critical as patent filings have been increasing over the years. Despite comprehensive studies in the area, there remain several issues in classifying patents on IPC hierarchical levels. Not only structural complexity but also shortage of patents in the lower level of the hierarchy causes the decline in classification performance. Therefore, we propose a new method of classification based on different criteria that are categories defined by the domain's experts mentioned in trend analysis reports, i.e. Patent Landscape Report (PLR). Several experiments were conducted with the purpose of identifying type of features and weighting methods that lead to the best classification performance using Support Vector Machine (SVM). Two types of features (noun and noun phrases) and five different weighting schemes (TF-idf, TF-rf, TF-icf, TF-icf-based, and TF-idcef-based) were experimented on.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.389-394
• /
• 2004

In this paper, we classified design patterns with special quality of pattern structure. Classification by clustering had expressed higher correctness degree than classification by facet. Therefore, can do that it is effective that classify design patterns using clustering algorithms that is automatic classification method. When we are searching design patterns, classification of design patterns can compare and analyze similar patterns because similar patterns is saved to same category. Also we can manage repository efficiently because of using and storing link information of patterns.

• International Journal of Contents
• /
• v.2 no.3
• /
• pp.11-17
• /
• 2006

The Clustering is representative method of components classification. The previous clustering methods that use cohesion and coupling cannot be effective because design pattern has focused on relation between classes. In this paper, we classified design patterns with features of object-oriented relationship. The result is that classification by clustering showed higher precision than classification by facet. It is effective that design patterns are classified by automatic clustering algorithm. When patterns are retrieved in classification of design patterns, we can use to compare them because similar pattern is saved to same category. Also we can manage repository efficiently because of storing patterns with link information.

• The Journal of the Korea Contents Association
• /
• v.19 no.11
• /
• pp.499-506
• /
• 2019

Virtual reality, which is attracting attention due to the 4th Industrial Revolution and commercialization of 5G technology, is expanding its scope from gaming to tourism, leisure, and education, and the VR market is expected to expand continuously. As the VR market scales up in Korea, theme parks combining virtual reality contents are spreading around the city center. Unlike the existing theme parks, VR Theme Park is a small amusement culture space that is organized indoors, and you can enjoy attractions (ride) that can be enjoyed in an amusement park with virtual reality contents. Virtual reality content, which has the same characteristics as a theme park whose purpose is to experience extraordinary experiences, provides high immersion and presence in combination with the physical stimulus of attraction. The virtual reality content combined with the attraction cannot be classified accurately with the existing classification method, so a new classification method is proposed according to the experience type and the installation type. The contents were categorized through the case of the domestic VR theme park, and the planning direction for the creation of the virtual reality attraction contents that was going on was sought.

• The Journal of the Korea Contents Association
• /
• v.9 no.3
• /
• pp.108-117
• /
• 2009

Current mobile content's classifications have two major problems. One violates a principle of classification and the other reveals limitations in dealing with various convergence services. This study proposes a new mobile content's classification system to resolve these problems by adapting Al Ries's principle of symmetric and asymmetric transition. Symmetric mobile contents take a form of mutation in convergence process; therefore, the contents would appear different from their originals whereas asymmetric type combines mobile contents in an autonomous way. This new system not only demonstrates a clearer classification but also implies the trend of mobile content development and services. The current suggests that symmetric type is preferable and symmetric type of mobile contents is re-developed to become a symmetric type as much as the technology can support. Nonetheless, it is found that asymmetric type would still be serviced to some extent. Thus, new mobile content's classification, proposed in this research, provides a more constructive understating of mobile content's directions in the era of digital convergence and a ground for comparative analysis of mobile content's development or positioning strategies.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.7
• /
• pp.4942-4949
• /
• 2015

The construction information classification which developed for the purpose of adapting informatization in construction industry suggests construction information with standardization. Currently, standardization of construction information is highly necessary for BIM that is main background of alteration as construction industry informatization. The authors suggest improvement of construction information classification. Particularly, extension contents for each facets are suggested to contain BIM. In case of applying extended classification to BIM, interoperability of information will be enhanced and it is effective to integrate information in phase of using BIM.

• The Journal of the Korea Contents Association
• /
• v.7 no.1
• /
• pp.215-224
• /
• 2007

The classification of the satellite images is basic part in Remote sensing. In classification of the satellite images, class separability feature is very effective accuracy of the images classified. For improving classification accuracy, It is necessary to study classification methode than analysis of class separability feature deciding classification probability. In this study, IKONOS, SPOT 5, Landsat TM, were resampled to sizes 1m grid. Above images were calculated the class separability prior to the step for classification of pixels. This Study concludes, each image was measured by the rate of class separability, values classified were showed highly about $1,600{\sim}2,000$.

• 한국체육학회지인문사회과학편
• /
• v.55 no.2
• /
• pp.111-121
• /
• 2016

The purpose of this study was to establish a classification systems of sport culture contents for enhancing cultural competitiveness. The main research method for this was the Delphi survey based on the literature. Through a literature review, the definition of sport cultural contents and the classification system of major industrial countries were analyzed. The Delphi survey by a panel of 25 was conducted over a two-round. The results were as follows: first, the sport culture contents includes an artistic creativity, which have a working form that enables delivery to the public and organizing sports activities, and classify as a technology for high value-added content and services. Second, the 'culture' for the large category item, which include the cultural heritage and culture art. The 'media' recognize the importance of temporal concepts with traditional media and it was classified as new media. In the case of items in the large category 'events' are classified as national and international sporting events and festivals, the 'services' main category item is divided into social, care, educational services. Finally, the 'others' includes such as copyrighted items.

• Journal of Korea Multimedia Society
• /
• v.12 no.6
• /
• pp.867-875
• /
• 2009

More and richer information sources and services are available on the web everyday. However, harmful information, such as adult content, is not appropriate for all users, notably children. Since internet is a worldwide open network, it has a limit to regulate users providing harmful contents through each countrie's national laws or systems. Additionally it is not a desirable way of developing a certain system-specific classification technology for harmful contents, because internet users can contact with them in diverse ways, for example, porn sites, harmful spams, or peer-to-peer networks, etc. Therefore, it is being emphasized to research and develop context-based core technologies for classifying harmful contents. In this paper, we propose an efficient text filter for blocking harmful texts of web documents using context-based technologies and examine which algorithms for feature selection, the process that select content terms, as features, can be useful for text categorization in all content term occurs in documents, are suitable for classifying harmful contents through implementation and experiment.