• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.312-319
• /
• 2021

This study aims to help users to have a more satisfying encounter based on the problems found by comparing and analyzing similar applications. That is, an application that derives intermediate points through the subway, which is a public transportation means, and provides information on nearby convenience facilities was proposed. The middle point calculation process uses the dijkstra algorithm, which stores the minimum number of nodes in the stored path from the first input location to the last location. The stack and arraylist are used to search all paths from the first input position to the last position, and then the path with the smallest number of nodes is selected. After that, the number of stations in the route is divided in half and the resulting station is output. In addition, this study provides information on convenience facilities near intermediate points in order to have differences from similar applications. It categorizes within a 1km radius of the point and provides a function that helps to conveniently identify only facilities around the middle point. In particular, by visualizing the number of convenience facilities with radar charts and numbers, it is possible to grasp the commercial district around the midpoint at a glance.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1114-1121
• /
• 2021

In this paper, we implement a recipe recommendation system using ingredient harmonization analysis based on recipe data. The proposed system receives an image of a food ingredient purchase receipt to recommend ingredients and recipes to the user. Moreover, it performs preprocessing of the receipt images and text extraction using the OCR algorithm. The proposed system can recommend recipes based on the combined data of ingredients. It collects recipe data to calculate the combination for each food ingredient and extracts the food ingredients of the collected recipe as training data. And then, it acquires vector data by learning with a natural language processing algorithm. Moreover, it can recommend recipes based on ingredients with high similarity. Also, the proposed system can recommend recipes using replaceable ingredients to improve the accuracy of the result through preprocessing and postprocessing. For our evaluation, we created a random input dataset to evaluate the proposed recipe recommendation system's performance and calculated the accuracy for each algorithm. As a result of performance evaluation, the accuracy of the Word2Vec algorithm was the highest.

• Journal of Korea Multimedia Society
• /
• v.24 no.12
• /
• pp.1673-1681
• /
• 2021

Item cold start is a well studied problem in the research field of recommender systems. Still, many existing collaborative filters cannot recommend items accurately when only a few user-item interaction data are available for newly introduced items (Cold items). We propose a interaction feature prediction method to mitigate item cold start problem. The proposed method predicts the interaction features that collaborative filters can calculate for the cold items. For prediction, in addition to content features of the cold-items used by state-of-the-art methods, our method exploits the interaction features of k-nearest content neighbors of the cold-items. An attention network is adopted to extract appropriate information from the interaction features of the neighbors by examining the contents feature similarity between the cold-item and its neighbors. Our evaluation on a real dataset CiteULike shows that the proposed method outperforms state-of-the-art methods 0.027 in Recall@20 metric and 0.023 in NDCG@20 metric.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.999-1006
• /
• 2022

The use of appropriate fonts is not only an aesthetic point of view, but also a factor influencing the reinforcement of meaning. However, it is a difficult process and wastes a lot of time for general users to choose a font that suits their needs and emotions. Therefore, in this study, keywords and fonts to be used in the experiment were selected for emotion-based font recommendation, and keyword values for each font were calculated through an experiment to check the correlation between keywords and fonts. Using the experimental results, a prototype of a keyword-based font recommendation system was designed and the possibility of the system was tested. As a result of the usability evaluation of the font recommendation system prototype, it received a positive evaluation compared to the existing font search system, but the number of fonts was limited and users had difficulties in the process of associating keywords suitable for their desired situation. Therefore, we plan to expand the number of fonts and conduct follow-up research to automatically recommend fonts suitable for the user's situation without selecting keywords.

• Journal of Digital Contents Society
• /
• v.17 no.5
• /
• pp.307-316
• /
• 2016

A user usually gets the same suggesting results as everyone else in most of the multimedia social services, nowadays. To address the challenging problem of personalization in the social network, we propose a method which exploits user's activities, user's moods, and user's friend relationships from the social network to build a decision-making system. Depending on a current state of the user's mood, this system infers the most appropriated video for the user. In the system, the user evaluates a set of the given recommendation methods which extract from the user's database social network and assigns a vague value to each method by a weight. Then, we find the fuzzy collection solution for the system and classify the set of methods into subsets, and order the subsets based on its local dominance to choose the best appropriate method. Finally, we conduct an experiment using the YouTube API with a lot of video types. The experiment result shows that the channel recommendation system appropriately affords the user's character, it is more satisfying than the current YouTube based on an evaluation of several users.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.617-618
• /
• 2014

본 논문에서는 보다 정확하고 효과적인 평판분석을 위하여 서비스 산업별 타겟으로 하는 수집사이트를 프로파일링 하는 방법을 제안한다. 먼저 각 서비스에 특화된 타겟 사이트를 추출하고 등록하고 각 서비스에 관련한 정보 및 의견 공유 게시판과 지식인 추천/질문 등 지식 공유 사이트를 추출한다. 또한 업종별 주요 사이트를 선택하고 등록하여 유효 데이터 수집한다. 이를 통해 실시간 수집 데이터의 활용 기술을 이용하여 수집원 프로파일링을 통한 미디어별 수집 주기 산정하고 수집 엔진의 유연한 확장성을 활용한 실시간 수집 제반 기술 확대할 수 있다. 또한 지속적인 수집원 변경관리를 수행한다. 즉, 신규 생성, 변경, 삭제되는 사이트에 대한 변경관리를 수행하고 지속적인 수집량 모니터링을 통한 수집여부를 점검하며 수집 필터링 규칙에 대한 튜닝으로 데이터 품질 확보하도록 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.729-732
• /
• 2007

맞춤형 방송 솔루션은 디지털 TV 서비스 외에도, 향후 유망기술로 주목 받고 있는 IPTV, DMB 등의 다양한 방송 및 멀티미디어 서비스에 적용이 가능하다. 본 연구에서는 맞춤형 방송 서비스의 객관적인 평가를 위한 성능평가 방법을 제안한다. 정보검색 시스템 평가에 사용하는 정확률, 재현률과 역순위 평균 수치를 적용하여 맞춤형방송의 추천결과와 실제 시청자의 시청 프로그램의 차이를 분석하여 맞춤형 방송 시스템의 정확도와 사용자 만족도를 통계적으로 평가 가능한 메카니즘을 제안한다. 그 동안 평가가 이루어지지 않았던 맞춤형 방송 서비스를 복합적으로 평가하는 방법론을 제안함으로써, 맞춤형 방송 시스템의 지속적인 성능향상과 연구개발에 기여할 것이다. 또한, 맞춤형 방송 서비스의 산업화와 다양한 장비로의 확산에도 기여할 것으로 기대된다.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.391-406
• /
• 2024

The current distribution of digital news has the potential to produce politically biased information for users as a result of individual choices and media selection based on those choices. Consequently, this research explored the factors affecting individual news selection and the effects of opinion changes and political participation that can occur when news tailored to users' partisan preferences is recommended. The phenomenon of selective exposure has been shown to be stronger when individuals utilize more limited information processing, experience higher discussion efficacy among groups with similar political beliefs. Furthermore, When a selective exposure group was randomly provided with a one-way message news that matched their partisan leanings, it was found that opinion consolidation, opinion-reinforcing information processing, and online political participation. On the other hand, when they were randomly presented with two-way messaging news in which the media balanced two competing partisan positions, they were found to be more likely to understand the other side's views and arguments, and more willing to adjust their existing opinions. We are confirmed that the balanced use of various opinions is very important in deliberative democratic process.

• Journal of Broadcast Engineering
• /
• v.20 no.6
• /
• pp.848-861
• /
• 2015

In this paper, we introduce a new web-based PR video making service system. Many video editing tools have required tough editing skill or scenario planning stage for a just simple PR video making. Some users may prefer a simple and fast way than sophisticated and complex functionality. To solve this problem, it is important to provide easy user interface and intelligent classification and recommendation scheme. Therefore, we propose a new template classification and recommendation scheme using a topic modeling method. The proposed scheme has the big advantage of being able to handle the unstructured meta data as well as structured one.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.1024-1034
• /
• 2019

Preference for music is determined by a variety of factors, and identifying characteristics that reflect specific factors is important for music recommendations. In this paper, we propose a method to extract the singing voice related music features reflecting various musical characteristics by using a model learned for singer identification. The model can be trained using a music source containing a background accompaniment, but it may provide degraded singer identification performance. In order to mitigate this problem, this study performs a preliminary work to separate the background accompaniment, and creates a data set composed of separated vocals by using the proven model structure that appeared in SiSEC, Signal Separation and Evaluation Campaign. Finally, we use the separated vocals to discover the singing voice related music features that reflect the singer's voice. We compare the effects of source separation against existing methods that use music source without source separation.