• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.193-198
• /
• 2023

We study a learner-customized lecture recommendation project using deep learning. Recommendation systems can be easily found on the web and apps, and examples using this feature include recommending feature videos by clicking users and advertising items in areas of interest to users on SNS. In this study, the sentence similarity Word2Vec was mainly used to filter twice, and the course was recommended through the Surprise library. With this system, it provides users with the desired classification of course data conveniently and conveniently. Surprise Library is a Python scikit-learn-based library that is conveniently used in recommendation systems. By analyzing the data, the system is implemented at a high speed, and deeper learning is used to implement more precise results through course steps. When a user enters a keyword of interest, similarity between the keyword and the course title is executed, and similarity with the extracted video data and voice text is executed, and the highest ranking video data is recommended through the Surprise Library.

• Proceedings of the Korean Society for Information Management Conference
• /
• 2004.08a
• /
• pp.167-172
• /
• 2004

기존의 문서분류는 학습문서에 출현하는 자질에 대해 가중치를 계산하여 그 순위에 따라 상위 자질로 구성된 지식베이스를 사용하였다. 그리고 새로운 문서가 들어왔을 때 자질 지식베이스를 근거로 새 문서를 색인하였다. 결국 자질 지식베이스와 정확히 일치하지 않는 키워드는 색인대상에서 제외되는 문제가 있었다. 본 고에서는 이 문제를 해결하기 위하여 분류될 문서의 특징을 나타내는 범주별 자질과 유사한의미를 가지나 형태가 변형되어 기술된 단어에 대하여 유사어 사전을 구축하였으며 이를 통해 새로운 문서가 범주에 할당될 가능성을 높여 자동 문서 범주화 시스템의 성능을 향상시키고자 한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10a
• /
• pp.33-35
• /
• 1999

인터넷의 발달과 더불어 매일같이 제공되는 수많은 정보로부터 자신에게 필요한 정보만을 추출하는데는 많은 시간과 노력이 소모된다. 이러한 정보수집의 어려움에서 정보를 쉽고 효율적으로 찾기 위해서 웹 문서 시각화 시스템을 구현하였다. 웹 문서 시각화 시스템은 사용자가 검색하는 정보는 과거에 검색했던 웹 문서를 다시 방문하는 경험에서 착안하였다. 이를 위해 인터넷 익스플로러를 통해서 방문 중인 웹 문서의 URL, 키워드, 문서간의 유사성을 추출하여 시각화 한다. 시각화 알고리즘으로 FastMap을 사용하였다. 본 논문에서 FastMap은 웹문서간의 유사성, 즉 상대적인 거리 객체 형태를 2-차원 공간으로 표현하는 알고리즘이다. 2차원 공간으로 매핑된 주변에 있는 웹 문서 객체들을 확대하면 방문중인 웹 문서와 유사성이 있는 문서를 쉽게 찾을 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.1234-1237
• /
• 2011

본 논문은 시맨틱 기술을 활용한 OpenAPI 조합가능 서비스 검색 기법 개발을 위해 관련 연구들을 분석하여 적용방안을 도출하였다. OpenAPI 조합가능 서비스 검색이란 선택된 서비스의 출력과 매칭이 되는 입력을 가진 서비스나 선택된 서비스의 입력과 매칭되는 출력을 가진 서비스를 찾는 것이다. 본 논문에서는 기존의 키워드 기반의 조합 가능 서비스 검색 기법의 한계를 시맨틱 기술을 활용하여 해결하기 위한 관련연구로 SAWSDL-MX2와 LOG4SWS.KOM에 관한 논문들을 분석했다. SAWSDL-MX2에서는 세가지 매칭 기법과 이에 따른 유사도 분석 기법을 제시하였고, LOG4SWS.KOM에서는 두가지 매칭 기법과 이에 따른 유사도 분석 방법을 제시하였다. 관련 연구들에서 분석된 내용을 바탕으로 OpenAPI 조합 가능 서비스에 대한 매칭 기법의 정의, 유사도 분석 기법의 정의가 추후 진행되어야 한다.

• Korean Journal of Food Science and Technology
• /
• v.51 no.2
• /
• pp.176-181
• /
• 2019

The aim of this study was to analyze dieting practices and tendencies in 2016 using big data. The keywords related to diet were collected from the portal site Naver and analyzed through simple frequency, N-gram, keyword network, and analysis of seasonality. The results showed that exercise had the highest frequency in simple frequency analysis. However, diet menu appeared most frequently in N-gram analysis. In addition, analysis of seasonality showed that the interest of subjects in diet increased steadily from February to July and peaked in October 2016. The monthly frequency of the keyword highfat diet was highest in October, because that showed the 'Low Carbohydrate High Fat' TV program. Although diet showed a certain pattern on a yearly basis, the emergence of new trendy diets in mass media also affects the pattern of diet. Therefore, it is considered that continuous monitoring and analysis of diet is needed rather than periodic monitoring.

• Management & Information Systems Review
• /
• v.36 no.1
• /
• pp.273-290
• /
• 2017

Internet of Things (IoT) is considered to be the next wave of Information Technology transformation after the Internet has changed the process of doing business. Since the domain of IoT ranging from the sensor technology to service to the users is wide, the structure of the research domain is not delineated clearly. To do that we suggest to use the Technology Stack Model proposed by Porter et al.(2014) to measure the maturity level of IoT in organizations. Based on the Stack Model, for the general understandings of IoT, we do keyword analyses on the academic papers whose major research issue is IoT. It is found that the current status of IoT application from the perspectives of cloud and big data analytics is not active, meaning that the real value of IoT has not been realized. We also examine the cases which deal with the part of cloud process which is crucial for value accrual. Based on these findings, we suggest the future direction of IoT research. We also propose that IT is to value chain what IoT is to the Stack Model to derive value in organizations.

• Journal of Environmental Impact Assessment
• /
• v.26 no.5
• /
• pp.303-314
• /
• 2017

The purpose of this study was to determine domestic and foreign ecological asset research trends. We aimed to understand ecological assets research directions and trends by comprehensively analyzing 12 keywords, including those similar to keywords for comparable assets, to identify related fields and regions. Extensive analysis of domestic and foreign studies was conducted through keyword network analysis of textural information. This approach is helpful for understanding the flow of information and identifying research directions. Foreign studies based on sustainability were connected with 'Economic assessment', 'Management' and 'Policy' areas. It was difficult to determine domestic research trends because there are fewer domestic studies than foreign. There were studies that sought to identify economic value of developing regions. This research can be used to guide the research direction for future ecosystem asset analysis in Korea.

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.209-220
• /
• 2017

As a semantic data is widely used in various applications such as Knowledge Bases and Semantic Web, needs for effective search over a large amount of RDF data have been increasing. Previous keyword search methods based on distinct root semantics only retrieve a set of answer trees having different root nodes. Thus, they often find answer trees with similar meanings or low query relevance together while those with the same root node cannot be retrieved together even if they have different meanings and high query relevance. We propose a new method to find diverse and relevant answers to the query by permitting duplication of root nodes among them. We present an efficient query processing algorithm using path indexes to find top-k answers given a maximum amount of root duplication a set of answer trees can have. We show by experiments using a real dataset that the proposed approach can produce effective answer trees which are less redundant in their content nodes and more relevant to the query than the previous method.

• Journal of KIISE:Databases
• /
• v.33 no.5
• /
• pp.538-550
• /
• 2006

XML Keyword search enables us to get information easily without knowledge of structure of documents and returns specific and useful partial document results instead of whole documents. Element level query processing makes it possible, but computational complexity, as the number of documents grows, increases significantly overhead costs. In this paper, we present document-level ranking scheme over XML documents which predicts results of element-level processing to reduce processing cost. To do this, we propose the notion of 'keyword proximity' - the correlation of keywords in a document that affects the results of element-level query processing using path information of occurrence nodes and their resemblances - for document ranking process. In benefit of document-centric view, it is possible to reduce processing time using ranked document list or filtering of low scored documents. Our experimental evaluation shows that document-level processing technique using ranked document list is effective and improves performance by the early termination for top-k query.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.5
• /
• pp.550-555
• /
• 2018

This study aims to identify and compare contents and keywords used in articles related to blockchain applications to various industries. The text mining and Semantic Network Analysis, as methods of keyword network analysis, were used to analyze articles including terms of 'finance' 'energy' and 'logistics', which media and government frequently mentioned as areas that can apply blockchain technologies. For this study, data were collected from 43,093 articles from January, 2017 through July, 2018. Data crawling was carried out by using Python BeautifulSoup and data cleaning was performed in order to eliminate mutual redundancies of the three terms. After that, text mining and semantic network analysis were performed using Textom and UCInet for network analysis between keywords. The results showed that all the three terms were similar in terms of 'technology', but there were differences in the contents of 'government policy' or 'industry' issues. In addition, there were differences in frequencies and centralities of these terms.