• Journal of KIISE
• /
• v.44 no.2
• /
• pp.179-185
• /
• 2017

Document summarization aims to generate a summary that is consistent and contains the highly related sentences in a document. In this study, we implemented for document summarization that extracts highly related sentences from a whole document by considering both similarities and entailment relations between sentences. Accordingly, we proposed a new algorithm, TextRank-NLI, which combines a Recurrent Neural Network based Natural Language Inference model and a Graph-based ranking algorithm used in single document extraction-based summarization task. In order to evaluate the performance of the new algorithm, we conducted experiments using the same datasets as used in TextRank algorithm. The results indicated that TextRank-NLI showed 2.3% improvement in performance, as compared to TextRank.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.443-450
• /
• 2005

In this paper, we propose an intelligent model lot extraction of the web user's preference and present the results of evaluation. For this purpose, we analyze shortcomings of current information retrieval engine being used and reflect preference weights on learner. As it doesn't depend on frequency of each word but intelligently learns patterns of user behavior, the mechanism Provides the appropriate set of results about user's questions. Then, we propose the concept of preference trend and its considerations and present an algorithm for extracting preference with examples. Also, we design an intelligent model for extraction of behavior patterns and propose HTML index and process of intelligent learning for preference decision. Finally, we validate the proposed model by comparing estimated results(after applying the Preference) of document ranking measurement.

• Journal of KIISE:Databases
• /
• v.37 no.6
• /
• pp.343-350
• /
• 2010

We can get answers we want to know via questioning in Knowledge Search Service (KSS) based on Q&A Community. However, it is getting more difficult to find credible documents in enormous documents, since many anonymous users regardless of credibility are participate in answering on the question. In previous works in KSS, researchers evaluated the quality of documents based on textual information, e.g. recommendation count, click count and non-textual information, e.g. answer length, attached data, conjunction count. Then, the evaluation results are used for enhancing search performance. However, the non-textual information has a problem that it is difficult to get enough information by users in the early stage of Q&A. The textual information also has a limitation for evaluating quality because of judgement by partial factors such as answer length, conjunction counts. In this paper, we propose the QualityRank algorithm to improve the problem by textual and non-textual information. This algorithm ranks the relevant and credible answers by considering textual/non-textual information and user centrality based on Social Network Analysis(SNA). Based on experimental validation we can confirm that the results by our algorithm is improved than those of textual/non-textual in terms of ranking performance.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.160-163
• /
• 2017

개인적인 활동들 전반에 걸쳐 스마트폰이 PC를 대체하고 있는 것으로 조사되었다. 스마트폰 사용자의 증가와 함께 다양한 어플리케이션이 개발되고 있고, 알람 어플리케이션은 혼자서 생활하는 시간이 많은 현대인들에게는 꼭 필요한 어플리케이션 가운데 하나이다. 이에 본 논문에서는 현대인들의 생활 패턴을 고려하여 기본적인 알람기능에 빅데이터를 이용한 알람음악 랭킹 제공, 날씨와 교통정보 제공, 일정관리 기능 등을 추가한 '빅데이터를 이용한 스마트 알람 시스템'을 개발하였다. 본 논문에서 개발한 어플리케이션은 바쁜 현대인의 아침시간에 여러 개의 어플리케이션을 사용할 필요가 없도록 사용자 편리성을 높인 알람 어플리케이션으로 개발하였다.

• The Journal of Society for e-Business Studies
• /
• v.14 no.4
• /
• pp.167-179
• /
• 2009

This paper introduces techniques to recommend standardized tags using tag ontology. Tag recommendation consists of TWCIDF and TWCITC; the former technique automatically tags a large quantity of already existing document groups, and the latter recommends tagging for new documents. Tag groups are created through several processes, including preprocessing, standardization using tag ontology, automatic tagging and defining ranks for recommendation. In the preprocessing process, in order to search semantic compound nouns, words are combined to establish basic word groups. In the standardization process, typographical errors and similar words are processed. As a result of experiments conducted on the basis of techniques presented in this paper, it is proved that real-time automatic tagging and tag recommendation is possible while guaranteeing the accuracy of tag recommendation.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.4
• /
• pp.1-9
• /
• 2010

In the new generation of Web, commonly called "Web 2.0", blogging has facilitated the publishing information or his/her opinion on the web. Various blog retrieval algorithms have been proposed to search for blogs more effectively. However, actually keyword-based searching or link-analysis blog ranking system cannot satisfy the user's requirement. In this paper, we suggest a topic-based weighting blog retrieval system in which the links between blog writings and searching words are considered to improve the search results. Our system extracts topics from each blog and weights them much higher than other guide words. In the comparison with other systems, we see that the proposed topic-base system has better recall rate of search results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.154-156
• /
• 2003

웹 상의 데이터 표현 및 교환의 표준으로 각광받는 XML은 논리적 구조와 내용 정보를 이용하여 보다 정확한 검색이 가능하다. 그러나 더욱 빠른 속도로 증가하는 많은 양의 데이터에 대해 보다 정확하고 풍부한 검색을 하기 위해서 메타데이터를 활용하는 방법이 고려되었고, RDF와 같은 메타데이터 기술 언어들에 대한 연구가 많이 이루어지고 있다. RDF는 XML의 문법 구조를 이용하여 작성되므로 XML 문서를 RDF 형태로 작성한다던가, 약간의 수정을 통해 기존 XML 문서를 RDF 형태로 변환하는 것이 가능하다. XML의 RDF 변환은 여러 이점 때문에 앞으로 활성화될 가능성이 크기 때문에 RDF의 특성을 고려한 저장 및 검색에 대한 연구가 필요하다. 따라서 본 논문에서는 XML을 기본적인 RDF 형태로 변환하는 기본적인 규칙을 소개하고 변환된 RDF 문서를 위한 저장 구조를 제안한다. 제안한 저장 구조는 기존 웹 애플리케이션과의 쉬운 연동을 위하여 관계형 데이터베이스를 기반으로 구성되며, 리소스/속성/값의 RDF 기본 구조를 고려한 세 종류의 테이블로 구성된다. 또한 본 논문에서는 RDF 문서에 대한 키워드 질의 처리를 고려하여, 질의 처리 결과의 단위를 리소스로 정의한다. 그리고 주어진 키워드들에 대한 질의 처리 결과로 반환된 리소스들 간의 중요도를 평가하기 위하여 키워드간의 근접도, 키워드 내포 정도, 다양한 속성 관계를 맺고 있는 다른 리소스들을 고려한 랭킹 평가 기법을 제안한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.3
• /
• pp.1422-1427
• /
• 2013

This paper proposed a movie retrieval system based on sentimental keywords extracted from user's movie reviews. At first, sentimental keyword dictionary is manually constructed by applying morphological analysis to user's movie reviews, and then keyword weights in the dictionary are calculated for each movie with TF-IDF. By using these results, the proposed system classify sentimental categories of movies and rank classified movies. Without reading any movie reviews, users can retrieve movies through queries composed by sentimental keywords.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.392-394
• /
• 2021

This paper is a thesis on an automatic restaurant recommendation application for tourists traveling to travel destinations. when you run the application at any travel destination in KOREA, it is an application that recommends desired services such as Korean, Chinese, Western, etc, regardless of the type of food, so that restaurant rankings are poured out in tourist destinations. not only recommending restaurants, but also collecting related information DB so that you can easily find restaurants in tourist destinations through reviews and stars such as hygiene conditions, prices, and compliance with quarantine regulations due to the recent coronavirus. the application was developed

• Journal of Digital Convergence
• /
• v.11 no.2
• /
• pp.271-277
• /
• 2013

Typical semantic search query for Semantic search promises to provide more accurate result than present-day keyword matching-based search by using the knowledge base represented logically. Existing keyword-based retrieval system is Preference for the semantic interpretation of a user's query is not the meaning of the user keywords of interconnect, you can not search. In this paper, we propose a method that can provide accurate results to meet the user's search intent to user preference based evaluation by ranking search. The proposed scheme is Integrated ontology-based knowledge base built on the formal structure of the semantic interpretation process based on ontology knowledge base system.