• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.294-296
• /
• 2012

본 논문은 일본어 웹 문서 말뭉치로부터 의존 구문 구조 관점으로 바라본 단어들의 동시발생(co-occurrence) 정보를 사용하여 서브토픽 마이닝(subtopic mining)을 수행하는 방법론을 제안한다. 우리는 의존 구문 구조를 반영하는 간단한 패턴들을 사용하여 서브토픽들을 추출 및 생성하고, 제안한 수식을 바탕으로 순위화한다. 본 방법론은 기존의 주요 상용 검색 서비스에서 제공하는 연관 검색어 및 추천 검색어를 사용한 방법론보다 좋은 성능을 보였다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.265-274
• /
• 2021

In this paper, we propose a novel recommendation model based on Doc2vec using search keywords and product details. Until now, a lot of prior studies on recommender systems have proposed collaborative filtering (CF) as the main algorithm for recommendation, which uses only structured input data such as customers' purchase history or ratings. However, the use of unstructured data like online customer review in CF may lead to better recommendation. Under this background, we propose to use search keyword data and product detail information, which are seldom used in previous studies, for product recommendation. The proposed model makes recommendation by using CF which simultaneously considers ratings, search keywords and detailed information of the products purchased by customers. To extract quantitative patterns from these unstructured data, Doc2vec is applied. As a result of the experiment, the proposed model was found to outperform the conventional recommendation model. In addition, it was confirmed that search keywords and product details had a significant effect on recommendation. This study has academic significance in that it tries to apply the customers' online behavior information to the recommendation system and that it mitigates the cold start problem, which is one of the critical limitations of CF.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.267-273
• /
• 2020

검색 엔진에 입력되는 질의 중 입력 빈도는 낮지만 상대적으로 길이가 긴 질의를 롱테일 질의라고 일컫는다. 롱테일 질의가 전체 검색 로그에서 차지하는 비중은 높은 반면, 그 형태가 매우 다양하고 검색 의도가 상세하며 개별 질의의 양은 충분하지 않은 경우가 많기 때문에 해당 질의에 대한 적절한 검색어를 추천하는 것은 어려운 문제다. 본 논문에서는 롱테일 질의 입력 시 적절한 검색어 추천을 제공하기 위하여 질의-문서 클릭 정보를 활용한 추출기반 모델 및 Seq2seq와 GPT-2 기반 생성모델을 활용한 질의 확장 방법론을 제안한다. 실험 및 결과 분석을 통하여 제안 방법이 기존에 대응하지 못했던 롱테일 질의를 자연스럽게 확장할 수 있음을 보였다. 본 연구 결과를 실제 서비스에 접목함으로써 사용자의 검색 편리성을 증대하는 동시에, 언어 모델링 기반 질의 확장에 대한 가능성을 확인하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.3
• /
• pp.1795-1805
• /
• 2015

This paper analyzed the recommendation stories in the online journalism on the user's news choice by college students in two ways. One way is recommendation stories, and the other one is their arrangement and the index of use. From the results of the analysis, 7 out of 11 types of recommendation stories had positive effects on selecting news stories, while the 4 other types had little effect. Most of the recommendation stories that had little effect on the user's news selection were 'comments' or 'things' related to tweets' on SNS. The arrangements of new stories and the searched keywords had some effects on the user's news choice but had no effect on the index of use. In addition, the hours of using news stories and the types of recommendation stories were mostly correlated with each other. Consequently, formal factors, such as the arrangement of news stories and the recommendation stories of online journalism, had positive effects on the user's news selection, as well as headlines and keywords of news stories.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.570-574
• /
• 2006

KISTI portal service is currently presenting the documents with many terminologies, so users can't find the results having their intention by using an umbrella query. In this paper, we suggest user oriented retrieval service that reflects query auto-complete, related-word suggestion and query expansion that uses nouns and relationships of U-WIN which is known as a semantic language resource. We intend to advance the retrieval satisfaction of current science & technology information service by using U-WIN's semantic information and improve the service environment that user can retrieve what they want quickly and exactly.

• Annual Conference of KIPS
• /
• 2005.11a
• /
• pp.457-460
• /
• 2005

본 논문에서는 추론규칙을 통한 온톨로지 구축과 이를 이용한 지능형 여행 추천 시스템 적용 방법에 대해 제안하고자 한다. 사용자 프로파일과 질의어 분석을 통한 사용자 성향 분석은 메타데이터 파일로 작성되며, 지능형 여행 추천을 위한 여행 온톨로지 및 Description Logic을 기반으로 생성된 추론 규칙은 정보 저장소에 저장한다. 온톨로지를 이용한 정보 검색은 다양하고 복잡한 조건에서 검색이 가능하였고, 인스턴스 추가시 각 클래스의 재생성 과정 없이 규칙의 설정만으로 쉽게 인스턴스를 추가할 수 있었다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.2
• /
• pp.65-72
• /
• 2012

In the 21st century of information systems, the amount of information resources are ever increasing and the role of information searching system is becoming criticalto easily acquire required information from the web. Generally, it requires the user to have enough pre-knowledge and superior capabilities to identify keywords of information to effectively search the web. However, most of the users undertake searching of the information without holding enough pre-knowledge and spend a lot of time associating key words which are related to their required information. Furthermore, many search engines support the keywords searching system but this only provides collection of similar words, and do not provide the user with exact relational search information with the keywords. Therefore this research report proposes a method of offering expanded user relationship search keywords by analyzing user query patterns to provide the user a system, which conveniently support their searching of the information.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.2
• /
• pp.29-35
• /
• 2023

This paper makes it possible to search videos based on sentence by improving the previous research which automatically generates rich metadata from videos and searches videos by key words. For search by sentence, morphemes are analyzed for each sentence, keywords are extracted, weights are assigned to each keyword, and some videos are recommended by applying a ranking algorithm developed in the previous research. In order to evaluate performance of video search in this paper, a sufficient amount of videos and sufficient number of user experiences are re required. However, in the current situation where these are insufficient, three indirect evaluation methods were used: evaluation of overall user satisfaction, comparison of recommendation scores and user satisfaction, and evaluation of user satisfaction by video categories. As a result of performance evaluation, it was shown that the rich metadata construction and video recommendation implementation in this paper give users high search satisfaction.

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.297-299
• /
• 2011

자연어 처리 분야 중 현재 가장 활용도가 높은 분야는 질의어 추천기능, 단어 자동 완성 기능 등으로 정보검색에서 사용자가 입력한 문자들을 바탕으로 질의어를 완성해주는 것이다. 이러한 기능을 위해서는 문서 내용을 고려한 N-Gram 데이터 구축이 필수적이다. 본 논문에서는 문서 편집기나 검색엔진의 질의어 추천 등에 많이 활용되는 N-Gram 데이터의 전문용어별 구축을 위해 위키피디아 문서를 이용하는 방안을 제시하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.1
• /
• pp.103-109
• /
• 2009

We build the web document recommending agent system which offers a certain amount of web documents to each user by monitoring and learning the user's action of web browsing. We also propose a method of query expansion using the Korean thesaurus. The queries to search for new web documents generate a candidate set using the Korean thesaurus. We extract the words which are mostly correlated with the queries, among the words in the candidate set, by using TF-IDF and mutual information. Then, we expand the query. If we adopt the system of query expansion, we can recommend a lot of web documents which have potential interests to users. We thus conclude that the system of query expansion is more effective than a base system of recommending web-documents to users.