• Journal of the Korean Society for information Management
• /
• v.19 no.3
• /
• pp.111-122
• /
• 2002

In order to investigate information seeking behavior of web search users, this study analyzes transaction logs posed by users of NAVER, a major Korean Internet search service. We present a session definition method for Web transaction log analysis, a way of cleaning original logs and a query classification method. We also propose a query term definition method that is necessary for Korean Web transaction log analysis. It is expected that this study could contribute to the development and implementation of more effective Web search systems and services.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.15-21
• /
• 2010

본 논문은 검색 서비스 기능 중에 빼놓을 수 없는 기능인 한글 검색 질의어(query) 교정 시스템을 '야후!'에서 구축하며 분석한 한글 오타 패턴 그리고 사용자 로그를 기반으로 설계한 질의어 교정 서비스에 대한 설명을 하고 있다. 이 교정 서비스는 현재 '야후! 코리아'에 적용되어 있으며, 한글을 고려한 키스트 로크를 기반으로 한 설계 방식 그리고 동적으로 에러모델을 구축하는 방법을 소개하고 있으며 또한 구축된 모델의 성능을 다른 검색 서비스와 비교한 결과를 소개한다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2002.11a
• /
• pp.266-272
• /
• 2002

Recently various of information suppliers provide information via WWW so the necessary of search engine grows larger. However the efficiency of most search engines is low comparatively because of using simple pattern match technique between user's query and web document. And a manifest contents of query for special expert field so much worse A specialized search engine returns the specialized information depend on each user's search goal. It is trend to develop specialized search engines in many countries. For example, in America, there are a site that searches only the recently updated headline news and the federal law and the government and and so on. However, most such engines don't satisfy the user's needs. This paper proposes the specialized search engine for ocean information that uses user's query related with ocean and search engine uses the association rules in web data mining. So specialized search engine for ocean provides more information related to ocean because of raising recall about user's query

• The KIPS Transactions:PartB
• /
• v.11B no.5
• /
• pp.537-544
• /
• 2004

The Web is rich with various sources of information. Due to the massive and heterogeneous web document collections, users want to find various types of target pages. Each type of information for Web search has designated queries. If a user query is not a designated query, then we cannot have good result documents. Different strategies are needed to utilize the goodness of each type of information for a search engine. If we know the property of information, then we can refine candidate pages and rank them delicately. Various experiments are conducted to show the properties of each type of information. Therefore, we show an appropriate combining formula to utilize the properties of each type of information. In addition, for a service finding task, we propose Service Link Information that utilizes the existence of mechanisms for a user interaction.

• Journal of the Korean Society for information Management
• /
• v.20 no.2
• /
• pp.27-41
• /
• 2003

Query logs are online records that capture user interactions with information retrieval systems and all the search processes. Query log analysis offers ad advantage of providing reasonable and unobtrusive means of collecting search information from a large number of users. In this paper, query logs of NAVER, a major Korean Internet search service, were analyzed to investigate the information seeking behabior of NAVER users. The query logs were collected over one week from various collecions such as comprehensive search, directory search and web ducument searc. It is expected that this study could contribute to the development and implementation of more effective web search systems and services.

• Journal of the Korean Society for Library and Information Science
• /
• v.39 no.2
• /
• pp.147-160
• /
• 2005

This study examines trends of web query types and topics submitted to NAVER during one year period by analyzing query logs and click logs. There was a seasonal difference in the distribution of query types. Query type distribution was also different between weekdays and weekends, and between different days of the week. The log data show seasonal changes in terms of the topics of queries. Search topics seem to change between weekdays and weekends, and between different days of the week. However, there was little change in overall patterns of search behavior across one year. The implications for system designers and web content providers are discussed.

• Journal of Information Technology Applications and Management
• /
• v.13 no.4
• /
• pp.121-140
• /
• 2006

UDDI (Universal Discovery Description and Integration) Registry is used for Web Services registration and search. UDDI offers the search result to the keyword-based query. UDDI supports WSDL registration but it does not supports WSDL search. So it is required that contents based search and ranking using name and description in UDDI registration information and WSDL. This paper proposes a retrieval engine considering contents of services registered in the UDDI and WSDL. It uses Vector Space Model for similarity comparison between contents of those. UDDI registry information hierarchy and WSDL hierarchy are considered during searching process. This engine suppports two discovery methods. One is Keyword-based search and the other is template-based search supporting ranking for user's query. Template-based search offers how service interfaces correspond to the query for WSDL documents. Proposed retrieval engine can offer search result more accurately than one which UDDI offers and it can retrieve WSDL which is registered in UDDI in detail.

• Journal of information and communication convergence engineering
• /
• v.11 no.4
• /
• pp.268-273
• /
• 2013

In this paper, we propose a document classification model using Web documents as a part of the training corpus in order to resolve the imbalance of the training corpus size per category. For the purpose of retrieving the Web documents closely related to each category, the proposed document classification model calculates the matching score between word features and each category, and generates a Web search query by combining the higher-ranked word features and the category title. Then, the proposed document classification model sends each combined query to the open application programming interface of the Web search engine, and receives the snippet results retrieved from the Web search engine. Finally, the proposed document classification model adds these snippet results as Web documents to the training corpus. Experimental results show that the method that considers the balance of the training corpus size per category exhibits better performance in some categories with small training sets.

• Journal of KIISE:Databases
• /
• v.36 no.4
• /
• pp.321-326
• /
• 2009

Since the existing Web search engines don't sufficiently reflect user's search intent, it is very difficult to find out accurate information that users want to find. Therefore, a lot of researches, study for personalized search, to enhance satisfaction of Web search results by analyzing search pattern and applying it to search are in progress in these days. Web searchers can more efficiently find information and easily obtain appropriate information through the personalized search. In this paper, we propose the personalized search based on community through the analysis of web users' query patterns and interest. Consequently, when applying query frequency, interest and community to web search, we are able to the confirm that the search results which hit to the search intent of the individual are provided.

• Journal of the Korean Society for Library and Information Science
• /
• v.45 no.2
• /
• pp.209-228
• /
• 2011

Web search behavior has evolved. People now search using many diverse information devices in various situations. To monitor these scattered and shifting search patterns, an improved way of learning and analysis are needed. Traditional web search studies relied on the server transaction logs and single query instance analysis. Since people use multiple smart devices and their searching occurs intermittently through a day, a bundled query research could look at the whole context as well as penetrating search needs. To observe and analyze bundled queries, we developed a proprietary research software set including a log catcher, query bundling tool, and bundle monitoring tool. In this system, users' daily search logs are sent to our analytic server, every night the users need to log on our bundling tool to package his/her queries, a built in web survey collects additional data, and our researcher performs deep interviews on a weekly basis. Out of 90 participants in the study, it was found that a normal user generates on average 4.75 query bundles a day, and each bundle contains 2.75 queries. Query bundles were categorized by; Query refinement vs. Topic refinement and 9 different sub-categories.