• 한국경영정보학회:학술대회논문집
• /
• 2007.11a
• /
• pp.623-628
• /
• 2007

본 논문은 Web 2.0시대의 새로운 컨텐츠 매체로 각광받고 있는 블로그와 관련하여 태그 기반의 검색 알고리즘을 제안하고자 한다. 최근 블로그 검색과 관련하여 태그 기반의 블로그 검색 서비스가 등장하기 시작했지만, 현재 제공되는 태그 기반의 검색 서비스는 태그의 유무와 컨텐트의 최신성을 주요 기준으로 삼고, 태그와 컨텐트 간의 관련성을 제대로 고려하지 않아 검색 결과가 만존스럽지 못하는 경우가 많다. 따라서 본 논문에서는 태그와 컨텐트와의 관련성을 실수화하고 이를 주요 기준으로 검색 결과의 순위를 결정하는 PTRank 알고리즘을 제안하였다. PTRank 알고리즘에서는 1) 태그가 피드의 제목에 포함되었는지 여부, 2) 태그가 피드의 설명에 나타나는 회수, 3) 태그가 아이템의 제목에 포함되었는지 여부, 4) 태그가 아이템의 설명에 나타나는 횟수, 5) 피드 내에서 태그의 IDF값, 6) 사용자의 검색 행위를 이용해 태그와 컨텐트간의 관련성을 실수화하였다. 실험 결과, PTRank 모델 및 학습 알고리즘이 태그 기반의 피드 검색에서 잘 작동하며 검색에 효과적으로 활용될 수 있다는 것을 알 수 있었다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.11
• /
• pp.1121-1125
• /
• 2010

This paper proposes a Wikipedia-based feedback method for in-depth blog distillation whose goal is to find blogs that represent in-depth thoughts or analysis on a given query. The proposed method uses Wikipedia articles which are relevant to the query. TREC Blogs08 collection which is a large-scale blog corpus and English Wikipedia dump were used for experiments, The proposed method significantly increased the retrieval performance including MAP over the conventional post based feedback method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.11a
• /
• pp.48-49
• /
• 2010

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06a
• /
• pp.92-93
• /
• 2010

• Journal of KIISE:Databases
• /
• v.37 no.5
• /
• pp.258-266
• /
• 2010

In this paper, we introduce a novel method to predict next update of blogs. The number of RSS feeds registered on meta-blogs is on the order of several million. Checking for updates is very time consuming and imposes a heavy burden on network resources. Since blog search engine has limited resources, there is a fix number of blogs that it can visit on a day. Nevertheless we need to maximize chances of getting new data, and the proposed method which predicts update probability on blogs could bring better chances for it. Also this work is important to avoid distributed denial-of-service attack for the owners of blogs. Furthermore, for the internet as whole this work is important, too, because our approach could minimize traffic. In this study, we assumed that there is a specific pattern to when a blogger is actively posting, in terms of days of the week and, more specifically, hours of the day. We analyzed 15,119 blogs to determine a blogger's posting preference. This paper proposes a method to predict the update probability based on a blogger's posting history and preferred days of the week. We applied proposed method to 12,115 blogs to check the precision of our predictions. The evaluation shows that the model has a precision of 0.5 for over 93.06% of the blogs examined.

• The KIPS Transactions:PartD
• /
• v.16D no.6
• /
• pp.845-850
• /
• 2009

Use of XML data has been increased with growth of Web 2.0 environment. XML is recognized its advantages by using based technology of RSS or ATOM for transferring information from blogs and news feed. Bitmap clustering is a method to keep index in main memory based on Relational DBMS, and which performed better than the other XML indexing methods during the evaluation. Existing method generates too many clusters, and it causes deterioration of result of searching quality. This paper proposes k-Bitmap clustering method that can generate user defined k clusters to solve above-mentioned problem. The proposed method also keeps additional inverted index for searching excluded terms from representative bits of k-Bitmap. We performed evaluation and the result shows that the users can control the number of clusters. Also our method has high recall value in single term search, and it guarantees the searching result includes all related documents for its query with keeping two indices.