Search

Showing total 2 results
Start Over Topic data mining Topic information retrieval Language japanese
2 results

1. Techniques of Document Clustering: A Review.

Author

Kishida, Kazuaki

Subjects
*INFORMATION architecture, *ALGORITHMS, *INFORMATION retrieval, *INFORMATION resources management, *ELECTRONIC records, *DATA mining
Abstract

The document clustering technique is widely recognized as a useful tool for information retrieval, organizing web documents, text mining and so on. The purpose of this paper is to review various document clustering techniques, and to discuss research issues for enhancing effectiveness or efficiency of the clustering methods. We explore extensive literature on non-hierarchical methods (single-pass methods), hierarchical methods (single-link, complete-link, etc.), dimensional reduction methods (LSI, principal component analysis, etc.), probabilistic methods, data mining techniques, and so on. In particular, this paper focuses on typical techniques, such as the k-means algorithm, the leader-follower algorithm, self-organizing map (SOM). single- or complete-link methods, bisecting k-means methods, latent semantic indexing (LSI), Gaussian-Mixture model and so on. After reviewing the techniques and algorithms, we discuss research issues on document clustering: computational complexity, feature extraction (selection of words), methods for defining term weights and similarity, and evaluation of results. [ABSTRACT FROM AUTHOR]

Published
2003

2. 医療文献からの類似項目発見についての考察

Subjects
文献検索, Literature based mining, Information retrieval, データマイニング, 情報検索, 文献からの知識発見, Data mining, Document search
Abstract

近年，医療分野において研究成果を記述した学術論文が蓄積されている．それに伴い，学術文献内に出現する項目の関係を用いて，新たな項目間の発見を目指す文献に基づく知識発見(Literature based mining)が行われている．ここでいう項目とは病名，薬名，化学物質名，遺伝子名などを指す． 我々も，文献に基づく知識発見について研究しており，特に病名・遺伝子と薬名・化学物質名との関係に着目している．新たな類似項目が発見された時には，分子シミュレーションを行い，類似関係を実証する．我々の調査対象は米国国立医学図書館が作成した医学，生命科学分野に特化した文献データベース MEDLINEである．本論文では，研究の全体像を述べると共に現在我々が保持しているデータ量について述べる．また，いくつかの検討している分析方法やアルゴリズムについて考察する．, Recently, the scientific papers that describe the result of research in the medical field have been accumulated. With increasing these papers, Literature based mining have been tried to discover new knowledge by using the relationship of items in the source literature. The item here indicates the name of a disease, the medicine, the chemical, and the gene, etc. We also research the knowledge discovery based on the literature. We especially pay attention to the relation between disease and gene, and between medicine and chemical. When we discover new relationship, we confirm it using molecular dynamics computer simulation. Our target database is the MEDLINE which is the literature database specializes in the medical and biological science, and is made by the United States National Library of Medicine. In this paper, we describe over view of our research plan and the amount of data, and consider our method of analysis.

Published
2008