• The KIPS Transactions:PartB
• /
• v.14B no.7
• /
• pp.513-522
• /
• 2007

Clustering is to divide given data and automatically find out the hidden meanings in the data. It analyzes data, which are difficult for people to check in detail, and then, makes several clusters consisting of data with similar characteristics. On-Line Document Clustering System, which makes a group of similar documents by use of results of the search engine, is aimed to increase the convenience of information retrieval area. Document clustering is automatically done without human interference, and the number of clusters, which affect the result of clustering, should be decided automatically too. Also, the one of the characteristics of an on-line system is guarantying fast response time. This paper proposed a method of determining the number of clusters automatically by geometrical information. The proposed method composed of two stages. In the first stage, centers of clusters are projected on the low-dimensional plane, and in the second stage, clusters are combined by use of distance of centers of clusters in the low-dimensional plane. As a result of experimenting this method with real data, it was found that clustering performance became better and the response time is suitable to on-line circumstance.

• Journal of the Korean Society for information Management
• /
• v.21 no.4 s.54
• /
• pp.173-185
• /
• 2004

Fast and high-quality document clustering algorithms play an important role in providing data exploration by organizing large amounts of information into a small number of meaningful clusters. Many papers have shown that the hierarchical clustering method takes good-performance, but is limited because of its quadratic time complexity. In contrast, with a large number of variables, K-means has a time complexity that is linear in the number of documents, but is thought to produce inferior clusters. In this paper, Condor system using K-Means algorithm Compares with regular method that the initial centroids have been established in advance, our method performance has been improved a lot.

• Journal of Korean Library and Information Science Society
• /
• v.34 no.3
• /
• pp.211-229
• /
• 2003

In this paper, we have focused that the number of word in the web document affects definite clustering performance. Our experimental results have clearly shown the relationship between the amounts of word and its impact on clustering performance. We also have presented an algorithm that can be supplemented of the contrast portion through co-links frequency of web documents. Testing bench of this research is 1,449 web documents included on 'Natural science' category among the Naver Directory. We have clustered these objects by term-based clustering, link-based clustering, and hybrid clustering method, and compared the output results with originally allocated category of Naver directory.

• The Journal of the Korea Contents Association
• /
• v.9 no.3
• /
• pp.1-8
• /
• 2009

This paper proposes a new document clustering system using fuzzy logic-based genetic algorithm (GA) and semantic vector expansion technology. It has been known in many GA papers that the success depends on two factors, the diversity of the population and the capability to convergence. We use the fuzzy logic-based operators to adaptively adjust the influence between these two factors. In traditional document clustering, the most popular and straightforward approach to represent the document is vector space model (VSM). However, this approach not only leads to a high dimensional feature space, but also ignores the semantic relationships between some important words, which would affect the accuracy of clustering. In this paper we use latent semantic analysis (LSA)to expand the documents to corresponding semantic vectors conceptually, rather than the individual terms. Meanwhile, the sizes of the vectors can be reduced drastically. We test our clustering algorithm on 20 news groups and Reuter collection data sets. The results show that our method outperforms the conventional GA in various document representation environments.

• Journal of Intelligence and Information Systems
• /
• v.17 no.1
• /
• pp.111-125
• /
• 2011

Clustering is a process of grouping similar or relevant documents into a cluster and assigning a meaningful concept to the cluster. By this process, clustering facilitates fast and correct search for the relevant documents by narrowing down the range of searching only to the collection of documents belonging to related clusters. For effective clustering, techniques are required for identifying similar documents and grouping them into a cluster, and discovering a concept that is most relevant to the cluster. One of the problems often appearing in this context is the detection of a complex concept that overlaps with several simple concepts at the same hierarchical level. Previous clustering methods were unable to identify and represent a complex concept that belongs to several different clusters at the same level in the concept hierarchy, and also could not validate the semantic hierarchical relationship between a complex concept and each of simple concepts. In order to solve these problems, this paper proposes a new clustering method that identifies and represents complex concepts efficiently. We developed the Hierarchical Overlapping Clustering (HOC) algorithm that modified the traditional Agglomerative Hierarchical Clustering algorithm to allow overlapped clusters at the same level in the concept hierarchy. The HOC algorithm represents the clustering result not by a tree but by a lattice to detect complex concepts. We developed a system that employs the HOC algorithm to carry out the goal of complex concept detection. This system operates in three phases; 1) the preprocessing of documents, 2) the clustering using the HOC algorithm, and 3) the validation of semantic hierarchical relationships among the concepts in the lattice obtained as a result of clustering. The preprocessing phase represents the documents as x-y coordinate values in a 2-dimensional space by considering the weights of terms appearing in the documents. First, it goes through some refinement process by applying stopwords removal and stemming to extract index terms. Then, each index term is assigned a TF-IDF weight value and the x-y coordinate value for each document is determined by combining the TF-IDF values of the terms in it. The clustering phase uses the HOC algorithm in which the similarity between the documents is calculated by applying the Euclidean distance method. Initially, a cluster is generated for each document by grouping those documents that are closest to it. Then, the distance between any two clusters is measured, grouping the closest clusters as a new cluster. This process is repeated until the root cluster is generated. In the validation phase, the feature selection method is applied to validate the appropriateness of the cluster concepts built by the HOC algorithm to see if they have meaningful hierarchical relationships. Feature selection is a method of extracting key features from a document by identifying and assigning weight values to important and representative terms in the document. In order to correctly select key features, a method is needed to determine how each term contributes to the class of the document. Among several methods achieving this goal, this paper adopted the $x^2$�� statistics, which measures the dependency degree of a term t to a class c, and represents the relationship between t and c by a numerical value. To demonstrate the effectiveness of the HOC algorithm, a series of performance evaluation is carried out by using a well-known Reuter-21578 news collection. The result of performance evaluation showed that the HOC algorithm greatly contributes to detecting and producing complex concepts by generating the concept hierarchy in a lattice structure.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.2
• /
• pp.150-156
• /
• 2008

Many papers have shown that the hierarchical clustering method takes good-performance, but is limited because of its quadratic time complexity. In contrast, with a large number of variables, K-means reduces a time complexity. Think of the factor of simplify, high-quality and high-efficiency, we combine the two approaches providing a new system named CONDOR system with hierarchical structure based on document clustering using K-means algorithm. Evaluated the performance on different hierarchy depth and initial uncertain centroid number based on variational relative document amount correspond to given queries. Comparing with regular method that the initial centroids have been established in advance, our method performance has been improved a lot.

• Journal of Internet Computing and Services
• /
• v.10 no.3
• /
• pp.141-150
• /
• 2009

XML is increasingly important in data exchange and information management. A large amount of efforts have been spent in developing efficient techniques for accessing, querying, and storing XML documents. In this paper, we propose a new method to cluster XML documents efficiently. A new prepresentative path called a virtul path which can represent both the structure and the contents of a XML document is proposed for the feature of a XML document. A method to apply the well known hierarchical clustering techniques to the representative paths to cluster XML documents is also proposed. The experiment shows that the true clusters are formed in a compact shape when a virtual path is used for the feature of a XML document.

• Journal of KIISE:Software and Applications
• /
• v.33 no.2
• /
• pp.243-251
• /
• 2006

In multi-document summarization by means of salient sentence extraction, it is important to remove redundant information. In the removal process, the similarities and differences of sentences are considered. In this paper, we propose a method for multi-document summarization which extracts salient sentences without having redundant sentences by way of cohesive term clustering method that utilizes co-occurrence Information. In the cohesive term clustering method, we assume that each term does not exist independently, but rather it is related to each other in meanings. To find the relations between terms, we cluster sentences according to topics and use the co-occurrence information oi terms in the same topic. We conduct experimental tests with the DUC(Document Understanding Conferences) data. In the tests, our method shows better performance of summarization than other summarization methods which use term co-occurrence information based on term cohesion of document or sentence unit, and simple statistical information.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.12
• /
• pp.2607-2613
• /
• 2012

The results of inherent knowledge based generic summarization are influenced by the composition of sentence in document set. In order to resolve the problem, this papser propses a new generic document summarization which uses clustering of semantic feature of document and coherence of document cluster. The proposed method clusters sentences using semantic feature deriving from NMF(non-negative matrix factorization), which it can classify document topic group because inherent structure of document are well represented by the sentence cluster. In addition, the method can improve the quality of summarization because the importance sentences are extracted by using coherence of sentence cluster and the cluster refinement by re-cluster. The experimental results demonstrate appling the proposed method to generic summarization achieves better performance than generic document summarization methods.

• Journal of KIISE:Software and Applications
• /
• v.36 no.5
• /
• pp.347-352
• /
• 2009

Nonnegative matrix factorization (NMF) is a popular method for multivariate analysis of nonnegative data, the goal of which is decompose a data matrix into a product of two factor matrices with all entries in factor matrices restricted to be nonnegative. NMF was shown to be useful in a task of clustering (especially document clustering). In this paper we present an algorithm for orthogonal nonnegative matrix factorization, where an orthogonality constraint is imposed on the nonnegative decomposition of a term-document matrix. We develop multiplicative updates directly from true gradient on Stiefel manifold, whereas existing algorithms consider additive orthogonality constraints. Experiments on several different document data sets show our orthogonal NMF algorithms perform better in a task of clustering, compared to the standard NMF and an existing orthogonal NMF.