Document Clustering based on Topic Maps

A major computational burden, while performing document clustering, is the calculation of similarity measure between a pair of documents. Similarity measure is a function that assigns a real number between 0 and 1 to a pair of documents, depending upon the degree of similarity between them. A value of zero means that the documents are completely dissimilar whereas a value of one indicates that the documents are practically identical. Traditionally, vector-based models have been used for computing the document similarity. The vector-based models represent several features present in documents. These approaches to similarity measures, in general, cannot account for the semantics of the document. Documents written in human languages contain contexts and the words used to describe these contexts are generally semantically related. Motivated by this fact, many researchers have proposed seman-tic-based similarity measures by utilizing text annotation through external thesauruses like Word...

2017

ing for each document. There are several possible extensions to this work:  The proposed document clustering approach has many practical applications. One direction is to apply this technique on some specific application area along with application specific optimizations to see the outcome. For example: web search results can be clustered using this approach. The snippets for each cluster are generated to see the quality of these snippets.  In the proposed approach each term, whether it is from lexical chain or from topic maps, has an equal effect on similarity calculation for a pair of documents. One possible direction is to introduce discriminative feature weighting for the features in this approach. Discriminative feature weighting has encouraging results for both text clustering and classification tasks.

In text categorization problem the most used method for documents representation is based on words frequency vectors called VSM (Vector Space Model). This representation is based only on words from documents and in this case loses any “word context” information found in the document. In this article we make a comparison between the classical method of document representation and a method called Suffix Tree Document Model (STDM) that is based on representing documents in the Suffix Tree format. For the STDM model we proposed a new approach for documents representation and a new formula for computing the similarity between two documents. Thus we propose to build the suffix tree only for any two documents at a time. This approach is faster, it has lower memory consumption and use entire document representation without using methods for disposing nodes. Also for this method is proposed a formula for computing the similarity between documents, which improves substantially the clustering quality. This representation method was validated using HAC - Hierarchical Agglomerative Clustering. In this context we experiment also the stemming influence in the document preprocessing step and highlight the difference between similarity or dissimilarity measures to find “closer” documents.

2014

Clustering is an automatic learning technique aimed at grouping a set of objects into subsets or clusters. Objects in the same cluster should be as similar as possible, whereas objects in one cluster should be as dissimilar as possible from objects in the other clusters. Document clustering has become an increasingly important task in analysing huge documents. The challenging aspect to analyse the enormous documents is to organise them in such a way that facilitates better search and knowledge extraction without introducing extra cost and complexity. Document clustering has played an important role in many fields like information retrieval and data mining. In this paper, first Document Clustering has been proposed using Hierarchical Agglomerative Clustering and K-Means Clustering Algorithm.Here, the approach is purely based on the frequency count of the terms present in the documents where context of the documents are totally ignored. Therefore, the method is modified by incorporati...

IEEE Transactions on Knowledge and Data Engineering, 2000

Phrase has been considered as a more informative feature term for improving the effectiveness of document clustering. In this paper, we propose a phrase-based document similarity to compute the pairwise similarities of documents based on the Suffix Tree Document (STD) model. By mapping each node in the suffix tree of STD model into a unique feature term in the Vector Space Document (VSD) model, the phrase-based document similarity naturally inherits the term tf-idf weighting scheme in computing the document similarity with phrases. We apply the phrase-based document similarity to the group-average Hierarchical Agglomerative Clustering (HAC) algorithm and develop a new document clustering approach. Our evaluation experiments indicate that the new clustering approach is very effective on clustering the documents of two standard document benchmark corpora OHSUMED and RCV1. The quality of the clustering results significantly surpasses the results of traditional single-word tf-idf similarity measure in the same HAC algorithm, especially in large document data sets. Furthermore, by studying the property of STD model, we conclude that the feature vector of phrase terms in the STD model can be considered as an expanded feature vector of the traditional single-word terms in the VSD model. This conclusion sufficiently explains why the phrase-based document similarity works much better than the single-word tf-idf similarity measure.

The most popular way for representing documents is the vector space model, because of its speed and versatility. The vector space model has some drawbacks. To overcome the bag of words problems, text documents are treated as a sequence of words and documents are retrieved based on sharing of frequent word sequences from text databases. The sequential relationship between the words and documents is preserved using a suffix tree data structure. Syntax based disambiguation is attempted by enriching the text document representations by background knowledge provided in a core ontology. Word Net is used for this purpose in our model. This work aims to extend a document representation model which is elegant by combining the versatility of the vector space model, the increased relevance of the suffix tree document model and also retains the relationship between words like synonyms. The effectiveness and the relevance of this concept based model compared to the existing models is evaluated by a partitioning clustering technique and then a systematic comparative study of the impact of similarity measures in conjunction with different types of vector space representation on cluster quality is performed. This document model will be called the Concept Based Vector Suffix Tree Document Model (CBVSTDM).

International Journal of Advanced Computer Science and Applications, 2016

Document clustering is an unsupervised machine learning method that separates a large subject heterogeneous collection (Corpus) into smaller, more manageable, subject homogeneous collections (clusters). Traditional method of document clustering works around extracting textual features like: terms, sequences, and phrases from documents. These features are independent of each other and do not cater meaning behind these word in the clustering process. In order to perform semantic viable clustering, we believe that the problem of document clustering has two main components: (1) to represent the document in such a form that it inherently captures semantics of the text. This may also help to reduce dimensionality of the document and (2) to define a similarity measure based on the lexical, syntactic and semantic features such that it assigns higher numerical values to document pairs which have higher syntactic and semantic relationship. In this paper, we propose a representation of document by extracting three different types of features from a given document. These are lexical α, syntactic β and semantic γ features. A meta-descriptor for each document is proposed using these three features: first lexical, then syntactic and in the last semantic. A document to document similarity matrix is produced where each entry of this matrix contains a three value vector for each lexical α, syntactic β and semantic γ. The main contributions from this research are (i) A document level descriptor using three different features for text like: lexical, syntactic and semantics. (ii) we propose a similarity function using these three, and (iii) we define a new candidate clustering algorithm using three component of similarity measure to guide the clustering process in a direction that produce more semantic rich clusters. We performed an extensive series of experiments on standard text mining data sets with external clustering evaluations like: F-Measure and Purity, and have obtained encouraging results.

Proceedings of the 16th international conference on World Wide Web - WWW '07, 2007

In this paper, we propose a new similarity measure to compute the pairwise similarity of text-based documents based on suffix tree document model. By applying the new suffix tree similarity measure in Group-average Agglomerative Hierarchical Clustering (GAHC) algorithm, we developed a new suffix tree document clustering algorithm (NSTC). Experimental results on two standard document clustering benchmark corpus OHSUMED and RCV1 indicate that the new clustering algorithm is a very effective document clustering algorithm. Comparing with the results of traditional word term weight tf-idf similarity measure in the same GAHC algorithm, NSTC achieved an improvement of 51% on the average of F-measure score. Furthermore, we apply the new clustering algorithm in analyzing the Web documents in online forum communities. A topic oriented clustering algorithm is developed to help people in assessing, classifying and searching the the Web documents in a large forum community.

2010

Document clustering is one of the most major techniques to group documents automatically. This technique is to divide a given set of documents into a certain number of clusters automatically. In this technique, the first step is ’feature extraction’ from documents. As a feature used in the conventional methods, we frequently use a set of words that contains nouns and verbs. Although words are used as features in a generic clustering framework, some previous research proposes the clustering method using the other features based on vector space model such as kernel methods and adaptive sprinkling. However, in previous research of document clustering, the method of appending new feature vectors obtained by using relationship between the existing documents and other documents has not been reported yet. So, we propose a new method for clustering documents using the relationship between the existing documents and other documents to acquire the more useful clusters for users. Our method ca...