Search

Your search keyword '"Murty, M. Narasimha"' showing total 4 results
Start Over Author "Murty, M. Narasimha" Topic artificial intelligence
4 results on '"Murty, M. Narasimha"'

2. Topic based semantic clustering using Wikipedia knowledge.

Author

Ramudu, Bharathi and Murty, M Narasimha

Abstract

Most of the existing academic social networks like DBLP and ArnetMiner suggest co-authors as being related to each author based on author and co-author details, which are extracted from academic researchers publications. Instead of this approach, we propose grouping of researchers on basis of topics they have in common that would make more semantic sense for a user searching for publications in an area. Identification of relevant complete topic phrases, discovery of equivalent topics (synonyms) using terms in document collections are fundamental problems in information retrieval, natural language processing, pattern recognition, etc. In this paper, we propose a novel approach, wherein we identify a document's topic using Wikipedia. Wikipedia is the largest freely available crowd-source with up-to-date knowledge containing around 20 millions topics. The document collection we consider is the collection of academic researchers' publications. The documents are mapped to topics extracted from documents using Wikipedia to generate a document-topic representation. Then clustering algorithm is applied on document-topic representation to group semantically related researchers and generate a topic based academic researchers' Social Network. We use Adaptive Rough Fuzzy Leader (ARFL) a soft clustering algorithm, since each researcher can have expertise in more than one area and they can belong to more than one group. We present the empirical evaluation of our proposed scheme. We also demonstrate how our proposed solution is scalable to various domain areas and can be used to design topic based retrieval systems. [ABSTRACT FROM PUBLISHER]

Published
2012
Full Text
View/download PDF

3. Partition based pattern synthesis technique with efficient algorithms for nearest neighbor classification

Author

Viswanath, P., Murty, M. Narasimha, and Bhatnagar, Shalabh

Subjects
*CLASSIFICATION, *INFORMATION science, *ARTIFICIAL intelligence, *COGNITIVE science
Abstract

Abstract: Nearest neighbor (NN) classifier is a popular non-parametric classifier. It is conceptually a simple classifier and shows good performance. Due to the curse of dimensionality effect, the size of training set needed by it to achieve a given classification accuracy becomes prohibitively large when the dimensionality of the data is high. Generating artificial patterns can reduce this effect. In this paper, we propose a novel pattern synthesis method called partition based pattern synthesis which can generate an artificial training set of exponential order when compared with that of the given original training set. We also propose suitable faster NN based methods to work with the synthetic training patterns. Theoretically, the relationship between our methods and conventional NN methods is established. The computational requirements of our methods are also theoretically established. Experimental results show that NN based classifiers with synthetic training set can outperform conventional NN classifiers and some other related classifiers. [Copyright &y& Elsevier]

Published
2006
Full Text
View/download PDF

4. Highly Sparse Reductions to Kernel Spectral Clustering

Author

Rocco Langone, Raghvendra Mall, Johan A. K. Suykens, Maji, Pradipta, Ghosh, Ashish, Murty, M Narasimha, Ghosh, Kuntal, and Pal, Sankar K

Subjects
SISTA, Property (programming), Computer science, business.industry, Model selection, Pattern recognition, 02 engineering and technology, Extension (predicate logic), 01 natural sciences, Spectral clustering, Set (abstract data type), 010104 statistics & probability, ComputingMethodologies_PATTERNRECOGNITION, Balanced line, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Artificial intelligence, 0101 mathematics, business, Eigenvalues and eigenvectors, Selection (genetic algorithm)
Abstract

Kernel spectral clustering is a model-based spectral clustering method formulated in a primal-dual framework. It has a powerful out-of-sample extension property and a model selection procedure based on the balanced line fit criterion. This paper is an improvement of a previous work which sparsified the kernel spectral clustering method using the line structure of the data projections in the eigenspace. However, the previous method works only in the case of well formed and well separated clusters as in other cases the line structure is lost. In this paper, we propose two highly sparse extensions of kernel spectral clustering that can overcome these limitations. For the selection of the reduced set we use the concept of angles between the data projections in the eigenspace. We show the effectiveness and the amount of sparsity obtained by the proposed methods for several synthetic and real world datasets. © Springer-Verlag 2013. ispartof: pages:163-169 ispartof: Proc. of the 5th International Conference on Pattern Recognition and Machine Intelligence vol:8251 LNCS pages:163-169 ispartof: 5th International Conference on Pattern Recognition and Machine Intelligence location:Kolkata, India date:Dec - Dec 2013 status: published

Published
2013

Catalog

Books, media, physical & digital resources
See catalog results