C
C. Chandra Sekhar
Researcher at Indian Institute of Technology Madras
Publications - 54
Citations - 406
C. Chandra Sekhar is an academic researcher from Indian Institute of Technology Madras. The author has contributed to research in topics: Support vector machine & Kernel method. The author has an hindex of 11, co-authored 54 publications receiving 331 citations. Previous affiliations of C. Chandra Sekhar include Indian Institutes of Technology.
Papers
More filters
Journal ArticleDOI
GMM-based intermediate matching kernel for classification of varying length patterns of long duration speech using support vector machines.
A. D. Dileep,C. Chandra Sekhar +1 more
TL;DR: The posterior probability weighted DKs (including the proposed IMKs) are proposed to improve their classification performance and reduce the number of support vectors to improve its discrimination ability.
Proceedings ArticleDOI
Extraction of fixed dimension patterns from varying duration segments of consonant-vowel utterances
TL;DR: This paper proposes an approach for detection of VOP, based on dynamic time alignment between a reference pattern of a CV class and the pattern of an utterance of that class, and shows that the hypothesised VOPs using the proposed approach have less deviation from their actual locations.
Proceedings ArticleDOI
Representation and feature selection using multiple kernel learning
A. D. Dileep,C. Chandra Sekhar +1 more
TL;DR: Issues in the MKL algorithm are addressed in the framework of support vector machines (SVM) and studies on the representation and feature selection are presented for an image categorization task.
Proceedings ArticleDOI
A density based method for multivariate time series clustering in kernel feature space
S. Chandrakala,C. Chandra Sekhar +1 more
TL;DR: This paper proposes a density based clustering method in kernel feature space for clustering multivariate time series data of varying length and presents heuristic methods to find the initial values of the parameters used in this proposed algorithm.
Journal ArticleDOI
Large margin mixture of AR models for time series classification
TL;DR: The LMAR and LMMAR models provide a generative interpretation that enables utilization of the rejection option in the high risk classification tasks and give a better classification performance compared to the SVM based classifier.