Loganathan C - Academia.edu (original) (raw)
Papers by Loganathan C
Fuzzy K-means clustering algorithm is very much useful for exploring the structure of a set of pa... more Fuzzy K-means clustering algorithm is very much useful for exploring the structure of a set of patterns, especially
when the clusters are overlapping. K-means algorithm is simple with low time complexity, and can process the
large data set quickly. But conventional K-means algorithm cannot get high clustering precise rate, and easily be affected by clustering center random initialized and isolated points. This paper proposes an algorithm to compute initial cluster centers for K-means clustering. A cutting plane is used to partition the data in a cell that divides cell in to two smaller cells. The plane is perpendicular to the data axis with high variance and is intended to reduce the sum squared errors of the two cells while at the same time keeping the two cells apart. Cells are partitioned one at a time till the number of cells equals to the predefined number of clusters, K. The centers of the K cells become the initial cluster centers for K-means. The experimental results suggest that the proposed algorithm is effective, converge to better clustering results than those of the random initialization method. The research also indicated the proposed algorithm would greatly improve the likelihood of every cluster containing some data in it. The research also
indicated the proposed algorithm would greatly improve the likelihood of every cluster containing some data in it
Messages are given as text needs to be coded to avoid loss of secrecy in any transaction. This ha... more Messages are given as text needs to be coded to avoid loss of secrecy in any transaction. This has been
atempted through any media requires tough crypto analysis and can be handled through known ciphers. This job gets
better results while combining the text message along with other building blocks of multimedia. In this paper a noval
approach to insert a message on an image and both of them are passed in a media using elliptic curve crypto systems.
Caution is made in maintaining the quality of the vehicle image carrying the text. We have presented the algorithm and
illustrated through standard images used in image analysis.
In Hill cipher encryption, the complexity of finding the inverse of the matrix during decryption ... more In Hill cipher encryption, the complexity of finding the inverse of the matrix during decryption is eliminated by
adopting self invertible matrices. Authentication of hall tickets for candidates during examination is a complicated
issue. By evaluating simultaneously encryption and decryption with details of text, photos and signatures of candidates it is easy to prevent malpractices during examination held at various centres of the institution.
In this paper we present a method to identify the fuzzy game value of the matrix with interval da... more In this paper we present a method to identify the fuzzy game value of the matrix with interval data. The ideas of crisp matrix games are extended to fuzzy matrix games. This result is applied to a three player game.
In this paper, we evaluate the value of the fuzzy game matrix with interval data using the approx... more In this paper, we evaluate the value of the fuzzy game matrix with interval data using the approximate method. This
method gives an approximate solution for the value of the game and the true value can be determined to any degree of
accuracy. This method assumes that each player will play in such a manner so as to maximize the expected gain or to
minimize the expected loss.
Improvement in sensing and storage devices and impressive growth in applications such as Internet... more Improvement in sensing and storage devices and impressive growth in applications such as Internet search, digital imaging, and video surveillance have generated many high-volume, high-dimensional data. The raise in both the quantity and the kind of data requires improvement in techniques to understand, process and summarize the data. Categorizing data into reasonable groupings is one of the most essential techniques for understanding and learning. This is performed with the help of technique called clustering. This clustering technique is widely helpful in fields such as pattern recognition, image processing, and data analysis. The commonly used clustering technique is K-Means clustering. But this clustering results in misclassification when large data are involved in clustering. To overcome this disadvantage, Fuzzy-Possibilistic C-Means (FPCM) algorithm can be used for clustering. FPCM combines the advantages of Possibilistic C-Means (PCM) algorithm and fuzzy logic. For further improving the performance of clustering, penalized and compensated constraints are used in this paper. Penalized and compensated terms are embedded with the modified fuzzy possibilistic clustering method's objective function to construct the clustering with enhanced performance. The experimental result illustrates the enhanced performance of the proposed clustering technique when compared to the fuzzy possibilistic c-means clustering algorithm.
IJCSIS Volumes by Loganathan C
Fuzzy K-means clustering algorithm is very much useful for exploring the structure of a set of pa... more Fuzzy K-means clustering algorithm is very much useful for exploring the structure of a set of patterns, especially
when the clusters are overlapping. K-means algorithm is simple with low time complexity, and can process the
large data set quickly. But conventional K-means algorithm cannot get high clustering precise rate, and easily be affected by clustering center random initialized and isolated points. This paper proposes an algorithm to compute initial cluster centers for K-means clustering. A cutting plane is used to partition the data in a cell that divides cell in to two smaller cells. The plane is perpendicular to the data axis with high variance and is intended to reduce the sum squared errors of the two cells while at the same time keeping the two cells apart. Cells are partitioned one at a time till the number of cells equals to the predefined number of clusters, K. The centers of the K cells become the initial cluster centers for K-means. The experimental results suggest that the proposed algorithm is effective, converge to better clustering results than those of the random initialization method. The research also indicated the proposed algorithm would greatly improve the likelihood of every cluster containing some data in it. The research also
indicated the proposed algorithm would greatly improve the likelihood of every cluster containing some data in it
Messages are given as text needs to be coded to avoid loss of secrecy in any transaction. This ha... more Messages are given as text needs to be coded to avoid loss of secrecy in any transaction. This has been
atempted through any media requires tough crypto analysis and can be handled through known ciphers. This job gets
better results while combining the text message along with other building blocks of multimedia. In this paper a noval
approach to insert a message on an image and both of them are passed in a media using elliptic curve crypto systems.
Caution is made in maintaining the quality of the vehicle image carrying the text. We have presented the algorithm and
illustrated through standard images used in image analysis.
In Hill cipher encryption, the complexity of finding the inverse of the matrix during decryption ... more In Hill cipher encryption, the complexity of finding the inverse of the matrix during decryption is eliminated by
adopting self invertible matrices. Authentication of hall tickets for candidates during examination is a complicated
issue. By evaluating simultaneously encryption and decryption with details of text, photos and signatures of candidates it is easy to prevent malpractices during examination held at various centres of the institution.
In this paper we present a method to identify the fuzzy game value of the matrix with interval da... more In this paper we present a method to identify the fuzzy game value of the matrix with interval data. The ideas of crisp matrix games are extended to fuzzy matrix games. This result is applied to a three player game.
In this paper, we evaluate the value of the fuzzy game matrix with interval data using the approx... more In this paper, we evaluate the value of the fuzzy game matrix with interval data using the approximate method. This
method gives an approximate solution for the value of the game and the true value can be determined to any degree of
accuracy. This method assumes that each player will play in such a manner so as to maximize the expected gain or to
minimize the expected loss.
Improvement in sensing and storage devices and impressive growth in applications such as Internet... more Improvement in sensing and storage devices and impressive growth in applications such as Internet search, digital imaging, and video surveillance have generated many high-volume, high-dimensional data. The raise in both the quantity and the kind of data requires improvement in techniques to understand, process and summarize the data. Categorizing data into reasonable groupings is one of the most essential techniques for understanding and learning. This is performed with the help of technique called clustering. This clustering technique is widely helpful in fields such as pattern recognition, image processing, and data analysis. The commonly used clustering technique is K-Means clustering. But this clustering results in misclassification when large data are involved in clustering. To overcome this disadvantage, Fuzzy-Possibilistic C-Means (FPCM) algorithm can be used for clustering. FPCM combines the advantages of Possibilistic C-Means (PCM) algorithm and fuzzy logic. For further improving the performance of clustering, penalized and compensated constraints are used in this paper. Penalized and compensated terms are embedded with the modified fuzzy possibilistic clustering method's objective function to construct the clustering with enhanced performance. The experimental result illustrates the enhanced performance of the proposed clustering technique when compared to the fuzzy possibilistic c-means clustering algorithm.