Narendra Chaudhari - Academia.edu (original) (raw)

Uploads

Papers by Narendra Chaudhari

Lecture Notes in Computer Science, 2009

In this paper, we propose a novel fast training algorithm called Constructive Semi-Supervised Cla... more In this paper, we propose a novel fast training algorithm called Constructive Semi-Supervised Classification Algorithm (CS-SCA) for neural network construction based on the concept of geometrical expansion. Parameters are updated according to the geometrical location of the training samples in the input space, and each sample in the training set is learned only once. It's a semisupervised based approach, the training samples are semi-labeled i.e. for some samples, labels are known and for some samples, data labels are not known. The method starts with clustering, which is done by using the concept of geometrical expansion. In clustering process various clusters are formed. The clusters are visualizes in terms of hyperspheres. Once clustering process over labeling of hyperspheres is done, in which class is assigned to each hypersphere for classifying the multi-dimensional data. This constructive learning avoids blind selection of neural network structure. The method proposes here is exhaustively tested with different benchmark datasets and it is found that, on increasing value of training parameters number of hidden neurons and training time both are getting decrease. Through our experimental work we conclude that CS-SCA result in simple neural network structure by less training time.

Journal of Computer Science, 2006

We propose the design of output codes for solving the classification problem in Fast Covering Lea... more We propose the design of output codes for solving the classification problem in Fast Covering Learning Algorithm (FCLA). For a complex multi-class problem normally the classifiers are constructed by combining the outputs of several binary ones. In this paper, we use the basic methods of decomposition; one per class (OPC) and Error Correcting Output Code (ECOC) with FCLA, binary to binary mapping algorithm as a base binary learner. The methods have been tested on Fisher's wellknown Iris data set and experimental results show that the classification ability is improved by using ECOC method.

ABSTRACT Without Abstract

International Journal of Advanced Engineering Technology, 2017

This paper review the generation of slag from an integrated steel plant; focusing on, slag genera... more This paper review the generation of slag from an integrated steel plant; focusing on, slag generated in blast furnace during process of iron making and through EAF / BOF during process of steel making .The slag generated from BF and EAF/BOF are having different characteristic. The different type of slag having different chemical and physical properties, this depends on the chemical properties of input raw material charged in process of iron/steel making and this slag used in different process as raw material. Blast Furnace slag production ranges from about 220 to 370 kilograms per metric ton of pig iron produced; although lower grade ores may yield much higher slag fractions. Steel making process in electric arc furnaces generates up to 15 % of slag, which is, based on its properties, classified as non-hazardous waste. Disposal of such material requires large surfaces and it is rather unfavorable in economic terms.

277 Abstract— worldwide, the small and medium enterprises (SMEs) have been accepted as the engine... more 277 Abstract— worldwide, the small and medium enterprises (SMEs) have been accepted as the engine of economic growth and for promoting equitable development. The SMEs constitute over 90% of total enterprises in most of the economies and are credited with generating the highest rates of employment growth and account for a major share of industrial production and exports (www.msme.gov.in). In India too, the SMEs play a pivotal role in the overall industrial economy of the country. SMEs are playing significant role in supply chains of larger organizations. Singh, Garg, and Deshmukh (2004) opine that to sustain the importance and performances, SMEs are feeling more pressures to improve their competitiveness as compared to past protective markets. However, the potential of SMEs is often not realized because of problems commonly related to wide size, isolation, market opportunities, standards/quality, supply chains, logistics and technology, innovation, etc. In order to enable SMEs tide-o...

International Journal of Computer Applications, 2014

Bioinformatics is faced with accelerating increase of data set sizes originating from powerful hi... more Bioinformatics is faced with accelerating increase of data set sizes originating from powerful high-throughput measuring devices. Extensive computational power is the basic requirement for solving problems in bioinformatics. One of the key solutions to time-efficient data processing is the proper implementation of computational intensive tasks using parallel technology. A large number of cores is combined into a single chip to improve the overall performance of the multicore processors. This depicts the current trends in processor architecture. This paper proposes a new software-only speculative parallelization scheme for implementing RNA Secondary Structure Prediction algorithm in parallel. The scheme is developed after a systematic evaluation of the design options available. It is also shown to be efficient, robust and to outperform previously proposed schemes used for parallel implementation of RNA Secondary Structure Prediction.

Lecture Notes in Computer Science, 2009

In this paper, we propose a novel fast training algorithm called Constructive Semi-Supervised Cla... more In this paper, we propose a novel fast training algorithm called Constructive Semi-Supervised Classification Algorithm (CS-SCA) for neural network construction based on the concept of geometrical expansion. Parameters are updated according to the geometrical location of the training samples in the input space, and each sample in the training set is learned only once. It's a semisupervised based approach, the training samples are semi-labeled i.e. for some samples, labels are known and for some samples, data labels are not known. The method starts with clustering, which is done by using the concept of geometrical expansion. In clustering process various clusters are formed. The clusters are visualizes in terms of hyperspheres. Once clustering process over labeling of hyperspheres is done, in which class is assigned to each hypersphere for classifying the multi-dimensional data. This constructive learning avoids blind selection of neural network structure. The method proposes here is exhaustively tested with different benchmark datasets and it is found that, on increasing value of training parameters number of hidden neurons and training time both are getting decrease. Through our experimental work we conclude that CS-SCA result in simple neural network structure by less training time.

Journal of Computer Science, 2006

We propose the design of output codes for solving the classification problem in Fast Covering Lea... more We propose the design of output codes for solving the classification problem in Fast Covering Learning Algorithm (FCLA). For a complex multi-class problem normally the classifiers are constructed by combining the outputs of several binary ones. In this paper, we use the basic methods of decomposition; one per class (OPC) and Error Correcting Output Code (ECOC) with FCLA, binary to binary mapping algorithm as a base binary learner. The methods have been tested on Fisher's wellknown Iris data set and experimental results show that the classification ability is improved by using ECOC method.

ABSTRACT Without Abstract

International Journal of Advanced Engineering Technology, 2017

This paper review the generation of slag from an integrated steel plant; focusing on, slag genera... more This paper review the generation of slag from an integrated steel plant; focusing on, slag generated in blast furnace during process of iron making and through EAF / BOF during process of steel making .The slag generated from BF and EAF/BOF are having different characteristic. The different type of slag having different chemical and physical properties, this depends on the chemical properties of input raw material charged in process of iron/steel making and this slag used in different process as raw material. Blast Furnace slag production ranges from about 220 to 370 kilograms per metric ton of pig iron produced; although lower grade ores may yield much higher slag fractions. Steel making process in electric arc furnaces generates up to 15 % of slag, which is, based on its properties, classified as non-hazardous waste. Disposal of such material requires large surfaces and it is rather unfavorable in economic terms.

277 Abstract— worldwide, the small and medium enterprises (SMEs) have been accepted as the engine... more 277 Abstract— worldwide, the small and medium enterprises (SMEs) have been accepted as the engine of economic growth and for promoting equitable development. The SMEs constitute over 90% of total enterprises in most of the economies and are credited with generating the highest rates of employment growth and account for a major share of industrial production and exports (www.msme.gov.in). In India too, the SMEs play a pivotal role in the overall industrial economy of the country. SMEs are playing significant role in supply chains of larger organizations. Singh, Garg, and Deshmukh (2004) opine that to sustain the importance and performances, SMEs are feeling more pressures to improve their competitiveness as compared to past protective markets. However, the potential of SMEs is often not realized because of problems commonly related to wide size, isolation, market opportunities, standards/quality, supply chains, logistics and technology, innovation, etc. In order to enable SMEs tide-o...

International Journal of Computer Applications, 2014

Bioinformatics is faced with accelerating increase of data set sizes originating from powerful hi... more Bioinformatics is faced with accelerating increase of data set sizes originating from powerful high-throughput measuring devices. Extensive computational power is the basic requirement for solving problems in bioinformatics. One of the key solutions to time-efficient data processing is the proper implementation of computational intensive tasks using parallel technology. A large number of cores is combined into a single chip to improve the overall performance of the multicore processors. This depicts the current trends in processor architecture. This paper proposes a new software-only speculative parallelization scheme for implementing RNA Secondary Structure Prediction algorithm in parallel. The scheme is developed after a systematic evaluation of the design options available. It is also shown to be efficient, robust and to outperform previously proposed schemes used for parallel implementation of RNA Secondary Structure Prediction.