A text classification system’s learning is substantially dependent on the input features and thei... more A text classification system’s learning is substantially dependent on the input features and their process of extraction and selection. The solitary drive encouraging feature selection practice is to lessen the dimensionality of the problem at hand; thus, facilitating the process of classification. Among several problem areas, text categorization is one area where feature selection plays a vital role. It is well-known that text categorization suffers from the curse of dimensionality. This results in the creation of feature space which may have redundant or irrelevant features leading to the creation of a poor classifier. Therefore, to build an intelligent classifier feature, selection is an important process. This paper has a fourfold objective: Firstly, it aims to create a word to vector space using a widely used score. Secondly, it intends to optimize text feature space using a nature-inspired algorithm. Thirdly, it aims at comparing classification performance of three prominently...
International Journal of Innovative Research in Computer and Communication Engineering, 2014
The continuing need for effective information retrieval, paved way for the inception of semantic ... more The continuing need for effective information retrieval, paved way for the inception of semantic web. It’s, unquestionably, as extension to the existing web which holds distributed information lacking any logical relationships amongst themselves. This information can be integrated and related to each other using ontologies. The paper intends to bridge this gap by providing a brief idea of ontology creation and how can it help in implementation of semantic web. The paper has threefold objective. Firstly the paper provides multiple definitions of ontology and their relation with in semantic web. Secondly the paper briefly provides a comparison of several ontology development tools justifies the use of a particular ontology tool. Lastly the paper proposes an ontology on data structures.
166 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Pvt. Ltd. Abstract—Re... more 166 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Pvt. Ltd. Abstract—Representation of distributed information, with a well defined meaning understandable for different parties, is the major challenge of Semantic Web. Several solutions have been built up. Use of Ontologies is one of the solutions to challenges faced by semantic web. This paper highlights importance of ontologies. This paper has three fold objectives. Firstly the paper throws light on how a semantic web based tool helps producing information using ontologies. Secondly, paper highlights the importance of ontology. Lastly a comparison of various tools for ontology development has been presented on various parameters
E-Learning, in the most basic terms, is utilization of technology to support delivery of educatio... more E-Learning, in the most basic terms, is utilization of technology to support delivery of education without any restriction on time and place. It can be said that it's a nonlinear and flexible approach, when viewed with reference to the traditional approach. Traditional way of learning or the classroom learning largely bounds the learner in terms of content and time besides putting restriction on user's pace of learning. By making excellent use of information and communication technologies, E-Learning offers competent solution to learner, keeping in view one's environment variables and reader's own pace and learning power. The learner controls his interactions with the content modules. Of course the efficiency of any E- learning module depends on learner's attitude towards learning and technology, availability and access to technology and his own technical background. This paper broadly analyzes the desire and openness, of students, to accept E-Learning as platfor...
• The SIMD model of parallel computing fitted into calibration sub-processes. • A methodology for... more • The SIMD model of parallel computing fitted into calibration sub-processes. • A methodology formulated around GPU hardware architecture and memory hierarchy. • Combines primitives of parallel implementations of ABC algorithm and k-NN algorithm. • NVIDIA Tesla C2050 GPU used with CUDA programming framework. • Over 10× acceleration achieved in calibration process. a b s t r a c t General purpose data parallel computing with graphical processing unit (GPU) is much structured today with NVIDIA R ⃝ CUDA and other parallel programming frameworks. Exploiting the CUDA programming framework, the present work proposes a novel methodology formulated around the GPU hardware architecture and memory hierarchy to accelerate the calibration process of a classification model named eNN10. Primarily developed for avalanche forecasting, eNN10 is based on brute force k-nearest neighbours (k-NN) approach and employs snow-meteorological variables to search for past days with similar conditions. The events associated with past similar days are then analysed to generate forecast. The model is required to be calibrated regularly to ensure higher degree of forecast accuracy in terms of Heidke skill score (HSS). The calibration of eNN10 is carried out by Artificial Bee Colony (ABC) algorithm, a swarm intelligence driven population based metaheuristic algorithm, and it requires thousands of HSS evaluations during the complete calibration process. A MATLAB sequential code for calibration runs for over 400 minutes and the proposed methodology delivered about 10× acceleration in calibration process. The methodology combines primitives of parallel implementations of brute force k-NN algorithm with that of population based metaheuristic algorithms and is scalable to deal with other similar real-world problems. The major objective of this paper is to highlight the methodology and associated future research areas.
A text classification system’s learning is substantially dependent on the input features and thei... more A text classification system’s learning is substantially dependent on the input features and their process of extraction and selection. The solitary drive encouraging feature selection practice is to lessen the dimensionality of the problem at hand; thus, facilitating the process of classification. Among several problem areas, text categorization is one area where feature selection plays a vital role. It is well-known that text categorization suffers from the curse of dimensionality. This results in the creation of feature space which may have redundant or irrelevant features leading to the creation of a poor classifier. Therefore, to build an intelligent classifier feature, selection is an important process. This paper has a fourfold objective: Firstly, it aims to create a word to vector space using a widely used score. Secondly, it intends to optimize text feature space using a nature-inspired algorithm. Thirdly, it aims at comparing classification performance of three prominently...
International Journal of Innovative Research in Computer and Communication Engineering, 2014
The continuing need for effective information retrieval, paved way for the inception of semantic ... more The continuing need for effective information retrieval, paved way for the inception of semantic web. It’s, unquestionably, as extension to the existing web which holds distributed information lacking any logical relationships amongst themselves. This information can be integrated and related to each other using ontologies. The paper intends to bridge this gap by providing a brief idea of ontology creation and how can it help in implementation of semantic web. The paper has threefold objective. Firstly the paper provides multiple definitions of ontology and their relation with in semantic web. Secondly the paper briefly provides a comparison of several ontology development tools justifies the use of a particular ontology tool. Lastly the paper proposes an ontology on data structures.
166 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Pvt. Ltd. Abstract—Re... more 166 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Pvt. Ltd. Abstract—Representation of distributed information, with a well defined meaning understandable for different parties, is the major challenge of Semantic Web. Several solutions have been built up. Use of Ontologies is one of the solutions to challenges faced by semantic web. This paper highlights importance of ontologies. This paper has three fold objectives. Firstly the paper throws light on how a semantic web based tool helps producing information using ontologies. Secondly, paper highlights the importance of ontology. Lastly a comparison of various tools for ontology development has been presented on various parameters
E-Learning, in the most basic terms, is utilization of technology to support delivery of educatio... more E-Learning, in the most basic terms, is utilization of technology to support delivery of education without any restriction on time and place. It can be said that it's a nonlinear and flexible approach, when viewed with reference to the traditional approach. Traditional way of learning or the classroom learning largely bounds the learner in terms of content and time besides putting restriction on user's pace of learning. By making excellent use of information and communication technologies, E-Learning offers competent solution to learner, keeping in view one's environment variables and reader's own pace and learning power. The learner controls his interactions with the content modules. Of course the efficiency of any E- learning module depends on learner's attitude towards learning and technology, availability and access to technology and his own technical background. This paper broadly analyzes the desire and openness, of students, to accept E-Learning as platfor...
• The SIMD model of parallel computing fitted into calibration sub-processes. • A methodology for... more • The SIMD model of parallel computing fitted into calibration sub-processes. • A methodology formulated around GPU hardware architecture and memory hierarchy. • Combines primitives of parallel implementations of ABC algorithm and k-NN algorithm. • NVIDIA Tesla C2050 GPU used with CUDA programming framework. • Over 10× acceleration achieved in calibration process. a b s t r a c t General purpose data parallel computing with graphical processing unit (GPU) is much structured today with NVIDIA R ⃝ CUDA and other parallel programming frameworks. Exploiting the CUDA programming framework, the present work proposes a novel methodology formulated around the GPU hardware architecture and memory hierarchy to accelerate the calibration process of a classification model named eNN10. Primarily developed for avalanche forecasting, eNN10 is based on brute force k-nearest neighbours (k-NN) approach and employs snow-meteorological variables to search for past days with similar conditions. The events associated with past similar days are then analysed to generate forecast. The model is required to be calibrated regularly to ensure higher degree of forecast accuracy in terms of Heidke skill score (HSS). The calibration of eNN10 is carried out by Artificial Bee Colony (ABC) algorithm, a swarm intelligence driven population based metaheuristic algorithm, and it requires thousands of HSS evaluations during the complete calibration process. A MATLAB sequential code for calibration runs for over 400 minutes and the proposed methodology delivered about 10× acceleration in calibration process. The methodology combines primitives of parallel implementations of brute force k-NN algorithm with that of population based metaheuristic algorithms and is scalable to deal with other similar real-world problems. The major objective of this paper is to highlight the methodology and associated future research areas.
Uploads
Papers by Pallavi Grover