amrita ticku - Academia.edu (original) (raw)

Papers by amrita ticku

Research paper thumbnail of IoT Based Human Biometric System Using Alcohol Level Detection for Driver Security

Research paper thumbnail of Utility of MATLAB and usage of backpropogation in Artificial neural network

International Journal of Innovative Research in Technology, Jun 6, 2020

The artificial neural network back propagation algorithm is implemented in Matlab Language. This ... more The artificial neural network back propagation algorithm is implemented in Matlab Language. This implementation is compared with several other software packages. The effect of reducing the number of iterations in the performance of the algorithm is studied. The speed of the back propagation program, backprop, written in Matlab language is compared with the speed of several other back propagation programs which are written in the C language. The speed of the Matlab program backprop is, also compared with the C program quickprop which is a variant of the back propagation algorithm. It is shown that the Matlab program backprop is about 4.5 to 7 times faster than the C programs.

Research paper thumbnail of Vader protocol based sentiment analysis technique using LSTM for Stock Trend prediction

Research paper thumbnail of Analysis of new differential evolution variants to solve multi-modal problems

IAES International Journal of Artificial Intelligence, Sep 1, 2023

Differential evolution algorithm (DE) requires diversified population to solve multi-modalproblem... more Differential evolution algorithm (DE) requires diversified population to solve multi-modalproblems. DE supports non-distributed population. DE versions include differential evolution algorithm with best selection method and species evolution (DESBS) and differential evolution algorithm with hierarchical fair competition modal (HFCDE). This article analyzed the efficiency of HFCDE and DESBS to solve the multi-modal s' problems. HFCDE and DESBS support non-distributed population structured. HFCDE starts with set of feasible solution then it distributed them in the different hierarchy. HFCDE provides the fair competition. DESBS is another semi distributed, differential evolution algorithm.It starts with set of feasible solutions (population). Later it identifies the niches and create the subgroups within population. Both HFCDE and DESBS have outperformed the other variant of state-of-art variants of DE. Here, the performance of DESBS and the HFCDE are rigorously tested on the multimodal problems. The success of DESBS over HFCDE in multi-modal difficulties managed to overcome the phenomena of elitism to resolve the complex problems, it has been observed that DESBS performs better than HFCDE in complex multi-modal scenarios.

Research paper thumbnail of Detailed Investigation of Cybersecurity in Governance

Zenodo (CERN European Organization for Nuclear Research), Feb 9, 2023

Research paper thumbnail of Development Of Test-Bed For Performance Evaluation Of Multiprotocol Label Switching

International journal of engineering research and technology, Dec 3, 2013

Multiprotocol Label Switching is called multi protocol because its techniques are applicable to a... more Multiprotocol Label Switching is called multi protocol because its techniques are applicable to any network layer protocol. The need to converge voice, data, and multimedia networks over IP based protocols has lead to the development of label switching. In this paper we develop a test bed to evaluate the performance of MPLS protocol. Simulations of this MPLS protocol have been carried out with the simulator Omnet++4.0.MPLS improves packet-forwarding performance in the network, supports Qos for service differentiation, supports network scalability, Integrates IP and ATM in the network and builds interoperable networks.

Research paper thumbnail of Comprehensive Survey of Deep Learning Applications in the Diagnosis of Epilepsy

Research paper thumbnail of Detailed Investigation About Cyber Security in Recent Times on Artificial Intelligence

Zenodo (CERN European Organization for Nuclear Research), Jan 24, 2023

Research paper thumbnail of Development Of Test-Bed For Performance Evaluation Of Multiprotocol Label Switching

Multiprotocol Label Switching is called multi protocol because its techniques are applicable to a... more Multiprotocol Label Switching is called multi protocol because its techniques are applicable to any network layer protocol. The need to converge voice, data, and multimedia networks over IP based protocols has lead to the development of label switching. In this paper we develop a test bed to evaluate the performance of MPLS protocol. Simulations of this MPLS protocol have been carried out with the simulator Omnet++4.0.MPLS improves packet-forwarding performance in the network, supports Qos for service differentiation, supports network scalability, Integrates IP and ATM in the network and builds interoperable networks.

Research paper thumbnail of Analysis of new differential evolution variants to solve multi-modal problems

IAES International Journal of Artificial Intelligence (IJ-AI)

Differential evolution algorithm requires diversified population to solve multi- modalproblems. D... more Differential evolution algorithm requires diversified population to solve multi- modalproblems. DE supports non-distributed population. Differential evolution algorithm (DE) versions include differential evolution algorithm with best selection method and species evolution (DESBS) and differential evolution algorithm with hierarchical fair competition modal (HFCDE). This article analyzed the efficiency of HFCDE and DESBS to solve the multi-modal s’ problems. HFCDE and DESBS support non-distributed population structured. HFCDE starts with set of feasible solution then it distributed them in the different hierarchy. HFCDE provides the fair competition. DESBS is another semi distributed, differential evolution algorithm.It starts with set of feasible solutions (population). Later it identifies the niches and create the sub-groups within population. Both HFCDE and DESBS have outperformed the other variant of state-of-art variants of DE. Here, the performance of DESBS and the HFCDE are ri...

Research paper thumbnail of Sentiment Analysis of Stocks Based on News Headlines Using NLP

Proceedings of the International Conference on Artificial Intelligence Techniques for Electrical Engineering Systems (AITEES 2022), 2023

In todays' world everyone starting from a child to an adult is studying stocks and is finding way... more In todays' world everyone starting from a child to an adult is studying stocks and is finding ways to earn more by studying the patterns of the market. Stock market is a compound interrelated system of various investors. It fluctuates frequently and hence is hard to predict what is yet to come. All the companies worldwide rely on these forecasts and speculations so that they can increase their profits. Everyday news on the economic front plays a vital role in defining the jump or drop in the prices of stocks. The market news helps the investor excessively in determining his bidding as it is immensely rich in information. In this study, we extract useful information from news headlines of a particular company to investigate the immediate impact of it on the company's stock growth. Having the text based dataset we use NLP and compare two approaches using two different algorithms which both collectively determine the sentiment of the news headline (whether it is positive/negative/neutral) in lieu with the company stocks.

Research paper thumbnail of Detailed Investigation About Cyber Security in Recent Times on Artificial Intelligence

Zenodo (CERN European Organization for Nuclear Research), Jan 24, 2023

Research paper thumbnail of Detailed Investigation of Cybersecurity in Governance

Zenodo (CERN European Organization for Nuclear Research), Feb 9, 2023

Research paper thumbnail of Accuracy Enhancement in Offline Signature Verification with The Use of Associative Memory

In today’s world every official document need the Hand written signature. To verify that the sign... more In today’s world every official document need the Hand written signature. To verify that the signature done by the person is accurate and not forged, human brain stores the pattern, data and everything so that they can recognize the pattern or data if again the same thing is shown. But the human mind receives the signal from eye which cannot catch the small variation in the pattern so it becomes a big problem for the legal issues. Associative memory is used for the verification of offline signature is used. In this paper we extend the algorithm for the memory of the system and how it will check the correct signature and forged signature .The Associative Memory Net (AMN) in correctly detecting forged signatures, fast. Here, the cost functions are handled with detail parametric studies and parallel processing using OpenMP. These algorithms are trained with the original or genuine signature and tested with a sample of ten very similar-looking forged signatures. The study concludes that...

Research paper thumbnail of Sentiment Analysis of Stocks Based on News Headlines Using NLP

Proceedings of the International Conference on Artificial Intelligence Techniques for Electrical Engineering Systems (AITEES 2022), 2020

In todays' world everyone starting from a child to an adult is studying stocks and is finding way... more In todays' world everyone starting from a child to an adult is studying stocks and is finding ways to earn more by studying the patterns of the market. Stock market is a compound interrelated system of various investors. It fluctuates frequently and hence is hard to predict what is yet to come. All the companies worldwide rely on these forecasts and speculations so that they can increase their profits. Everyday news on the economic front plays a vital role in defining the jump or drop in the prices of stocks. The market news helps the investor excessively in determining his bidding as it is immensely rich in information. In this study, we extract useful information from news headlines of a particular company to investigate the immediate impact of it on the company's stock growth. Having the text based dataset we use NLP and compare two approaches using two different algorithms which both collectively determine the sentiment of the news headline (whether it is positive/negative/neutral) in lieu with the company stocks.

Research paper thumbnail of Customer Segmentation - A Boon for Business

Research paper thumbnail of IoT Based Human Biometric System Using Alcohol Level Detection for Driver Security

Research paper thumbnail of Utility of MATLAB and usage of backpropogation in Artificial neural network

International Journal of Innovative Research in Technology, Jun 6, 2020

The artificial neural network back propagation algorithm is implemented in Matlab Language. This ... more The artificial neural network back propagation algorithm is implemented in Matlab Language. This implementation is compared with several other software packages. The effect of reducing the number of iterations in the performance of the algorithm is studied. The speed of the back propagation program, backprop, written in Matlab language is compared with the speed of several other back propagation programs which are written in the C language. The speed of the Matlab program backprop is, also compared with the C program quickprop which is a variant of the back propagation algorithm. It is shown that the Matlab program backprop is about 4.5 to 7 times faster than the C programs.

Research paper thumbnail of Vader protocol based sentiment analysis technique using LSTM for Stock Trend prediction

Research paper thumbnail of Analysis of new differential evolution variants to solve multi-modal problems

IAES International Journal of Artificial Intelligence, Sep 1, 2023

Differential evolution algorithm (DE) requires diversified population to solve multi-modalproblem... more Differential evolution algorithm (DE) requires diversified population to solve multi-modalproblems. DE supports non-distributed population. DE versions include differential evolution algorithm with best selection method and species evolution (DESBS) and differential evolution algorithm with hierarchical fair competition modal (HFCDE). This article analyzed the efficiency of HFCDE and DESBS to solve the multi-modal s' problems. HFCDE and DESBS support non-distributed population structured. HFCDE starts with set of feasible solution then it distributed them in the different hierarchy. HFCDE provides the fair competition. DESBS is another semi distributed, differential evolution algorithm.It starts with set of feasible solutions (population). Later it identifies the niches and create the subgroups within population. Both HFCDE and DESBS have outperformed the other variant of state-of-art variants of DE. Here, the performance of DESBS and the HFCDE are rigorously tested on the multimodal problems. The success of DESBS over HFCDE in multi-modal difficulties managed to overcome the phenomena of elitism to resolve the complex problems, it has been observed that DESBS performs better than HFCDE in complex multi-modal scenarios.

Research paper thumbnail of Detailed Investigation of Cybersecurity in Governance

Zenodo (CERN European Organization for Nuclear Research), Feb 9, 2023

Research paper thumbnail of Development Of Test-Bed For Performance Evaluation Of Multiprotocol Label Switching

International journal of engineering research and technology, Dec 3, 2013

Multiprotocol Label Switching is called multi protocol because its techniques are applicable to a... more Multiprotocol Label Switching is called multi protocol because its techniques are applicable to any network layer protocol. The need to converge voice, data, and multimedia networks over IP based protocols has lead to the development of label switching. In this paper we develop a test bed to evaluate the performance of MPLS protocol. Simulations of this MPLS protocol have been carried out with the simulator Omnet++4.0.MPLS improves packet-forwarding performance in the network, supports Qos for service differentiation, supports network scalability, Integrates IP and ATM in the network and builds interoperable networks.

Research paper thumbnail of Comprehensive Survey of Deep Learning Applications in the Diagnosis of Epilepsy

Research paper thumbnail of Detailed Investigation About Cyber Security in Recent Times on Artificial Intelligence

Zenodo (CERN European Organization for Nuclear Research), Jan 24, 2023

Research paper thumbnail of Development Of Test-Bed For Performance Evaluation Of Multiprotocol Label Switching

Multiprotocol Label Switching is called multi protocol because its techniques are applicable to a... more Multiprotocol Label Switching is called multi protocol because its techniques are applicable to any network layer protocol. The need to converge voice, data, and multimedia networks over IP based protocols has lead to the development of label switching. In this paper we develop a test bed to evaluate the performance of MPLS protocol. Simulations of this MPLS protocol have been carried out with the simulator Omnet++4.0.MPLS improves packet-forwarding performance in the network, supports Qos for service differentiation, supports network scalability, Integrates IP and ATM in the network and builds interoperable networks.

Research paper thumbnail of Analysis of new differential evolution variants to solve multi-modal problems

IAES International Journal of Artificial Intelligence (IJ-AI)

Differential evolution algorithm requires diversified population to solve multi- modalproblems. D... more Differential evolution algorithm requires diversified population to solve multi- modalproblems. DE supports non-distributed population. Differential evolution algorithm (DE) versions include differential evolution algorithm with best selection method and species evolution (DESBS) and differential evolution algorithm with hierarchical fair competition modal (HFCDE). This article analyzed the efficiency of HFCDE and DESBS to solve the multi-modal s’ problems. HFCDE and DESBS support non-distributed population structured. HFCDE starts with set of feasible solution then it distributed them in the different hierarchy. HFCDE provides the fair competition. DESBS is another semi distributed, differential evolution algorithm.It starts with set of feasible solutions (population). Later it identifies the niches and create the sub-groups within population. Both HFCDE and DESBS have outperformed the other variant of state-of-art variants of DE. Here, the performance of DESBS and the HFCDE are ri...

Research paper thumbnail of Sentiment Analysis of Stocks Based on News Headlines Using NLP

Proceedings of the International Conference on Artificial Intelligence Techniques for Electrical Engineering Systems (AITEES 2022), 2023

In todays' world everyone starting from a child to an adult is studying stocks and is finding way... more In todays' world everyone starting from a child to an adult is studying stocks and is finding ways to earn more by studying the patterns of the market. Stock market is a compound interrelated system of various investors. It fluctuates frequently and hence is hard to predict what is yet to come. All the companies worldwide rely on these forecasts and speculations so that they can increase their profits. Everyday news on the economic front plays a vital role in defining the jump or drop in the prices of stocks. The market news helps the investor excessively in determining his bidding as it is immensely rich in information. In this study, we extract useful information from news headlines of a particular company to investigate the immediate impact of it on the company's stock growth. Having the text based dataset we use NLP and compare two approaches using two different algorithms which both collectively determine the sentiment of the news headline (whether it is positive/negative/neutral) in lieu with the company stocks.

Research paper thumbnail of Detailed Investigation About Cyber Security in Recent Times on Artificial Intelligence

Zenodo (CERN European Organization for Nuclear Research), Jan 24, 2023

Research paper thumbnail of Detailed Investigation of Cybersecurity in Governance

Zenodo (CERN European Organization for Nuclear Research), Feb 9, 2023

Research paper thumbnail of Accuracy Enhancement in Offline Signature Verification with The Use of Associative Memory

In today’s world every official document need the Hand written signature. To verify that the sign... more In today’s world every official document need the Hand written signature. To verify that the signature done by the person is accurate and not forged, human brain stores the pattern, data and everything so that they can recognize the pattern or data if again the same thing is shown. But the human mind receives the signal from eye which cannot catch the small variation in the pattern so it becomes a big problem for the legal issues. Associative memory is used for the verification of offline signature is used. In this paper we extend the algorithm for the memory of the system and how it will check the correct signature and forged signature .The Associative Memory Net (AMN) in correctly detecting forged signatures, fast. Here, the cost functions are handled with detail parametric studies and parallel processing using OpenMP. These algorithms are trained with the original or genuine signature and tested with a sample of ten very similar-looking forged signatures. The study concludes that...

Research paper thumbnail of Sentiment Analysis of Stocks Based on News Headlines Using NLP

Proceedings of the International Conference on Artificial Intelligence Techniques for Electrical Engineering Systems (AITEES 2022), 2020

In todays' world everyone starting from a child to an adult is studying stocks and is finding way... more In todays' world everyone starting from a child to an adult is studying stocks and is finding ways to earn more by studying the patterns of the market. Stock market is a compound interrelated system of various investors. It fluctuates frequently and hence is hard to predict what is yet to come. All the companies worldwide rely on these forecasts and speculations so that they can increase their profits. Everyday news on the economic front plays a vital role in defining the jump or drop in the prices of stocks. The market news helps the investor excessively in determining his bidding as it is immensely rich in information. In this study, we extract useful information from news headlines of a particular company to investigate the immediate impact of it on the company's stock growth. Having the text based dataset we use NLP and compare two approaches using two different algorithms which both collectively determine the sentiment of the news headline (whether it is positive/negative/neutral) in lieu with the company stocks.

Research paper thumbnail of Customer Segmentation - A Boon for Business