Abhishek Chunawale
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Abhishek Chunawale has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Sumedha Sirsikar
Maharashtra Institute of Technology
3 shared publications - Mangesh Bedekar
Dr. Vishwanath Karad MIT World Peace University
2 shared publications - Anand Patekar
Department of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
1 shared publication - Anoushka Kachewar
Department of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
1 shared publication - Dipali Baviskar
Department of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
1 shared publication - Harsh Halwai
Department of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
1 shared publication
Latest Publications
- Open Access
- 1 February 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (2) , 920-931
Electroencephalogram based human emotion classification for valence and arousal using machine learning approach
Humans have unique ability to express emotions and electroencephalogram (EEG) signals are one of the sought-after ways to analyze a person’s emotional state. However, extracting proper emotion related features from EEG and finding corresponding emotion is challenging because of complex nature of emotions and underlying brain activities. The objective of this paper is to address this issue for more accurate emotion classification based on EEG. It also compares feature extraction methods namely fast fourier transform (FFT) and discrete wavelet transform (DWT). DEAP dataset is used for classification of human emotions through support vector machine (SVM) and K-nearest neighbor (KNN) algorithms by considering features such as standard deviation, mean, variance, power spectrum density (PSD) for FFT; and energy, entropy for DWT. It is observed that feature extraction from FFT yielded better results than DWT and KNN gave more accuracy of 96.61% for valence and 96.42% for arousal as compared to SVM. The proposed method based on PSD and FFT fared better than other existing ones in terms of accuracy when compared against different features and feature extraction techniques. This approach is expected to help researchers to understand feature extraction from EEG signals and decide proper features and techniques for better implementation.
- 1 February 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (2) , 920-931
Open AccessElectroencephalogram based human emotion classification for valence and arousal using machine learning approach
Humans have unique ability to express emotions and electroencephalogram (EEG) signals are one of the sought-after ways to analyze a person’s emotional state. However, extracting proper emotion related features from EEG and finding corresponding emotion is challenging because of complex nature of emotions and underlying brain activities. The objective of this paper is to address this issue for more accurate emotion classification based on EEG. It also compares feature extraction methods namely fast fourier transform (FFT) and discrete wavelet transform (DWT). DEAP dataset is used for classification of human emotions through support vector machine (SVM) and K-nearest neighbor (KNN) algorithms by considering features such as standard deviation, mean, variance, power spectrum density (PSD) for FFT; and energy, entropy for DWT. It is observed that feature extraction from FFT yielded better results than DWT and KNN gave more accuracy of 96.61% for valence and 96.42% for arousal as compared to SVM. The proposed method based on PSD and FFT fared better than other existing ones in terms of accuracy when compared against different features and feature extraction techniques. This approach is expected to help researchers to understand feature extraction from EEG signals and decide proper features and techniques for better implementation.
- Open Access
- 18 August 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (8s) , 376-380
Machine Learning based Stress Detection using Keyboard Typing Behavior
Emotion detection is one of those areas where technological advances have brought about significant changesin the human lifestyle. During COVID-19 pandemic, due to the work from home culture, use of computers and laptop was suddenly increased. Introduction of digital environments gave it a whole new dimension. Emotion detection is a virtual or computerized way to detect stress. People suffer from various kinds of stress in day to day activities and it is directly connected to their performance. The stress factor can be expressed through a number of ways and human behavior. The way in which humans interact with the computer can reveal the emotional state of the user, mainly the stress. Keyboard typing behavior or characteristics can be used for stress detection. This paper focuses on understanding typing behaviour of human and indicate their stress level. Relevant features are extracted from typing behavior of a user and used for training machine learning models for detection of stress. K-Nearest Neighbor algorithm gave highest accuracy of 84.21% with dimensionality reduction approach.
- 18 August 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (8s) , 376-380
Open AccessMachine Learning based Stress Detection using Keyboard Typing Behavior
Emotion detection is one of those areas where technological advances have brought about significant changesin the human lifestyle. During COVID-19 pandemic, due to the work from home culture, use of computers and laptop was suddenly increased. Introduction of digital environments gave it a whole new dimension. Emotion detection is a virtual or computerized way to detect stress. People suffer from various kinds of stress in day to day activities and it is directly connected to their performance. The stress factor can be expressed through a number of ways and human behavior. The way in which humans interact with the computer can reveal the emotional state of the user, mainly the stress. Keyboard typing behavior or characteristics can be used for stress detection. This paper focuses on understanding typing behaviour of human and indicate their stress level. Relevant features are extracted from typing behavior of a user and used for training machine learning models for detection of stress. K-Nearest Neighbor algorithm gave highest accuracy of 84.21% with dimensionality reduction approach.
- 8 August 2017
- Published by Springer Nature
- p. 155-163
RED: Residual Energy and Distance Based Clustering to Avoid Energy Hole Problem in Self-organized Wireless Sensor Networks
- 8 August 2017
- Published by Springer Nature
- p. 155-163
RED: Residual Energy and Distance Based Clustering to Avoid Energy Hole Problem in Self-organized Wireless Sensor Networks
- 1 December 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 244-248
Minimization of average energy consumption to prolong lifetime of Wireless Sensor Network
Wireless Sensor Network (WSN) is a network of sensor nodes that can sense the environment and send the information through wireless links to a sink. Wireless sensor nodes possess limited processing capability, storage and energy resources. The existence of sensor network depends on the life of sensor nodes i.e. ultimately on the energy consumption during its operation. Thus, in WSN, the efficient use of energy resources is very much necessary. Clustering is one of the approaches for energy saving in WSN. A cluster is a group of sensor nodes with one central entity named Cluster Head (CH). In this article, a new way of clustering for CH selection and cluster formation in WSNs is proposed; in which sensor network is divided into zones as per geographic locations of nodes. Clusters are formed within the zones by taking into account the residual energy of nodes and node distance. All cluster members send the sensed data to their respective CH. Hence, CH consumes more energy as it processes the collected data before forwarding it to Base Station (BS). Dropping of cluster head's residual energy below threshold value initiates cluster reformation. Performance analysis and simulation results are given with variations in number of nodes and transmission range. The obtained results show good performance of our algorithm in terms of reduced energy consumption, increased network lifetime and scalability of the network.
- 1 December 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 244-248
Minimization of average energy consumption to prolong lifetime of Wireless Sensor Network
Wireless Sensor Network (WSN) is a network of sensor nodes that can sense the environment and send the information through wireless links to a sink. Wireless sensor nodes possess limited processing capability, storage and energy resources. The existence of sensor network depends on the life of sensor nodes i.e. ultimately on the energy consumption during its operation. Thus, in WSN, the efficient use of energy resources is very much necessary. Clustering is one of the approaches for energy saving in WSN. A cluster is a group of sensor nodes with one central entity named Cluster Head (CH). In this article, a new way of clustering for CH selection and cluster formation in WSNs is proposed; in which sensor network is divided into zones as per geographic locations of nodes. Clusters are formed within the zones by taking into account the residual energy of nodes and node distance. All cluster members send the sensed data to their respective CH. Hence, CH consumes more energy as it processes the collected data before forwarding it to Base Station (BS). Dropping of cluster head's residual energy below threshold value initiates cluster reformation. Performance analysis and simulation results are given with variations in number of nodes and transmission range. The obtained results show good performance of our algorithm in terms of reduced energy consumption, increased network lifetime and scalability of the network.
- 1 January 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 204-208
Self-Organization Architecture and Model for Wireless Sensor Networks
In wireless sensor networks (WSN) there is a need to have self-formation capabilities into an existing variety of composite computing systems. In view of the fact that functionality is distributed over a number of components in a remote area, it would be an undefeatable task to accumulate and instruct each of them individually. All network entities should behave in a self-organized manner that means the global behavior emerges through their operation on local information and direct and/or indirect mutual interaction. Based on the all characteristics, we propose architecture of self-organizing wireless sensor networks (SOWSN). The motivation to develop this model is multifold. There is need to study various parameters of system that are required by the WSN to be self-organized. Hence we have come up with a model that forms a base for self-organization of WSN. This model provides ability to recognize and study significant parameters of a system along with its essential metrics.
- 1 January 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 204-208
Self-Organization Architecture and Model for Wireless Sensor Networks
In wireless sensor networks (WSN) there is a need to have self-formation capabilities into an existing variety of composite computing systems. In view of the fact that functionality is distributed over a number of components in a remote area, it would be an undefeatable task to accumulate and instruct each of them individually. All network entities should behave in a self-organized manner that means the global behavior emerges through their operation on local information and direct and/or indirect mutual interaction. Based on the all characteristics, we propose architecture of self-organizing wireless sensor networks (SOWSN). The motivation to develop this model is multifold. There is need to study various parameters of system that are required by the WSN to be self-organized. Hence we have come up with a model that forms a base for self-organization of WSN. This model provides ability to recognize and study significant parameters of a system along with its essential metrics.