Sunita Kumari Chaurasia
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sunita Kumari Chaurasia has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
S.R.N Reddy
2 shared publicationsIgdtuw Cse Dept
1 shared publicationS. R. N. Reddy
Department of Computer Science & Engineering, Indira Gandhi Delhi Technical University for Women (IGDTUW), Delhi-110006, India
1 shared publication
Latest Publications
- 22 September 2021
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 81 (1) , 1077-1108
State-of-the-art survey on activity recognition and classification using smartphones and wearable sensors
- 22 September 2021
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 81 (1) , 1077-1108
State-of-the-art survey on activity recognition and classification using smartphones and wearable sensors
- 30 August 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Engineering and Advanced Technology
- Vol. 8 (6) , 2143-2148
AI Assisted Human Activity Recognition (HAR)
Human Activity Recognition and assisting user on the basis of his context is attracting researchers since decade Researchers are working in the area to increase the accuracy of detection by various means. The challenging issue is to determine the correct supervised classifier for the detection purpose. This paper intent to examine the methodology used to recognize HAR and the impact of classifiers practiced in training and Testing. We have also tried to identify the suitable supervised machine learning model for HAR. Data of 30 Users with 561 features belonging to accelerometer and gyroscope sensor of smartphone from UCI repository is used for evaluation purpose. Nine different supervised machine learning Models are trained and tested on the dataset. The result concludes that HAR is a process which depends upon the classifiers used. It also conclude that out of 9 different Machine learning models ANN performs well and after that SVM, kNN, Random Forest and Extra Tree are equally good models for the purpose of HAR with Accuracy and execution time as the performance evaluation metric.
- 30 August 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Engineering and Advanced Technology
- Vol. 8 (6) , 2143-2148
AI Assisted Human Activity Recognition (HAR)
Human Activity Recognition and assisting user on the basis of his context is attracting researchers since decade Researchers are working in the area to increase the accuracy of detection by various means. The challenging issue is to determine the correct supervised classifier for the detection purpose. This paper intent to examine the methodology used to recognize HAR and the impact of classifiers practiced in training and Testing. We have also tried to identify the suitable supervised machine learning model for HAR. Data of 30 Users with 561 features belonging to accelerometer and gyroscope sensor of smartphone from UCI repository is used for evaluation purpose. Nine different supervised machine learning Models are trained and tested on the dataset. The result concludes that HAR is a process which depends upon the classifiers used. It also conclude that out of 9 different Machine learning models ANN performs well and after that SVM, kNN, Random Forest and Extra Tree are equally good models for the purpose of HAR with Accuracy and execution time as the performance evaluation metric.
- Open Access
- 30 October 2018
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 4 (16) , e5
Design and Implementation of Data Collection & Analysis Tool for Healthcare Parameter Monitoring using Inverse Low Pass Filter
BACKGROUND: Health of an individual can be determined by monitoring his daily activities. Human Activity Recognition (HAR) is a process, used to record these activities. The first step of HAR involves collection of Data from various Sensors, which is challenging due to the requirement of interfacing of hardware Sensors. Researchers either do hardware interfacing themselves or just analyse already available dataset. In the literature, many measures are proposed to calculate the human activities by using the Accelerometer data. But, different activities recorded by Accelerometer are not accurate due to presence of gravity component. Also there is requirement of identification of appropriate Machine learning algorithm for HAR.OBJECTIVE: To facilitate the researchers in hassle free data collection for monitoring HAR and finding optimal cutoff frequency to remove gravity component present in accelerometer data. Also detecting suitable machine learning model for HAR.METHODS: An application has been developed using Inverse low pass filter with an optimal threshold value to compute cut-off frequency for the removal of gravity component from accelerometer data. 8 different machine learning models are trained and tested for examining suitability of models on the collected data at server. Accuracy and execution time are considered as validation parameters.RESULTS: Gravity Component is removed and data gathered by using 0.6 alpha is providing 0.05 mean acceleration in static state. Among Various ML models k Nearest Neighbor is providing 93.17 percent of accuracy with 0.187486 sec of execution time.CONCLUSION: Removal of Gravity component is optimised by our method and out of various ML model kNN is identified as suitable model for HAR.
- 30 October 2018
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 4 (16) , e5
Open AccessDesign and Implementation of Data Collection & Analysis Tool for Healthcare Parameter Monitoring using Inverse Low Pass Filter
BACKGROUND: Health of an individual can be determined by monitoring his daily activities. Human Activity Recognition (HAR) is a process, used to record these activities. The first step of HAR involves collection of Data from various Sensors, which is challenging due to the requirement of interfacing of hardware Sensors. Researchers either do hardware interfacing themselves or just analyse already available dataset. In the literature, many measures are proposed to calculate the human activities by using the Accelerometer data. But, different activities recorded by Accelerometer are not accurate due to presence of gravity component. Also there is requirement of identification of appropriate Machine learning algorithm for HAR.OBJECTIVE: To facilitate the researchers in hassle free data collection for monitoring HAR and finding optimal cutoff frequency to remove gravity component present in accelerometer data. Also detecting suitable machine learning model for HAR.METHODS: An application has been developed using Inverse low pass filter with an optimal threshold value to compute cut-off frequency for the removal of gravity component from accelerometer data. 8 different machine learning models are trained and tested for examining suitability of models on the collected data at server. Accuracy and execution time are considered as validation parameters.RESULTS: Gravity Component is removed and data gathered by using 0.6 alpha is providing 0.05 mean acceleration in static state. Among Various ML models k Nearest Neighbor is providing 93.17 percent of accuracy with 0.187486 sec of execution time.CONCLUSION: Removal of Gravity component is optimised by our method and out of various ML model kNN is identified as suitable model for HAR.