Ajay Agarwal
Research Subjects
Research Interests
Ajay Agarwal has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Apeksha Aggarwal
Indian Institute of Technology Roorkee
2 shared publications - Abeer Ali Alnuaim
Department of Computer Science and Engineering College of Applied Studies King Saud University 11451 Riyadh Saudi Arabia
1 shared publication - Akshat Srivastava
School of Computer Science Engineering and Technology, Bennett University, Plot Nos 8-11, TechZone 2, Greater Noida 201310, India
1 shared publication - Aseel Alhadlaq
Department of Computer Science and Engineering, College of Applied Studies and Community Services, King Saud University, Riyadh, Saudi Arabia
1 shared publication - Dilbag Singh
Center of Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, NY 10016, USA
1 shared publication - Gaurav Agarwal
Veer Madho Singh Bhandari Uttarakhand Technical University
1 shared publication
Latest Publications
- 7 September 2023
- Published by Springer Nature
- p. 51-59
Fake News Detection Using Machine Learning
- Open Access
- 14 August 2023
- Published by Springer Nature in Research Square
Optimized Routing Protocol Using Metaheuristic Techniques for Improving Energy Efficiency in WSN
A Wireless Sensor Network is a group of small, autonomous sensor nodes which are connected wirelessly and have sensing, processing, and communication capabilities. These nodes are responsible for gathering and monitoring physical information from the surrounding environment. Since these nodes are typically compact and operate with a battery, they have a limited power supply which leads to a constrained network lifetime. Therefore, the use of energy effective techniques that lengthen the network’s lifespan is highly significant. Hence, the objective of this research work is to reduce energy usage and increase the network’s lifespan. Clustering and optimization techniques are commonly used in sensor network to enhance its stability and lifespan. In a clustered sensor network, cluster heads play a crucial role, as they are responsible for performing various tasks that consume more energy. This research work aims to enhance the effectiveness of the network by proposing a hybrid nature-inspired optimization algorithm named as Energy Efficient Yellow Saddle Goatfish Pelican Optimization algorithm (EEYSGPO) which uses Yellow Saddle Goatfish Algorithm to identify the optimum cluster head from a set of nodes. The parameters like residual energy, distance, delay, load, and communication quality are all used to select the optimized cluster head in the clusters. After choosing the optimized cluster head, pelican optimization technique is used to determine the best route for communication between cluster head and the base station, which is calculated on the basis of distance and residual energy. The MATLAB simulator is used for simulation and the obtained results of suggested technique were compared with several existing energy-efficient techniques such as EECHIGWO, SSMOECHS, FGWSTERP and LEACH-PRO using performance measures such as throughput, dead nodes and energy consumption. Simulation findings reveal that the optimal selection of cluster heads and routes in EEYSPO algorithm resolved the issues related to premature convergence and increase the lifetime and scalability of WSN. When compared to the EECHIGWO, SSMOECHS, LEACH-PRO and FGWSTERP protocols, the proposed methodology improves network stability by 57.28 %, 324.5 %, 571.72 % and 91.37 % respectively.
- 14 August 2023
- Published by Springer Nature in Research Square
Open AccessOptimized Routing Protocol Using Metaheuristic Techniques for Improving Energy Efficiency in WSN
A Wireless Sensor Network is a group of small, autonomous sensor nodes which are connected wirelessly and have sensing, processing, and communication capabilities. These nodes are responsible for gathering and monitoring physical information from the surrounding environment. Since these nodes are typically compact and operate with a battery, they have a limited power supply which leads to a constrained network lifetime. Therefore, the use of energy effective techniques that lengthen the network’s lifespan is highly significant. Hence, the objective of this research work is to reduce energy usage and increase the network’s lifespan. Clustering and optimization techniques are commonly used in sensor network to enhance its stability and lifespan. In a clustered sensor network, cluster heads play a crucial role, as they are responsible for performing various tasks that consume more energy. This research work aims to enhance the effectiveness of the network by proposing a hybrid nature-inspired optimization algorithm named as Energy Efficient Yellow Saddle Goatfish Pelican Optimization algorithm (EEYSGPO) which uses Yellow Saddle Goatfish Algorithm to identify the optimum cluster head from a set of nodes. The parameters like residual energy, distance, delay, load, and communication quality are all used to select the optimized cluster head in the clusters. After choosing the optimized cluster head, pelican optimization technique is used to determine the best route for communication between cluster head and the base station, which is calculated on the basis of distance and residual energy. The MATLAB simulator is used for simulation and the obtained results of suggested technique were compared with several existing energy-efficient techniques such as EECHIGWO, SSMOECHS, FGWSTERP and LEACH-PRO using performance measures such as throughput, dead nodes and energy consumption. Simulation findings reveal that the optimal selection of cluster heads and routes in EEYSPO algorithm resolved the issues related to premature convergence and increase the lifetime and scalability of WSN. When compared to the EECHIGWO, SSMOECHS, LEACH-PRO and FGWSTERP protocols, the proposed methodology improves network stability by 57.28 %, 324.5 %, 571.72 % and 91.37 % respectively.
- 2 June 2023
- Published by Springer Nature
- p. 645-654
An Intelligent Framework for Movie Recommendation Through Online Social Media
- 2 June 2023
- Published by Springer Nature
- p. 645-654
An Intelligent Framework for Movie Recommendation Through Online Social Media
- 23 June 2022
- Published by Springer Nature
- p. 1-9
Two-terminal Reliability Assessments of a Complex System
- 23 June 2022
- Published by Springer Nature
- p. 1-9
Two-terminal Reliability Assessments of a Complex System
- Top Cited PapersOpen Access
Two-Way Feature Extraction for Speech Emotion Recognition Using Deep Learning
Recognizing human emotions by machines is a complex task. Deep learning models attempt to automate this process by rendering machines to exhibit learning capabilities. However, identifying human emotions from speech with good performance is still challenging. With the advent of deep learning algorithms, this problem has been addressed recently. However, most research work in the past focused on feature extraction as only one method for training. In this research, we have explored two different methods of extracting features to address effective speech emotion recognition. Initially, two-way feature extraction is proposed by utilizing super convergence to extract two sets of potential features from the speech data. For the first set of features, principal component analysis (PCA) is applied to obtain the first feature set. Thereafter, a deep neural network (DNN) with dense and dropout layers is implemented. In the second approach, mel-spectrogram images are extracted from audio files, and the 2D images are given as input to the pre-trained VGG-16 model. Extensive experiments and an in-depth comparative analysis over both the feature extraction methods with multiple algorithms and over two datasets are performed in this work. The RAVDESS dataset provided significantly better accuracy than using numeric features on a DNN.
Top Cited PapersOpen AccessTwo-Way Feature Extraction for Speech Emotion Recognition Using Deep Learning
Recognizing human emotions by machines is a complex task. Deep learning models attempt to automate this process by rendering machines to exhibit learning capabilities. However, identifying human emotions from speech with good performance is still challenging. With the advent of deep learning algorithms, this problem has been addressed recently. However, most research work in the past focused on feature extraction as only one method for training. In this research, we have explored two different methods of extracting features to address effective speech emotion recognition. Initially, two-way feature extraction is proposed by utilizing super convergence to extract two sets of potential features from the speech data. For the first set of features, principal component analysis (PCA) is applied to obtain the first feature set. Thereafter, a deep neural network (DNN) with dense and dropout layers is implemented. In the second approach, mel-spectrogram images are extracted from audio files, and the 2D images are given as input to the pre-trained VGG-16 model. Extensive experiments and an in-depth comparative analysis over both the feature extraction methods with multiple algorithms and over two datasets are performed in this work. The RAVDESS dataset provided significantly better accuracy than using numeric features on a DNN.