Bhasker Pant
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
Research Subjects
Bhasker Pant has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Devesh Pratap Singh
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India
4 shared publications - Kumud Pant
Department of Biotechnology, Graphic Era Deemed to be University, Dehradun 248002, India
2 shared publications - Madhu Sharma Gaur
G.L. Bajaj Institute of Management and Technology, India
2 shared publications - Neelam Singh
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India
2 shared publications - Rayan H. Goudar
Visvesvaraya Technological University (VTU)
2 shared publications - Abdullah Alshehri
Faculty of Computing and Information, Al-Baha University, Al-Baha 65779, Saudi Arabia
1 shared publication
Latest Publications
- Open Access
- 7 October 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (10s) , 588-599
Forecasting the Missing Links in Heterogeneous Social Networks
Social network analysis has gained attention from several researchers in the past time because of its wide application in capturing social interactions. One of the aims of social network analysis is to recover missing links between the users which may exist in the future but have not yet appeared due to incomplete data. The prediction of hidden or missing links in criminal networks is also a significant problem. The collection of criminal data from these networks appears to be incomplete and inconsistent which is reflected in the structure in the form of missing nodes and links. Many machine learning algorithms are applied for this detection using supervised techniques. But, supervised machine learning algorithms require large datasets for training the link prediction model for achieving optimum results. In this research, we have used a Facebook dataset to solve the problem of link prediction in a network. The two machine learning classifiers applied are LogisticRegression and K-Nearest Neighbour where KNN has higher accuracy than LR. In this article, we have proposed an algorithm Graph Sample Aggregator with Low Reciprocity, (GraphSALR), for the generation of node embeddings in larger graphs which use node feature information.
- 7 October 2023
- Published by Auricle Technologies, Pvt., Ltd. in International Journal on Recent and Innovation Trends in Computing and Communication
- Vol. 11 (10s) , 588-599
Open AccessForecasting the Missing Links in Heterogeneous Social Networks
Social network analysis has gained attention from several researchers in the past time because of its wide application in capturing social interactions. One of the aims of social network analysis is to recover missing links between the users which may exist in the future but have not yet appeared due to incomplete data. The prediction of hidden or missing links in criminal networks is also a significant problem. The collection of criminal data from these networks appears to be incomplete and inconsistent which is reflected in the structure in the form of missing nodes and links. Many machine learning algorithms are applied for this detection using supervised techniques. But, supervised machine learning algorithms require large datasets for training the link prediction model for achieving optimum results. In this research, we have used a Facebook dataset to solve the problem of link prediction in a network. The two machine learning classifiers applied are LogisticRegression and K-Nearest Neighbour where KNN has higher accuracy than LR. In this article, we have proposed an algorithm Graph Sample Aggregator with Low Reciprocity, (GraphSALR), for the generation of node embeddings in larger graphs which use node feature information.
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2782 (1) , 020135
A review of network slicing based 5G
5G is planned as a multi-service network that can serve a variety of verticals with varying customer satisfaction and service needs. A key to accomplishing this
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2782 (1) , 020135
Open AccessA review of network slicing based 5G
5G is planned as a multi-service network that can serve a variety of verticals with varying customer satisfaction and service needs. A key to accomplishing this
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2521 (1)
TRS: Trusted recommendation system for fast changing dataset
Recommendations can be used in two ways, one for the mass audience and second for personalizing, in the mass audience like weather recommendation, stock recommendation, so this type of recommendation is based on past generic data. However, personalized recommendations are based on unique dataset of users, and at the time of changing location and demographic condition changing frequently, it becomes more complex. This type of recommendation entirely depends upon the dynamic profile of the user, and this dynamic behavior can alter by the third person for fake and biased suggestions. In this paper, we identify the way of altering the recommendation system (RS), and at the same time, we propose a framework TRS for trusted RS for dynamic dataset based. This paper defines the process for fake profile classification, detection, and prevention from profile injection. In this framework we propose two phases, in first phase online social network (OSN) fake profile detection using supervised machine learning and in second phase we detect profile injection used for recommendations influence using outlier detection.
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2521 (1)
Open AccessTRS: Trusted recommendation system for fast changing dataset
Recommendations can be used in two ways, one for the mass audience and second for personalizing, in the mass audience like weather recommendation, stock recommendation, so this type of recommendation is based on past generic data. However, personalized recommendations are based on unique dataset of users, and at the time of changing location and demographic condition changing frequently, it becomes more complex. This type of recommendation entirely depends upon the dynamic profile of the user, and this dynamic behavior can alter by the third person for fake and biased suggestions. In this paper, we identify the way of altering the recommendation system (RS), and at the same time, we propose a framework TRS for trusted RS for dynamic dataset based. This paper defines the process for fake profile classification, detection, and prevention from profile injection. In this framework we propose two phases, in first phase online social network (OSN) fake profile detection using supervised machine learning and in second phase we detect profile injection used for recommendations influence using outlier detection.
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2521 (1)
Unraveling anti-cancer potential of phytotoxins using machine learning approaches
Cancer is known to be one of the central diseases which have the second highest mortality rate worldwide, and because of its ubiquity there is no doubt an unfulfilled requirement to develop and discover new anti cancer drugs. Nevertheless, the conventional way of discovering drugs, and developing them is a tedious, extensive and costly task, therefore in-silico techniques and algorithms comes to rescue at this point. In this article, potential of phytotoxins with anti cancer properties is unravelled with the aid of machine learning techniques more specifically implementation of tensor flow algorithm using Recurrent Neural Network (RNN). Various bioinformatics database and software have also been used at various steps for analysis. Toxic plants-phytotoxin (TPPT) database has been used to retrieve information on phytotoxins. Anti-cancer and non anti-cancer molecules have been sued for training and testing. Mordred web server has been used for attribute extraction from anti-cancer, non anti-cancer and phytotoxins dataset. The data has been split into training and testing data with training size of 70% and 30% respectively. Activation function used was softmax. Optimizer used at the time of compilation of the model was adam optimizer. Loss function used was sparse_categorical_crossentropy. The algorithm was run for 25 epochs. Different layers were used in this model like Input layer TFOpLambda, GRU, Flatten and Dense. The numbers of instances used were 1623 and those used for testing from total were 487 and training was performed with the rest of them i.e. 1136. On analyzing and testing the phytotoxins, more than 50 percent of the chemicals were known to exhibit anti cancer properties, and this showed that they can be used as anti neoplastic agents. The results of this research finding require further validation in laboratory. Initial screening can help in eliminating non anti-cancer molecules with great accuracy. Therefore can be a valuable aid in cancer fighting drug regime.
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2521 (1)
Open AccessUnraveling anti-cancer potential of phytotoxins using machine learning approaches
Cancer is known to be one of the central diseases which have the second highest mortality rate worldwide, and because of its ubiquity there is no doubt an unfulfilled requirement to develop and discover new anti cancer drugs. Nevertheless, the conventional way of discovering drugs, and developing them is a tedious, extensive and costly task, therefore in-silico techniques and algorithms comes to rescue at this point. In this article, potential of phytotoxins with anti cancer properties is unravelled with the aid of machine learning techniques more specifically implementation of tensor flow algorithm using Recurrent Neural Network (RNN). Various bioinformatics database and software have also been used at various steps for analysis. Toxic plants-phytotoxin (TPPT) database has been used to retrieve information on phytotoxins. Anti-cancer and non anti-cancer molecules have been sued for training and testing. Mordred web server has been used for attribute extraction from anti-cancer, non anti-cancer and phytotoxins dataset. The data has been split into training and testing data with training size of 70% and 30% respectively. Activation function used was softmax. Optimizer used at the time of compilation of the model was adam optimizer. Loss function used was sparse_categorical_crossentropy. The algorithm was run for 25 epochs. Different layers were used in this model like Input layer TFOpLambda, GRU, Flatten and Dense. The numbers of instances used were 1623 and those used for testing from total were 487 and training was performed with the rest of them i.e. 1136. On analyzing and testing the phytotoxins, more than 50 percent of the chemicals were known to exhibit anti cancer properties, and this showed that they can be used as anti neoplastic agents. The results of this research finding require further validation in laboratory. Initial screening can help in eliminating non anti-cancer molecules with great accuracy. Therefore can be a valuable aid in cancer fighting drug regime.
- Open Access
- 24 April 2022
- Published by MDPI AG in Electronics
- Vol. 11 (9) , 1353
Hybrid Spatiotemporal Contrastive Representation Learning for Content-Based Surgical Video Retrieval
In the medical field, due to their economic and clinical benefits, there is a growing interest in minimally invasive surgeries and microscopic surgeries. These types of surgeries are often recorded during operations, and these recordings have become a key resource for education, patient disease analysis, surgical error analysis, and surgical skill assessment. However, manual searching in this collection of long-term surgical videos is an extremely labor-intensive and long-term task, requiring an effective content-based video analysis system. In this regard, previous methods for surgical video retrieval are based on handcrafted features which do not represent the video effectively. On the other hand, deep learning-based solutions were found to be effective in both surgical image and video analysis, where CNN-, LSTM- and CNN-LSTM-based methods were proposed in most surgical video analysis tasks. In this paper, we propose a hybrid spatiotemporal embedding method to enhance spatiotemporal representations using an adaptive fusion layer on top of the LSTM and temporal causal convolutional modules. To learn surgical video representations, we propose exploring the supervised contrastive learning approach to leverage label information in addition to augmented versions. By validating our approach to a video retrieval task on two datasets, Surgical Actions 160 and Cataract-101, we significantly improve on previous results in terms of mean average precision, 30.012 ± 1.778 vs. 22.54 ± 1.557 for Surgical Actions 160 and 81.134 ± 1.28 vs. 33.18 ± 1.311 for Cataract-101. We also validate the proposed method’s suitability for surgical phase recognition task using the benchmark Cholec80 surgical dataset, where our approach outperforms (with 90.2% accuracy) the state of the art.
- 24 April 2022
- Published by MDPI AG in Electronics
- Vol. 11 (9) , 1353
Open AccessHybrid Spatiotemporal Contrastive Representation Learning for Content-Based Surgical Video Retrieval
In the medical field, due to their economic and clinical benefits, there is a growing interest in minimally invasive surgeries and microscopic surgeries. These types of surgeries are often recorded during operations, and these recordings have become a key resource for education, patient disease analysis, surgical error analysis, and surgical skill assessment. However, manual searching in this collection of long-term surgical videos is an extremely labor-intensive and long-term task, requiring an effective content-based video analysis system. In this regard, previous methods for surgical video retrieval are based on handcrafted features which do not represent the video effectively. On the other hand, deep learning-based solutions were found to be effective in both surgical image and video analysis, where CNN-, LSTM- and CNN-LSTM-based methods were proposed in most surgical video analysis tasks. In this paper, we propose a hybrid spatiotemporal embedding method to enhance spatiotemporal representations using an adaptive fusion layer on top of the LSTM and temporal causal convolutional modules. To learn surgical video representations, we propose exploring the supervised contrastive learning approach to leverage label information in addition to augmented versions. By validating our approach to a video retrieval task on two datasets, Surgical Actions 160 and Cataract-101, we significantly improve on previous results in terms of mean average precision, 30.012 ± 1.778 vs. 22.54 ± 1.557 for Surgical Actions 160 and 81.134 ± 1.28 vs. 33.18 ± 1.311 for Cataract-101. We also validate the proposed method’s suitability for surgical phase recognition task using the benchmark Cholec80 surgical dataset, where our approach outperforms (with 90.2% accuracy) the state of the art.
- 5 November 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 123 (2) , 1789-1807
Big Data Knowledge Discovery as a Service: Recent Trends and Challenges
- 5 November 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 123 (2) , 1789-1807
Big Data Knowledge Discovery as a Service: Recent Trends and Challenges
- 25 September 2020
- Published by Springer Nature in Wireless Personal Communications
- Vol. 116 (4) , 2819-2833
µBIGMSA-Microservice-Based Model for Big Data Knowledge Discovery: Thinking Beyond the Monoliths
- 25 September 2020
- Published by Springer Nature in Wireless Personal Communications
- Vol. 116 (4) , 2819-2833
µBIGMSA-Microservice-Based Model for Big Data Knowledge Discovery: Thinking Beyond the Monoliths
- 13 April 2016
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 8 (2) , 493-514
An approach to solve the target coverage problem by efficient deployment and scheduling of sensor nodes in WSN
- 13 April 2016
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 8 (2) , 493-514
An approach to solve the target coverage problem by efficient deployment and scheduling of sensor nodes in WSN
- Open Access
- 8 September 2014
- Published by MECS Publisher in International Journal of Computer Network and Information Security
- Vol. 6 (10) , 64-71
Trust Metric based Soft Security in Mobile Pervasive Environment
In the decentralized and highly dynamic environment like Mobile Pervasive Environments (MPE) trust and security measurement are two major challenging issues for community researchers. So far primarily many of architectural frameworks and models developed and being used. In the vision of pervasive computing where mobile applications are growing immensely with the potential of low cost, high performance, and user centric solutions. This paradigm is highly dynamic and heterogeneous and brings along trust and security challenges regarding vulnerabilities and threats due to inherent open connectivity. Despite advances in the technology, there is still a lack of methods to measure the security and level of trust and framework for the assessment and calculation of the degree of the trustworthiness. In this paper, we explore security and trust metrics concerns requirement and challenges to decide the trust computations metric parameters for a self-adaptive self-monitoring trust based security assurance in mobile pervasive environment. The objective is to identify the trust parameters while routing and determine the node behavior for soft security trust metric. In winding up, we put our efforts to set up security assurance model to deal with attacks and vulnerabilities requirements of system under exploration.
- 8 September 2014
- Published by MECS Publisher in International Journal of Computer Network and Information Security
- Vol. 6 (10) , 64-71
Open AccessTrust Metric based Soft Security in Mobile Pervasive Environment
In the decentralized and highly dynamic environment like Mobile Pervasive Environments (MPE) trust and security measurement are two major challenging issues for community researchers. So far primarily many of architectural frameworks and models developed and being used. In the vision of pervasive computing where mobile applications are growing immensely with the potential of low cost, high performance, and user centric solutions. This paradigm is highly dynamic and heterogeneous and brings along trust and security challenges regarding vulnerabilities and threats due to inherent open connectivity. Despite advances in the technology, there is still a lack of methods to measure the security and level of trust and framework for the assessment and calculation of the degree of the trustworthiness. In this paper, we explore security and trust metrics concerns requirement and challenges to decide the trust computations metric parameters for a self-adaptive self-monitoring trust based security assurance in mobile pervasive environment. The objective is to identify the trust parameters while routing and determine the node behavior for soft security trust metric. In winding up, we put our efforts to set up security assurance model to deal with attacks and vulnerabilities requirements of system under exploration.
- 1 June 2014
- Published by Springer Nature in CSI Transactions on ICT
- Vol. 2 (2) , 97-107
Cluster head selection by randomness with data recovery in WSN
- 1 June 2014
- Published by Springer Nature in CSI Transactions on ICT
- Vol. 2 (2) , 97-107
Cluster head selection by randomness with data recovery in WSN