Mamta Bansal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mamta Bansal has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Rashi Rastogi
Department of Computer Application, Sir Chhotu Ram Institute of Engineering Technology (SCRIET), CCSU, Meerut, India
2 shared publications - Ajay Rana
Shobhit Institute of Engineering and Technology
1 shared publication Anamika Rana
Maharaja Surajmal Institute
1 shared publication- Anubhav Sharma
Department of Computer Science & Engineering Shobhit University Meerut India
1 shared publication - Avinav Pathak
Department of Computer Science, Shobhit Institute of Engineering & Technology (Deemed to-be University), Meerut, India
1 shared publication Birendra Kumar Sharma
Ajay Kumar Garg Engineering College- MCA
1 shared publication
Latest Publications
- Open Access
- 2 February 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (1) , 20064-20070
Effective Diabetes Prediction using an IoT-based Integrated Ensemble Machine Learning Framework
Diabetes, a prevalent chronic disease, affects a significant global population. Identifying and being aware of key variables promptly may substantially enhance results for both patients and public health efforts. Systematic methods such as monitoring diabetic patients allow the collection of extensive data from diabetic patients. When it comes to keeping track of a patient's health, IoT sensors, such as those used in diabetic patient monitoring systems, are invaluable. Blood glucose levels, body temperature, and location of a diabetic patient can be tracked and recorded through a monitoring device. In addition to monitoring patients, these data can be classified using Machine Learning (ML) methods. This study applies three ML models to three different diabetes datasets and analyzes their performance. According to the results, the fine-tuned random forest model achieved higher accuracy, i.e., 89%, 90%, and 99%.
- 2 February 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (1) , 20064-20070
Open AccessEffective Diabetes Prediction using an IoT-based Integrated Ensemble Machine Learning Framework
Diabetes, a prevalent chronic disease, affects a significant global population. Identifying and being aware of key variables promptly may substantially enhance results for both patients and public health efforts. Systematic methods such as monitoring diabetic patients allow the collection of extensive data from diabetic patients. When it comes to keeping track of a patient's health, IoT sensors, such as those used in diabetic patient monitoring systems, are invaluable. Blood glucose levels, body temperature, and location of a diabetic patient can be tracked and recorded through a monitoring device. In addition to monitoring patients, these data can be classified using Machine Learning (ML) methods. This study applies three ML models to three different diabetes datasets and analyzes their performance. According to the results, the fine-tuned random forest model achieved higher accuracy, i.e., 89%, 90%, and 99%.
- Open Access
- 18 October 2024
- Published by Dewspublication
- Vol. 20 (S13) , 382-391
Design Of A Block Chain Based Model For Secure And Effective Pharmaceutical Supply Chain
The manufacture and allocation of forged drugs is an urgent and increasingly critical worldwide issue, especially in pandemics. The imperfect Supply and Block Chain system in the pharmaceutical industry is one of the reasons for drug counterfeiting. Drugs ownership changes from manufacturers to wholesaler, distributor, and then pharmacist before it reaches the customer thus making it difficult to keep track of it. In this paper, we have to Design a Block chain based Model for Secure and Effective Pharmaceutical Supply Chain. Finally, the study mainly focuses on increasing the safety of pharmaceutical products and reducing the manual operation of the supply chain with the most efficient architecture. This initiative will use block chain technology to address issues and inefficiencies in the pharmaceutical supply chain. The traditional pharmaceutical supply chain is vulnerable to problems such as counterfeiting, data inconsistencies, and a lack of transparency. The proposed solution uses a decentralized and transparent block chain technology to improve the entire traceability, security, and efficiency of the supply chain.
- 18 October 2024
- Published by Dewspublication
- Vol. 20 (S13) , 382-391
Open AccessDesign Of A Block Chain Based Model For Secure And Effective Pharmaceutical Supply Chain
The manufacture and allocation of forged drugs is an urgent and increasingly critical worldwide issue, especially in pandemics. The imperfect Supply and Block Chain system in the pharmaceutical industry is one of the reasons for drug counterfeiting. Drugs ownership changes from manufacturers to wholesaler, distributor, and then pharmacist before it reaches the customer thus making it difficult to keep track of it. In this paper, we have to Design a Block chain based Model for Secure and Effective Pharmaceutical Supply Chain. Finally, the study mainly focuses on increasing the safety of pharmaceutical products and reducing the manual operation of the supply chain with the most efficient architecture. This initiative will use block chain technology to address issues and inefficiencies in the pharmaceutical supply chain. The traditional pharmaceutical supply chain is vulnerable to problems such as counterfeiting, data inconsistencies, and a lack of transparency. The proposed solution uses a decentralized and transparent block chain technology to improve the entire traceability, security, and efficiency of the supply chain.
- Top Cited PapersOpen Access
- 13 March 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 59 (1) , 238
Metaheuristic Algorithms for Optimization: A Brief Review
In the area of optimization, metaheuristic algorithms have attracted a lot of interest. For many centuries, human beings have utilized metaheuristic algorithms as a problem-solving approach. The application of these methods to combinatorial optimization problems has rapidly become a growing area of research, incorporating principles of natural selection, evolution, and problem-solving strategies. While conventional software engineering methods may not always be effective in resolving software issues, mathematical optimization using metaheuristics can offer a solution. As a result, metaheuristics have become an increasingly important part of modern optimization, with a large number of algorithms emerging over the last two decades. The purpose of this study is to present a quick overview of these algorithms so that researchers may choose and use the best metaheuristic method for their optimization issues. The key components and concepts of each type of algorithm have been discussed, highlighting their benefits and limitations. This paper aims to provide a comprehensive review of these algorithms, including evolution-based methods, swarm intelligence-based, physics-based, human-related, and hybrid metaheuristics by highlighting their key components and concepts and comparing and contrasting their similarities and differences. This work also addressed some of the difficulties associated with metaheuristic algorithms. Some practical uses of these metaheuristic algorithms were addressed.
- 13 March 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 59 (1) , 238
Top Cited PapersOpen AccessMetaheuristic Algorithms for Optimization: A Brief Review
In the area of optimization, metaheuristic algorithms have attracted a lot of interest. For many centuries, human beings have utilized metaheuristic algorithms as a problem-solving approach. The application of these methods to combinatorial optimization problems has rapidly become a growing area of research, incorporating principles of natural selection, evolution, and problem-solving strategies. While conventional software engineering methods may not always be effective in resolving software issues, mathematical optimization using metaheuristics can offer a solution. As a result, metaheuristics have become an increasingly important part of modern optimization, with a large number of algorithms emerging over the last two decades. The purpose of this study is to present a quick overview of these algorithms so that researchers may choose and use the best metaheuristic method for their optimization issues. The key components and concepts of each type of algorithm have been discussed, highlighting their benefits and limitations. This paper aims to provide a comprehensive review of these algorithms, including evolution-based methods, swarm intelligence-based, physics-based, human-related, and hybrid metaheuristics by highlighting their key components and concepts and comparing and contrasting their similarities and differences. This work also addressed some of the difficulties associated with metaheuristic algorithms. Some practical uses of these metaheuristic algorithms were addressed.
- 18 August 2023
- Published by Springer Nature
- p. 445-461
Assessment on Different IoT-Based Healthcare Services and Applications
- 18 August 2023
- Published by Springer Nature
- p. 445-461
Assessment on Different IoT-Based Healthcare Services and Applications
- 3 June 2022
- Published by Springer Nature
- p. 473-480
A Relative Study on Machine Learning-Based Algorithm Used to Develop Recommender Systems
- 3 June 2022
- Published by Springer Nature
- p. 473-480
A Relative Study on Machine Learning-Based Algorithm Used to Develop Recommender Systems
- 3 June 2022
- Published by Springer Nature
- p. 481-491
Secure Watermarking to Protect Colour Images on Social Media from Misuse
- 3 June 2022
- Published by Springer Nature
- p. 481-491
Secure Watermarking to Protect Colour Images on Social Media from Misuse
- 3 June 2022
- Published by Springer Nature
- p. 525-532
Software Testing and Test Case Optimization: Concepts and Trends
- 3 June 2022
- Published by Springer Nature
- p. 525-532
Software Testing and Test Case Optimization: Concepts and Trends
- 3 June 2022
- Published by Springer Nature
- p. 507-524
Analysis of Current Recommendation Techniques and Evaluation Metrics to Design an Improved Book Recommendation System
- 3 June 2022
- Published by Springer Nature
- p. 507-524
Analysis of Current Recommendation Techniques and Evaluation Metrics to Design an Improved Book Recommendation System
- 3 June 2022
- Published by Springer Nature
- p. 501-505
Breast Cancer Diagnosis with the Support of Machine Learning Algorithms
- 3 June 2022
- Published by Springer Nature
- p. 501-505
Breast Cancer Diagnosis with the Support of Machine Learning Algorithms
- 1 January 2021
- Published by Diva Enterprises Private Limited in ASIAN JOURNAL OF MULTIDIMENSIONAL RESEARCH
- Vol. 10 (10) , 288-294
An overview of deep learning in healthcare
In recent years, a rising expectation has been built around the analysis of huge quantities of data often accessible in companies, which has been examined by academics and effectively utilized by business. Big Data and Deep Learning are two of the most often used phrases in scientific circles nowadays. In this two-part study, we want to shed some light on the present status of these two distinct but connected areas of Data Science in order to better understand the current state and future development of healthcare. We begin by providing a basic review of the technical components of Big Data technologies, as well as an overview of the elements of Deep Learning methods as described in scientific literature. Then we focus on the application areas that may be claimed to have produced meaningful real-world success stories, with examples ranging from major technology firms to financial institutions, among others. The academic effort that has gone into bringing these technologies to the healthcare sector is then summarized and analyzed in two ways: first, the landscape of application examples is globally scrutinized according to the varying nature of medical data, including data forms in electronic health records, medical time signals, and medical images; and second, a specific application example is scrutinized according to the varying nature of medical data. The freely accessible MIMIC dataset includes a collection of toy application examples designed to assist novices get started with some principled, simple, and structured content and code. Current and future obstacles to the employment of both sets of methods in our future healthcare are discussed critically.
- 1 January 2021
- Published by Diva Enterprises Private Limited in ASIAN JOURNAL OF MULTIDIMENSIONAL RESEARCH
- Vol. 10 (10) , 288-294
An overview of deep learning in healthcare
In recent years, a rising expectation has been built around the analysis of huge quantities of data often accessible in companies, which has been examined by academics and effectively utilized by business. Big Data and Deep Learning are two of the most often used phrases in scientific circles nowadays. In this two-part study, we want to shed some light on the present status of these two distinct but connected areas of Data Science in order to better understand the current state and future development of healthcare. We begin by providing a basic review of the technical components of Big Data technologies, as well as an overview of the elements of Deep Learning methods as described in scientific literature. Then we focus on the application areas that may be claimed to have produced meaningful real-world success stories, with examples ranging from major technology firms to financial institutions, among others. The academic effort that has gone into bringing these technologies to the healthcare sector is then summarized and analyzed in two ways: first, the landscape of application examples is globally scrutinized according to the varying nature of medical data, including data forms in electronic health records, medical time signals, and medical images; and second, a specific application example is scrutinized according to the varying nature of medical data. The freely accessible MIMIC dataset includes a collection of toy application examples designed to assist novices get started with some principled, simple, and structured content and code. Current and future obstacles to the employment of both sets of methods in our future healthcare are discussed critically.