MANDAR KRISHNARAO MOKASHI

@viit.ac.in

Assistant Professor, Department of Information Technology
Vishwakarma Institute of Information Technology, Pune

MANDAR KRISHNARAO MOKASHI


            Download Compact PDF  
https://researchid.co/mkmokashi

RESEARCH INTERESTS

Wireless Sensor Network
21

Scopus Publications

Scopus Publications

  • Efficient solving of partial differential equations with new numerical methods
    Ravi Yadav, Tanveer Ahmad Wani, Mandar K. Mokashi, Shirish Jaysing Navale, Egambergan Khudaynazarov, Swati V. Khidse
    Journal of Interdisciplinary Mathematics, 2026
    Partial Differential Equations (PDEs) play a very significant role in the life of scientists and engineers who wish to model complex physical processes. However, standard solutions to math have a hard time striking the correct balance of accuracy, stability, and the speed at which they can be of their tasks. In this research, it is proposed that new numeric techniques can be used to solve PDEs at a fast rate, higher convergence rates and reduced costs of computing. The new techniques utilise a better summability and flexible discretisation to provide a large variety of border cases and nonlinearities. Most computer experiments demonstrate that the proposed methods are faster and more precise compared to alternative approaches to the standard PDE problems. It is now possible to have fluid dynamics, heat movement, and electromagnetism models that are more precise. The study demonstrates that big computer systems may be able to use scalable application.
  • Fractional calculus applications in deep learning architectures
    Chandrakant Deelip Kokane, Anjali S. More, Harsha Jitendra Sarode, Mandar K. Mokashi, Sonali Prashant Bhoite, Mahesh Ashok Bhandari
    Journal of Interdisciplinary Mathematics, 2026
    The study investigates the application of the concept of the addition of a deep learning system to fractional calculus (FC) in order to accelerate and generalize the learning process. We propose a different form of neural network FGNN, Fractional Gradient Neural Network. In this network, the derivatives of fractional order are the gradient dynamics in control resulting in the optimisation being more fluid and the extraction of features being improved. The approach is a hybrid between the ease of use and memory capabilities of FC and the adaptation and data-driven capabilities of DL. The picture and signaling datasets experimental results indicate that these networks are more precise, more robust and less prone to overfitting as compared to standard networks. These findings indicate that FC is an excellent mathematical foundation to make next-generation deep learning models improved.
  • Deep Learning for Colorectal Cancer Screening Using Colonoscopy Images
    Mandar K. Mokashi, Ganesh Shelke, Manali M. Shah, Mahesh S. Shinde, Mahesh Bhandari, Dattatray G. Takale, Parikshit N. Mahalle, Bipin Sule
    Lecture Notes in Networks and Systems, 2026
  • VIRTUAL PERFORMANCES AND AI-DRIVEN AUDIENCE ANALYTICS
    Swarna Swetha Kolaventi, Amritpal Sidhu, Rashmi Manhas, Mandar K Mokashi, Aneesh Wunnava, Suhas Gupta
    Shodhkosh Journal of Visual and Performing Arts, 2025
    Entertainment business is going into computers and this has brought a change in the manner shows are produced, shown and enjoyed. The creation of the interactive platform and the greater global connectivity has propelled the growth of the virtual performances as a dynamic substitute of the traditional live performances. These are digital programs, theatre, and music shows. In this paper, the author will discuss how artificial intelligence (AI) is changing crowd data to make virtual performance worlds more interesting, personal, and creative decision worlds. Newer, more established ways of gauging an audience that relied on feedbacks of individuals and basic demographics is being rapidly phased out by intelligent systems that have the capacity of capturing and analysing real time data. Using machine learning, AI models can be used to guess the emotion of people, how many they are engaged, and what they like with high precision. In other words, the methods of computer vision and emotion monitoring can determine the attention and emotions and natural language processing (NLP) can assist in making what is being said on social networks and chats, polls and social networks more complicated. Through this addition, virtual performance platforms can modify things such as lighting, music or even the performance of the story in real-time to better communicate to the audiences. Moreover, the information created on the basis of artificial intelligence will allow the manufacturers and producers to improve the strategies of their performance, increase the efficiency of marketing, and build the more profound emotional connections. The case studies of the digital experiences where AI has been used to enhance the experience, provide an example of how such systems may change the experiences to be more flexible, data-driven, and engaging.
  • FinTech-Enabled Microfinance for Last-Mile Communities: A Pathway Toward SDG 1 and SDG 8
    Kireeti Gudipati, Gurpreet Kaur, Chuanyong Ji, Mandar K Mokashi, Sonali Prashant Bhoite, Piyush K. Ingole
    Enterprise Development and Microfinance, 2025
    Microfinance made possible through FinTech has become an effective catalyst of financial inclusion to last-mile communities and a plausible route to Sustainable Development Goal (SDG) 1 (No Poverty) and SDG 8 (Decent Work and Economic Growth). Through digital payment solutions, mobile banking, AI-based credit check, and data analytics, FinTech can minimize the structure, i.e., geographic isolation, the absence of formal credit history, and elevated transaction rates. This digitalization enhances the ability of the microfinance institutions to provide affordable, scalable and inclusive financial services, which contribute to livelihood creation and economic resiliency of locals. Although the world is making strides in financial inclusion, last-mile communities are still overrepresented in the informal financial framework. Microfinance models Traditional models of microfinance are inefficient in their operations, have limited reach, are characterized by high default risk, and rely on manual processes. These limitations limit access to credit, savings, and insurance products on a timely basis and therefore limit entrepreneurial activities and sustainable production of employment. The existing literature tends to separate FinTech innovation and the development of microfinance, with little conceptualization of the two at an SDG-based level of critique. Evidence of the linkage of FinTech-enabling models of microfinance with quantifiable results on poverty alleviation and decent work in underserved areas is lacking. The current paper suggests a hybrid FinTech-microfinance model that builds on mobile apps, digital identities, risk scoring based on artificial intelligence, and real-time tracking to improve its outreach and financial sustainability. It contributes to the fact that technological innovation was aligned with the SDG indicators, showed better outcomes in income stability, growth of micro-enterprises, job creation, and inclusive economic development, thus contributing to SDG 1 and SDG 8.
  • Distinguishing Reality from Fabrication: A Review of Deepfake Detection Methods and Comparative Analysis
    Abhijeet Cholke, Puja Cholke, Mandar Mokashi, Sheetal Aher, Shrikant Shinde, Arun Ghandat
    2025 International Conference on Future Technologies Icft 2025, 2025
  • Zero Trust Security Architecture in Cyber Forensics: An Analysis Through Log Management
    Pooja Ramesh Oza, Vijay S. Gulhane, Parth Sharma, Monica Lamba, Mandar Krishnarao Mokashi, Deepali Patil
    Smart Innovation Systems and Technologies, 2025
  • SIFT and SURF: A Comparative Analysis of Feature Extraction Methods for Image Matching
    D. Anandhasilambarasan, Bharat Bhushan, S Ganga, Romil Jain, Pooja Varma, Mandar K. Mokashi
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
    The capacity to appropriately suit pics is a vital challenge in pc vision programs. To do this, feature extraction techniques inclusive of SIFT (Scale Invariant feature transform) and SURF (Speeded-Up sturdy functions) may be used. In this newsletter, a comparative evaluation of those two methods is provided with appreciate to their overall performance in terms of speed, accuracy, robustness, memory necessities, and generality. consequences display that SIFT typically outperforms SURF in terms of accuracy, robustness, and reminiscence necessities, at the same time as SURF plays extensively higher in phrases of velocity and generality. As such, the selection of feature extraction approach will depend upon the unique assignment handy and the options of the user.
  • Retracted: Real-time Abnormal Event Detection in Surveillance Videos using Spatio-temporal Features (2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) DOI: 10.1109/ICCCNT61001.2024.10724891)
    Udita Goyal, K S Kiran, Mandar Krishnarao Mokashi, V.J. Vijayalakshmi, Sudhanshu Dev, V Janakiraman
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
    This paper provides a novel technique for detecting abnormal occasions in surveillance motion pictures the usage of patio-temporal functions. This method is based totally on a convolutional neural community (CNN) architecture that extracts patio-temporal features from surveillance films by means of making use of motion and look cues. First, the video frames are pre-processed to reduce noise and extract salient capabilities. Then, an improved movement functions analysis network is developed to discover the motion styles of gadgets in movies. Eventually, the developed CNN is used to classify the movies into specific classes depending at the extracted movement and look statistics, for that reason figuring out strange activities. The proposed system done better accuracy and faster processing while in comparison to different traditional strategies for real-time peculiar event detection in surveillance movies.
  • Proposed Statistical Framework for Signal Modulation and Processing
    Naveen Kumar Rajendran, Mandar Krishnarao Mokashi, K. Ranjith Singh, Abhishek Singla, P. Chandrakala, Samaksh Goyal
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
    This paper proposes a new statistical framework to facilitate sign modulation and processing. The proposed framework models every signal as a combination of different distributions and makes use of advanced statistical inference techniques to perceive the parameters of each issue. additionally, the framework takes into account the temporal correlations between person samples of the sign and uses superior gadget gaining knowledge of techniques to lessen interference from noise, allowing for the a hit transmission of alerts over noisy channels. The proposed framework is designed to lessen the computational value of signal processing and for that reason enhance the transmission price of signals. The overall performance of the proposed framework is evaluated through both simulated and actual-world experiments. Consequences show that the proposed framework considerably improves the sign-to-noise ratio of transmitted indicators and also reduces the computational cost of sign processing.
  • Retraction Notice: Proposed Statistical Framework for Signal Modulation and Processing (2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) DOI: 10.1109/ICCCNT61001.2024.10725739)
    Naveen Kumar Rajendran, Mandar Krishnarao Mokashi, K. Ranjith Singh, Abhishek Singla, P. Chandrakala, Samaksh Goyal
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
  • Retraction Notice: Real-time Abnormal Event Detection in Surveillance Videos using Spatio-temporal Features (2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) DOI: 10.1109/ICCCNT61001.2024.10724891)
    Udita Goyal, K S Kiran, Mandar Krishnarao Mokashi, V.J. Vijayalakshmi, Sudhanshu Dev, V Janakiraman
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
  • Retraction Notice: SIFT and SURF: A Comparative Analysis of Feature Extraction Methods for Image Matching (2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) DOI: 10.1109/ICCCNT61001.2024.10726049)
    D. Anandhasilambarasan, Bharat Bhushan, S Ganga, Romil Jain, Pooja Varma, Mandar K. Mokashi
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
  • Real-time Abnormal Event Detection in Surveillance Videos using Spatio-temporal Features
    Udita Goyal, K S Kiran, Mandar Krishnarao Mokashi, V.J. Vijayalakshmi, Sudhanshu Dev, V Janakiraman
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
  • Retracted: SIFT and SURF: A Comparative Analysis of Feature Extraction Methods for Image Matching (2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) DOI: 10.1109/ICCCNT61001.2024.10726049)
    D. Anandhasilambarasan, Bharat Bhushan, S Ganga, Romil Jain, Pooja Varma, Mandar K. Mokashi
    2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
  • Noma Based Full Duplex 6G Communication Systems
    Asha Sohal, Sandeep Sharma, Sweta Agarwal, N. V. Balaji, D. Bhanu, Mandar Krishnarao Mokashi
    Proceedings of International Conference on Contemporary Computing and Informatics Ic3i 2023, 2023
  • Harnessing AI to Achieve the Best Results from Data Science
    Valarmathy AS, Mandar K. Mokashi, Sunil Kumar Jakhar, Vaishali Singh, Asha KS, Ashmeet Kaur
    2023 3rd International Conference on Smart Generation Computing Communication and Networking Smart Gencon 2023, 2023
  • Investigating the Potential of Data Mining for Resource Optimization
    Dhiraj Kumar Singh, B Reddy, Sheril S, Mandar K. Mokashi, Akhilendra Pratap Singh, Sudha D
    Proceedings 2023 IEEE International Conference on Paradigm Shift in Information Technologies with Innovative Applications in Global Scenario Icpsitiags 2023, 2023
  • Click based inferring of user search goals using pseudo document
    Harshada P. Bhambure, Mandar Mokashi
    IEEE International Conference on Computer Communication and Control Ic4 2015, 2016
  • Real-Time Flash-Flood Monitoring, Alerting and Forecasting System using Data Mining and wireless sensor Network
    Sagar Sudhakar Mane, M.K. Mokashi
    2015 International Conference on Communication and Signal Processing Iccsp 2015, 2015
  • A survey on inferring user search goals using feedback session
    Harshada Bhambure, Mandar Mokashi
    2015 International Conference on Pervasive Computing Advance Communication Technology and Application for Society Icpc 2015, 2015