RMRahul Prakash Mirajkar

Rahul Prakash Mirajkar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Rahul Prakash Mirajkar has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Binary hybrid pathfinder algorithm for efficient feature selection in resource-constrained embedded systems

    by Rahul Mirajkar, Premanand Ghadekar, Vijay Dasharath Chougule, Renuka Bhandari, Hridaynath Khandagale, Mahavir A. Devmane, Mangesh Hajare, Kuldeep B. Vayadande

    Feature selection is critical for embedded machine learning systems where computational resources and memory are severely constrained. This paper presents the binary quadratically interpolated hybrid pathfinder algorithm (BQIHPFA), a novel metaheuristic optimization method designed for efficient feature subset selection in resource-limited classification tasks. BQIHPFA adapts the continuous QIHPFA to binary search spaces through sigmoid transfer functions and employs a hybrid two-group enhancement strategy combining pathfinder dynamics with salp swarm algorithm-inspired exploration. We evaluate BQIHPFA against three established binary optimization algorithms (binary particle swarm optimization (BPSO), binary grey wolf optimizer (BGWO), and binary whale optimization (BWO)) on three benchmark datasets with varying dimensionalities: Língua Brasileira de Sinais (Brazilian Sign Language) movement (90 features), Parkinson's disease detection (22 features), and Sonar Rock vs. Mine (60 features). Experimental results demonstrate that BQIHPFA achieves competitive classification accuracy (average 83.57%) with substantial feature reduction (average 64.1%) while executing 5.2 times faster than complex baselines and consuming minimal memory (peak: 45-58 MB). Ablation experiments demonstrate that every algorithmic part makes a 8-24% contribution to the total performance. BQIHPFA offers an easy-to-use, non-specific feature selection method to automated resource-constrained embedded classification systems, applicable to be deployed to low-power computing environments, and internet of things (IoT) edge systems.

  • Open Access

    Synaptic shield: fusion of ResNext–50 and long short-term memory for enhanaced deepfake detection

    by Amit Mishra, Prajwal Chinchmalatpure, Govinda B. Sambare, Viomesh Kumar Singh, Atul Gulabrao Pawar, Rahul Prakash Mirajkar, Priyanka K. Takalkar, Kuldeep Vayadande

    Recent developments in deepfakes have created much anxiety about the authenticity of any digital content and thus, calls for implementing detection mechanisms that will work accordingly. This paper uses Synaptic Shield, a innovative deep learning (DL) framework which is customized to detect alterations by deepfakes with high precision levels. It employs both convolution neural networks (CNNs) as well as modules for time feature extractions to test spatial and motion indicators from video data. High-level preprocessing pipelines in combination with confidence scoring mechanism help make Synaptic Shield adaptive toward manipulation techniques such as FaceSwap and DeepFake. The accuracy of our model surpasses other deepfake detection models with a high accuracy of 98.3%. The above results are based on exhaustive experimentation on standard datasets like FaceForensics++, DeepFake detection challenge (DFDC), and Celeb DeepFake (Celeb-DF). Synaptic Shield is shown to be the best with outstanding results that maintain a higher confidence score equivalent to its precision and reliability. Scalability in having the capacity to accommodate various manipulation techniques and levels of video quality indicates robustness in offering an effective method toward ensuring integrity in digital media. The work is an important move forward in addressing the problems created by DeepFake technologies.

  • Anomaly Detection and Downtime Prediction in Server Logs Using Hybrid Machine Learning

    by Swati Gandhi, Sharda Yashwant Salunkhe, Dipa Dattatray Dharmadhikari, Ajit Patil, Yogesh Uttamrao Bodhe, Sonali Bhoite, Kuldeep Vayadande, Rahul Mirajkar

    In this paper, we propose a forensic framework that is able to forecast downtime errors at the server level based on anomaly detection to system logs, converted into features according to machine learning (ML) algorithms. We used hybrid of ML model, based on Isolation Forest and One-Class SVM algorithms for encoding normal patterns of behaviour systems to annotate deviations from the norm. Detailed log file parsing of any log files of interest, feature extraction, and weighted scoring in combining anomalies were among the actions included in my approach. The hybrid model achieved 99.54% accuracy on real-world live-server log validation data and 95.08% in recall, consistent with individual models, making it a not-so-common occurrence. It identified high-risk periods and scheduled the preventive maintenance. This shows the practical way to bring into predictive IT operations log analysis using ML techniques, which in turn answers a very direct critical need for improved reliability of infrastructure. Future work will be pointed toward this dimension in view of limitations related to dataset scope and sensitivity of models to changes in log format.

  • A Comprehensive Review on Comparative Analysis of Web Services

    by Kuldeep Vayadande, Varsha Jujare, Maya Yogesh Patil, Ajit R. Patil, Gitanjali Bhimrao Yadav, Sachin Balawant Takmare, Rahul Prakash Mirajkar

    This study analyzes web services' functionalities and performance, focusing on mobile platforms and Web3 applications. Web services enable communication between applications but pose challenges due to different types (RESTful, SOAP), data formats, and protocols. Choosing the right service requires considering scalability, security, and interoperability. The study provides a comparative analysis of web service types, evaluating their strengths, limitations, and suitability for mobile platforms. It highlights challenges in selecting services and offers insights to guide decision-making. By assessing performance metrics, this research aids in optimizing web-based applications, enhancing decision-making for Web3 and mobile development.