Prakash Kumar Jha

@soa.ac.in

Assistant Professor, Department of Computer Science and Engineering
Siksha O Anusandhan University

Prakash Kumar Jha


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

EDUCATION

M.Tech in Computer Science and Engineering from National Institute of Technology Patna
B.Tech in Computer Science and Engineering from MAKAUT,WB

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Science, Computer Networks and Communications, Management Science and Operations Research
7

Scopus Publications

2

Scholar Citations

1

Scholar h-index

Scopus Publications

  • GSM-Driven IoT-Based Smart Drip Monitoring System to Improve Patient Care*
    Anmol Rath, Prakash Kumar Jha, Shatarupa Dash
    2nd International Conference on Cognitive Green and Ubiquitous Computing IC Cgu 2025, 2025
  • A Combination of Ensemble with MCDM Approach for the Prediction of Alzheimer’s Disease through Audio Data
    Pradip Dhal, Prakash Kumar Jha, Ishan Ayus
    Machine Learning for Neurodegenerative Disorders Advancements and Applications, 2025
    The most prevalent kind of dementia is Alzheimer's disease (AD). It is a gradual condition that starts with mild memory loss and may develop into a loss of speech and capacity to react to surroundings. The brain regions responsible for cognition, memory, and language are affected by AD. The development of more sophisticated treatments for AD depends on early identification. Machine learning (ML) is a branch of artificial intelligence (AI) that helps computers learns from large and complex data sets by applying various probabilistic and optimization techniques. Researchers typically use ML to diagnose AD in its early stages. This work presented a hybrid framework for the early prediction of AD. First, our voting-based feature selection (FS) approach extracts the most essential features. After applying the pre-processing first, this framework uses the SVM-SMOTE oversampling method for handling the class imbalance issue. Then, the voting-based FS approach extracts the most essential features. Finally, a technique for order preference by similarity to ideal solution (TOPSIS)-based classifier has been applied to these optimal features. TOPSIS is a multi-criteria decision-making (MCDM) approach. To assess the resilience and predict how the model will perform with unknown data, the suggested framework has been tested using various classification techniques. The experimental result shows that this approach is superior in boosting classification accuracy.
  • Smart Iot-Enabled Saline Level Monitoring and Alert System
    Anmol Rath, Prakash Kumar Jha, Shatarupa Dash, Chetana Priyadarshini Das
    International Conference on Artificial Intelligence and Emerging Technologies Icaiet 2025, 2025
    The Internet of Things (IoT) has become a gamechanging technology that allows for real-time data collecting, analysis and automation in a number of fields, including smart cities, transportation, agriculture and especially healthcare. With its solutions for automated medication distribution, remote monitoring, and real-time health warnings, IoT has emerged as a game-changing instrument in the medical industry. The quality of treatment, hospital productivity, and patient safety are all greatly increasing as a result of its adoption into healthcare systems. The manual monitoring of intravenous (IV) saline bottles is a crucial issue that is often overlooked in hospitals. When an empty saline container fails to be noticed right away in a busy hospital setting, air may enter the patient's circulation and cause serious problems or even death. These occurrences demonstrate the necessity of an automated, dependable monitoring system to guarantee prompt alarms and lessen the workload for medical personnel. Using a Raspberry Pi Pico W microcontroller for processing, a Bluetooth module for wireless communication, and a HX711 load sensor to track the weight of saline bottles, this study introduces an Internet of Things-based solution. The system checks sensor latency to guarantee timely message delivery in addition to sending alarms when the saline level is low. Early findings show that the system is a dependable and effective tool in hospital settings, with little latency and excellent weight detection accuracy.
  • Enhancing Network Security through Real-Time Anomaly Detection in Network Traffic
    Prakash Kumar Jha, Saket Maheshwari, Rahul Kumar, Bipul Kumar
    Icoicc 2025 3rd International Conference on Intelligent and Cloud Computing, 2025
    Securing network infrastructures is essential in today's digital environment to guard against more complex cyberattacks. Network traffic anomaly detection is essential for spotting malicious activity like port scanning, Distributed Denial of Service (DDoS) attacks, and illegal access. The goal of this study is to improve network security by classifying network traffic as either "attack" or "not attack" using a variety of machine learning models. We use XGBoost, Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Logistic Regression to categorize network traffic data according to whether it exhibits dangerous or benign behavior. The study investigates how well these models detect attacks with high accuracy while retaining the ability to classify in real time. Our findings show that SVM and XGBoost perform better in terms of accuracy and resilience, especially when dealing with complicated, high-dimensional data. We also assess the trade-offs between computing efficiency and model complexity, highlighting the significance of real-time detection in averting possible dangers. The difficulties of managing unbalanced data and the requirement for scalable, flexible solutions to counter new threats are also covered in the paper. Lastly, we make recommendations for future lines of inquiry to enhance anomaly detection systems and further hone the categorization procedure.
  • Optimizing Resource Scheduling for Enhanced Efficiency in Cloud Computing
    Prayansh Mishra, Prakash Kumar Jha, Ipsita Sahoo, Bhaskar Chakraborty, Omkar Pratik Mishra
    Icoicc 2025 3rd International Conference on Intelligent and Cloud Computing, 2025
    Cloud computing is a significant advancement in computing technologies as it operates on a pay-as-you-use model. It provides resources like storage, servers, etc., through a virtual platform which makes the remote access of distributed resources easy and efficient. This enhances flexibility, scalability, and cost-effectiveness. Clients provide tasks to the cloud service providers through some software which follow the resource scheduling. Resource scheduling is a critical process that shows the overall performance of a system. It is the process to generate the schedule appropriately that decides which task will be mapped to which resource. Efficient resource scheduling ensures optimal utilization of a resource. Sometimes many tasks are provided to the same set resources repeatedly and the rest of the resources remain idle. This creates an uneven distribution of workload which may have impact on the efficiency of the system. For this reason, load balancing is required. It is an important component of cloud computing which is designed to distribute workloads and computing resources efficiently across multiple servers or VMs. Several resource scheduling algorithms have been proposed over the years addressing multiple factors and challenges. However, many of these could not adequately integrate several factors of resource scheduling and load balancing for an overall optimal performance. This project aims to address those problems and enhance efficiency in resource scheduling through a simulation-based evaluation. The project proposes an algorithm which considers task length and task priorities simultaneously and then finds their credit. Since load balancing is equally important as resource scheduling, the proposed method integrates credit-based task allocation with a load balancing strategy to improve Makespan and system performance. This maintains an even distribution of workload.
  • A Comparative Analysis of Regression Algorithms in Lung Cancer Prediction
    Prakash Kumar Jha, Avni Garg, Ramachandran T, Pavan Chaudhary, Jitha Janardhanan, Prabhjot Kaur
    2025 International Conference on Automation and Computation Autocom 2025, 2025
    Results of All Cancers The mortality of lung cancer is highest among both international and domestic patients. Hence, early detection and accurate prediction are essential for treatment-affecting diagnoses to improve patient outcomes. Throughout the evolution of machine learning technologies, numerous regression algorithms have been introduced to predict lung cancer using clinical data or imaging datasets. In this work, we perform a comparison analysis among linear regression, logistic regression, and all commonly used decision-tree-based methods, including Decision Trees (CART), Random Forests, and Support Vector Machines for predicting lung cancer diagnosis and prognosis. We use patient data from two hospitals, including age, smoking information, lung function tests, and imaging results, to perform the analysis. Our results suggest that random forest and support vector machines generally offer the best accuracy, sensitivity, and specificity in diagnosis and prognosis prediction. Decision tree algorithms also do but depend heavily on the feature set chosen. In addition, we show improved performance in predicting placebo response by combining different data sources. Our study demonstrates that selecting regression algorithms are critical in lung cancer prediction and that combining multiple data types can improve accuracy. Realizing this potential requires accurate prediction algorithms, which are not currently available; such information can lead to better-developed forecast models and improved lung cancer patients' diagnoses and outcomes.
  • Stacked Layer Based Deep Learning Approach for Fake News Classification
    Pradip Dhal, Deb Biswas, Pragya, Jayshree Patra, Manima Asish Mishra, Prakash Kumar Jha
    2023 7th International Conference on Computing Communication Control and Automation Iccubea 2023, 2023
    The growth of the global population”s social., economic., and emotional spheres is impacted by fake news and misleading information. Several classification systems are available to help identify fake news., but they needed to be quicker., more secure., and as trustworthy as was required at the time. Since manually confirming the integrity of news is difficult and expensive., researchers have shown much interest in this area. Different techniques for spotting fake news., including sentiment-based., image-based., content-based., social context-based., and hybrid context-based categorization., were examined. In this paper., we have developed a lightweight Deep Learning (DL) network for Fake News Classification (FNC)., a sentiment-based classification. The deep network consists of a stacked layer based feature extraction approach. Here stack layer first extracts the global features by the network of Bidirectional-Long Short Term Memory (Bi-LSTM) and Bidirectional-Gated Recurrent Unit (Bi-GRU). The results of the global features we have extracted the local features by the Convolutional Neural Network (CNN). The experiment findings show that the suggested lightweight DL network is more successful in producing superior results than the other methods.

RECENT SCHOLAR PUBLICATIONS

  • GSM-Driven IoT-Based Smart Drip Monitoring System to Improve Patient Care
    A Rath, PK Jha, S Dash
    2025 International Conference on Cognitive, Green and Ubiquitous Computing … , 2025
    2025
  • Smart Iot-Enabled Saline Level Monitoring and Alert System
    A Rath, PK Jha, S Dash, CP Das
    2025 International Conference on Artificial intelligence and Emerging … , 2025
    2025
  • Optimizing Resource Scheduling for Enhanced Efficiency in Cloud Computing
    P Mishra, PK Jha, I Sahoo, B Chakraborty, OP Mishra
    2025 International Conference on Intelligent and Cloud Computing (ICoICC), 1-6 , 2025
    2025
  • Enhancing Network Security through Real-Time Anomaly Detection in Network Traffic
    PK Jha, S Maheshwari, R Kumar, B Kumar
    2025 International Conference on Intelligent and Cloud Computing (ICoICC), 1-6 , 2025
    2025
  • A Combination of Ensemble with MCDM Approach for the Prediction of Alzheimer's Disease through Audio Data
    P Dhal, PK Jha, I Ayus
    Machine Learning for Neurodegenerative Disorders: Advancements and … , 2025
    2025
  • A Comparative Analysis of Regression Algorithms in Lung Cancer Prediction
    PK Jha, A Garg, P Chaudhary, J Janardhanan, P Kaur
    2025 International Conference on Automation and Computation (AUTOCOM), 905-910 , 2025
    2025
  • BINARY PSO BASED OPTIMIZATION APPROACH FOR THE IMAGE CLASSIFICATION OF MECHANICAL TOOLS
    P Dhal, PK Jha, P Kumar, B Pradhan
    Futuristic Trends in Mechanical Engineering Volume 3 Book 7 3, 1-10 , 2024
    2024
  • Stacked layer based deep learning approach for fake news classification
    P Dhal, D Biswas, J Patra, MA Mishra, PK Jha
    2023 7th International Conference On Computing, Communication, Control And … , 2023
    2023
    Citations: 2

MOST CITED SCHOLAR PUBLICATIONS

  • Stacked layer based deep learning approach for fake news classification
    P Dhal, D Biswas, J Patra, MA Mishra, PK Jha
    2023 7th International Conference On Computing, Communication, Control And … , 2023
    2023
    Citations: 2
  • GSM-Driven IoT-Based Smart Drip Monitoring System to Improve Patient Care
    A Rath, PK Jha, S Dash
    2025 International Conference on Cognitive, Green and Ubiquitous Computing … , 2025
    2025
  • Smart Iot-Enabled Saline Level Monitoring and Alert System
    A Rath, PK Jha, S Dash, CP Das
    2025 International Conference on Artificial intelligence and Emerging … , 2025
    2025
  • Optimizing Resource Scheduling for Enhanced Efficiency in Cloud Computing
    P Mishra, PK Jha, I Sahoo, B Chakraborty, OP Mishra
    2025 International Conference on Intelligent and Cloud Computing (ICoICC), 1-6 , 2025
    2025
  • Enhancing Network Security through Real-Time Anomaly Detection in Network Traffic
    PK Jha, S Maheshwari, R Kumar, B Kumar
    2025 International Conference on Intelligent and Cloud Computing (ICoICC), 1-6 , 2025
    2025
  • A Combination of Ensemble with MCDM Approach for the Prediction of Alzheimer's Disease through Audio Data
    P Dhal, PK Jha, I Ayus
    Machine Learning for Neurodegenerative Disorders: Advancements and … , 2025
    2025
  • A Comparative Analysis of Regression Algorithms in Lung Cancer Prediction
    PK Jha, A Garg, P Chaudhary, J Janardhanan, P Kaur
    2025 International Conference on Automation and Computation (AUTOCOM), 905-910 , 2025
    2025
  • BINARY PSO BASED OPTIMIZATION APPROACH FOR THE IMAGE CLASSIFICATION OF MECHANICAL TOOLS
    P Dhal, PK Jha, P Kumar, B Pradhan
    Futuristic Trends in Mechanical Engineering Volume 3 Book 7 3, 1-10 , 2024
    2024