panduranga vital terlapu - Profile on Academia.edu (original) (raw)

Papers by panduranga vital terlapu

International Journal of Computing and Digital Systems, 2025

Autonomous Autonomous vehicles mark a revolutionary change in transportation, promising improved ... more Autonomous Autonomous vehicles mark a revolutionary change in transportation, promising improved safety, accessibility and convenience. LiDAR (Light Detection and Ranging) sensors are critical for environmental perception in autonomous vehicles, yet generating precise, real-time models of dynamic driving environments remains challenging. It is used to scan the surroundings of these vehicles. To address this, we propose enhanced LiDAR data processing techniques that integrate the Iterative Closest Point (ICP) algorithm for spatial alignment of point clouds and DBSCAN clustering for robust object segmentation in complex scenarios. To address this, we propose enhanced LiDAR data processing techniques that integrate the Iterative Closest Point (ICP) algorithm for spatial alignment of point clouds and DBSCAN clustering for robust object segmentation in complex scenarios. We explore the Iterative Closest Point (ICP) technique on LiDAR point-cloud data to iteratively align and refine overlapping point clouds, improving surface representation. Additionally, we employ DBSCAN clustering on point cloud data to detect and group spatially dense regions, enhancing object detection in complex environments. We also investigated the combined use of DBSCAN and bounding-box methods, providing a robust solution for object detection. This clustering approach improves our object segmentation process, allowing for a more accurate identification of objects within the LiDAR data. Our research findings will help guide future studies on LiDAR data processing techniques. Our proposed method successfully balances the quality, file size, and accuracy of preprocessing, allowing for further value adjustments through rigorous testing and evaluation of existing datasets.

International Journal of Computing and Digital Systems, 2025

Embedded systems and smartphones are vital in real-time applications, inspiring our interaction w... more Embedded systems and smartphones are vital in real-time applications, inspiring our interaction with technology. Smartphones possess various sensors like accelerometers, gyroscopes, and magnetometers. Deep Learning (DL) models enhance the capabilities of sensors, enabling them to perform real-time analysis and decision-making with accuracy and speed. This study demonstrates an intelligent system that detects smartphone movements using deep learning (DL) techniques such as convolutional neural networks (CNN) and stacked autoencoders(SAEs). The dataset has six smartphone movements, with 921 samples split into 695 for training and 226 for testing. The best training performance was achieved by Auto-Encoder 1 and Auto-Encoder 2. The SAEs had high classification accuracy (CA) and AUC values of 0.996 and 1.0, respectively. Similarly, CNN performed well with CA and AUC values of 0.991 and 0.998. These results show that CNN and SAEs are effective in identifying smartphone movements. The findings help improve smartphone apps and understand how well they can identify movement. The study indicates that CNN and SAE are influential in accurately identifying smartphone movements. Future research can improve motion detection by integrating more sensor data and advanced models. Using advanced deep learning architectures like RNNs or transformers can enhance the understanding and accuracy of predicting smartphone movements.

IEEE ACCESS, 2024

Effort estimations are a crucial aspect of software development. The tasks should be completed be... more Effort estimations are a crucial aspect of software development. The tasks should be completed before the start of any software project. Accurate estimations increase the chances of project success, and inaccurate information can lead to severe issues. This study systematically reviewed the literature on effort-estimating models from 2015-2024, identifying 69 relevant studies from various publications to compile information on various software work estimation models. This review aims to analyze the models proposed in the literature and their classification, the metrics used for accuracy measurement, the leading model that has been chiefly applied for effort estimation, and the benchmark datasets available. The study utilized 542 relevant articles on software development, cost, effort, prediction, estimation, and modelling techniques in the search strategy. After 194 selections, the authors chose 69 articles to understand ML applications in SEE comprehensively. The researchers used a scoring system to assess each study's responses (from 0 to 5 points) to their research questions. This helped them identify credible studies with higher scores for a comprehensive review aligned with its objectives. The data extraction process identified 91% (63) of 69 studies as either highly or somewhat relevant, demonstrating a successful search strategy for analysis. The literature review on SEE indicates a growing preference for ML-based models in 59% of selected studies. 17% of the studies chosen favor hybrid models to overcome software development challenges. We qualitatively analyzed all the literature on software effort estimation using expert judgment, formal estimation techniques, ML-based techniques, and hybrid techniques. We discovered that researchers have frequently used ML-based models to estimate software effort and are currently in the lead. This study also explores the application of feature importance and selection in machine learning models for Software Effort Estimation (SEE) using popular algorithms like support Vector Machine (SVM), AdaBoost (AB), Gradient Boost (GB), and Random Forest (RF) with six benchmark datasets like CHINA, COCOMO-NASA2, COCOMO, COCOMO81, DESHARNAIS, and KITCHENHAM. We analyze the dataset descriptions and feature importance of the dataset analysis using ML models for choosing crucial play attributes in SEE.

Institute of Advanced Engineering and Science (IAES), 2024

Wireless sensor networks (WSNs) are of significant importance in many applications; nevertheless,... more Wireless sensor networks (WSNs) are of significant importance in many applications; nevertheless, their operational efficiency and longevity might be impeded by energy limitations. The low energy adaptive clustering hierarchy (LEACH) protocol has been specifically developed with the objective of achieving energy consumption equilibrium and regularly rotating cluster heads (CHs). This study presents a novel technique, namely the hierarchical fuzzy logic controller (HFLC), which is integrated with the LEACH protocol to enhance the process of CH selection and effectively prolong the network's operational lifespan. The HFLC system employs fuzzy logic as a means to address the challenges posed by uncertainty and imprecision. It assesses many aspects, including residual energy, node proximity, and network density, in order to make informed decisions. The combination of HFLC with LEACH demonstrates superior performance compared to the conventional LEACH protocol in terms of energy efficiency, stability, and network durability. This study emphasizes the potential of intelligent and adaptive mechanisms in improving the performance of WSNs by improving the survivability of nodes by reducing the energy consumption of the nodes during the communication of network process. It also paves the way for future research that integrates soft computing approaches into network protocols.

Drinkers Voice Recognition Intelligent System: An Ensemble Stacking Machine Learning Approach

Annals of data science, Jul 7, 2024

Indian journal of science and technology, May 14, 2024

Objectives: The research aims to enhance breast cancer detection accuracy and effectiveness using... more Objectives: The research aims to enhance breast cancer detection accuracy and effectiveness using deep transfer learning and pre-trained neural networks. It analyses breast ultrasound images and identifies important characteristics using pre-trained networks. The goal is to create a more efficient and accurate automated system for breast cancer detection. Methods: The study uses breast ultrasound cancer image data from the Kaggle Data Repository to extract informative features, identify cancer-related characteristics, and classify them into benign, malignant, and normal tissue. Pre-trained Deep Neural Networks (DNNs) extract these features and feed them into a 10-fold crossvalidation SVM classifier. The SVM is evaluated using various kernel functions to identify the best kernel for separating data points. This methodology aims to achieve accurate classification of breast cancer in ultrasound images. Findings: The study confirms the effectiveness of deep transfer learning for breast cancer detection in ultrasound images, with Inception V3 outperforming VGG-16 and VGG-19 in extracting relevant features. The combination of Inception V3 and the SVM classifier with a polynomial kernel achieved the highest classification accuracy, indicating its ability to model complex relationships. The study demonstrated an AUC of 0.944 and a classification accuracy of 87.44% using the Inception V3 + SVM polynomial. Novelty: This research demonstrates the potential of deep transfer learning and SVM classifiers for accurate breast cancer detection in ultrasound images. It integrates Inception V3, VGG-16, and VGG-19 for breast cancer detection, demonstrating improved classification accuracy. The combination of Inception V3 and SVM (polynomial) achieved a significant AUC (0.944) and classification accuracy (87.44%), outperforming other models tested. This research underscores the potential of these technologies for accurate breast cancer detection in ultrasound images.

International journal of computing and digital system/International Journal of Computing and Digital Systems, May 1, 2024

Alcohol consumption can lead to vocal health risks and long-term health issues for individuals. T... more Alcohol consumption can lead to vocal health risks and long-term health issues for individuals. The paper introduces a novel dataset that analyzes vowel vocalizations to detect early alcohol consumption. This study examines hidden parameters in vowel sounds, such as frequency, jitters, shimmer, and harmonic ratio, which can identify individuals who consume alcohol. It aims to identify subtle vocal patterns that serve as markers for alcohol consumption. This study analyzed 509 vowel vocalizations from 290 records of 46 alcohol-consuming individuals and 219 non-drinkers aged 22-34. The study used intelligent machine learning models and Incremental Hidden Layer Neurons Artificial Neural Networks (IHLN-ANNs) with Back-propagation to identify patterns indicative of alcohol consumption. The Random Forest (RF) model achieved 95.3% accuracy, while the BP-ANNs model showed 99.4% accuracy with five neurons in a hidden layer. The findings could be applied to developing smartphone applications to provide timely alerts and cautionary measures for alcohol consumption, reducing accident risks. The study highlights voice analysis's potential as a non-invasive and cost-effective tool for identifying alcohol consumers, offering potential avenues for future public health initiatives.

Feature importance for software development effort estimation using multi level ensemble approaches

Bulletin of Electrical Engineering and Informatics, Apr 1, 2024

Feature importance strategy that substantially impacts software development effort estimation (SD... more Feature importance strategy that substantially impacts software development effort estimation (SDEE) can help lower the dimensionality of dataset size. SDEE models developed to estimate effort, time, and wealth required to accomplish a software product on a limited budget are used more frequently by project managers as decision-support tool effort estimation algorithms trained on a dataset containing essential elements to improve their estimation accuracy. Earlier research worked on creating and testing various estimation methods to get accurate. On the other hand, ensemble produces superior prediction accuracy than single approaches. Therefore, this study aims to identify, develop, and deploy an ensemble approach feasible and practical for forecasting software development activities with limited time and minimum effort. This paper proposed a collaborative system containing a multi-level ensemble approach. The first level grabs the optimal features by adopting boosting techniques that impact the decided target; this subset features forward to the second level developed by a stacked ensemble to compute the product development effort concerning lines of code (LOC) and actual. The proposed model yields high accuracy and is more accurate than distinct models.

Optimizing Chronic Kidney Disease Diagnosis in Uddanam: A Smart Fusion of GA-MLP Hybrid and PCA Dimensionality Reduction

Procedia Computer Science, Dec 31, 2022

Diagnosing chronic kidney disease (CKD) accurately and quickly is a big challenge in healthcare t... more Diagnosing chronic kidney disease (CKD) accurately and quickly is a big challenge in healthcare today. In a region called Uddanam in Andhra Pradesh, India, there's a common kidney disease called Uddanam nephropathy. In the 1990s, Andhra Pradesh was the first place to find this kind of kidney disease. In our research, we gathered data about personal and clinical of 1,055 individuals living in the Uddanam area (Srikakulam district of Andhra Pradesh, India). Among these, 656 had CKD, and 399 did not. We looked at 37 different aspects of their health. We apply for feature reduction Principal component Analysis and for classification uses MLP and Intelligent GA-MLP Hybrid model. This method combines the power of Genetic Algorithms (GA) and Multi-Layer Perceptron neural networks (MLP) to make CKD diagnosis more accurate. By using GA's ability to find the best solutions and MLP's skill in recognizing patterns, this new method can overcome the problems with older ways of diagnosing CKD. We tested our method thoroughly, comparing it to other methods. Our results showed that our approach can find CKD early and accurately. By combining GA and MLP, we're not only helping improve medical diagnosis but also giving hope for better CKD management and patient care. Our hybrid model, which uses PCA (20 dimensions) along with GA and MLP, performed the best with an accuracy of 98.34% during training and 98.54% during testing.

IEEE access, 2024

According to WHO statistics for 2018, there are 1.2 million cases and 700,000 deaths from breast ... more According to WHO statistics for 2018, there are 1.2 million cases and 700,000 deaths from breast cancer (BC) each year, making it the second-highest cause of mortality for women globally. In recent years, advances in artificial (AI) intelligence and machine (ML) learning have shown incredible potential in increasing the accuracy and efficiency of BC diagnosis. This research describes an intelligent BC image analysis system that leverages the capabilities of transfer learning (TLs) with ensemble stacking ML models. As part of this research, we created a model for analyzing ultrasound BC images using cutting-edge TL models such as Inception V3, VGG-19, and VGG-16. We have implemented stacking of ensemble ML models, including MLP (Multi-Layer Perceptron) with different architectures (10 10, 20 20, and 30 30) and Support Vector Machines (SVM) with RBF and Polynomial kernels. We analyzed the effectiveness of the proposed system in performance parameters (accuracy (CA), sensitivity, specificity, and AUC). Compared to the results with existing BC diagnostic systems, the proposed method (Inception V3 + Staking) is superior, with performance parameters 0.947 of AUC and 0.858 of CA values. The proposed BCUI analysis system consists of data collection, pre-processing, transfer learning, ensemble stacking of ML models, and performance evaluation, with comparative analysis demonstrating its superiority over existing methods.

Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET 2024), Advances in Computer Science Research 112, 2024

bstract. Maize plant diseases can have a severe impact on agricultural productivity, making detec... more bstract. Maize plant diseases can have a severe impact on agricultural productivity, making detection and control challenging for farmers. Early identification of diseases is crucial for minimizing losses. This study proposes a new approach that integrates machine learning (ML) and deep learning (DL) algorithms to improve maize disease diagnosis and prognosis. The research employs traditional machine learning algorithms, such as Support Vector Machine (SVM) and Multilayer Perceptron (MLP), along with extracted features of Transfer Learning models, such as InceptionV3, VGG19, and Dense-Net201. The objective is to develop a robust system for early disease detection in maize leaves using image analysis. Optimization techniques, such as the Adam optimizer, and activation functions, such as tanh and sigmoid, are also explored. The results indicate that the Adam optimizer MLP achieves the highest accuracy (MLP(100,100) layers PCA(300) accuracy 0.95107) as well as SVM (RBF kernel) with PCA(100) accuracy (0.95585) exceptional other classification methods. This integrated approach promotes agricultural sustainability and crop yield by enabling prompt disease management. Keywords: Machine Learning, Deep Learning, Transfer Learner, SVM, MLP

Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET 2024), Advances in Computer Science Research 112, 2024

One of the most recent technologies is blockchain. It is used in many industries, including healt... more One of the most recent technologies is blockchain. It is used in many industries, including healthcare, finance, supply chain management, etc. A distributed technology called blockchain makes data or information available to all network nodes. It serves as an electronic database and stores information in digital form. Blockchain is widely famously employed in cryptocurrency systems to create a secure and decentralized transaction. However, a blockchain may be used to store any data. Not only is the data related to cryptocurrency but it is also used in healthcare systems. We can use blockchain to record and store patient data due to its distributed nature. But on the dot, this has become a problem as the data of the particular patient will be all over the network which increases the risk and complexity to handle the data. So, we are going to look for a solution and utilize it. Hence, the data will be authorized, authenticated, transparent, and highly secured. The patient controls access to the data by using the consortium type of the blockchain. The patient decides who can use and access their data, which can be the most beneficial aspect of using blockchain in healthcare.
Keywords: Medical record, Smart Contract, MetaMask, Decentralized system, Authentication

Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET 2024), Advances in Computer Science Research 112, 2024

Abstract. In recent times, generating understandable prompts for AI has been a significant proble... more Abstract. In recent times, generating understandable prompts for AI has been a significant problem, which in turn leads to inaccurate results. In essence, the problem is about finding ways to make AI understand and respond accurately to the prompts given to it, which is crucial for improving its overall performance and usefulness in various applications. This paper proposes a novel approach to enhance AI comprehension by generating tailored understanding prompts through prompt engineering techniques in Natural Language Processing, along with advanced Transformer-based deep learning models. Our project integrates these techniques to transform a base prompt into a set of diverse and comprehensive understanding prompts. To ensure data security, we have also implemented data encryption standard and Blowfish encryption algorithms to protect sensitive information during the transformation process. The resulting prompts will be used to train AI models, enabling them to grasp nuanced details and context when responding to user queries. The paper's significance lies in its potential to improve the quality of AI-generated responses across a range of applications, including natural language understanding, question answering, and content generation. Crucially, the developed web application, constructed using the MERN Stack, promises more reliable and insightful interactions with AI systems, effectively bridging the gap between human comprehension and AI-generated content.

Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET 2024), Advances in Computer Science Research 112, 2024

Parkinson's disease is a disorder of the nervous system that causes impairment and changes in cog... more Parkinson's disease is a disorder of the nervous system that causes impairment and changes in cognitive behavior. Voice analysis has become a crucial tool for diagnosing neurological conditions like PD, with symptoms typically appearing in people aged 50 or older. This research suggests new methods to improve early PD diagnostic methods, focusing on assessing aspects and fine-tuning hyperparameters of machine learning algorithms. The data set includes characteristics of both healthy and PD patients, aged 50 to 85. After processing, pertinent characters or traits are extracted from those voice recordings. In this research paper, we investigate Principal Component Analysis (PCA) for feature selection in conjunction with optimization techniques for training Support Vector Machine and Multilayer Perceptron models. The optimization techniques employed include the Firefly Algorithm, Particle Swarm Optimization (PSO), Grasshopper Optimizer, Grey Wolf Optimizer, and Genetic Algorithm (GA). Our study aims to assess the effectiveness of these optimization algorithms in enhancing the performance of MLP and SVM models on the dataset of Parkinson. The MLP and SVM accuracy rates of the optimization algorithms Firefly, PSO, Genetic, Grey Wolf, and Grasshopper were high; Firefly reached 97% (MLP) and 92% (SVM) accuracy, PSO 82% and 94.87% accuracy, while Genetic, Grasshopper, and Greywolf obtained 82% and 94% accuracy, respectively.

Journal of Emerging Technologies and Innovative Research (JETIR), 2024

Digital certificates serve as essential cryptographic tools in securing online transactions and c... more Digital certificates serve as essential cryptographic tools in securing online transactions and communications by verifying the identities of parties involved. The main goal of this project is to make digital certificates more secure using blockchain technology. Digital certificates are important for verifying things like education degrees, medical records, and financial transactions. However, current methods can be vulnerable to fraud and tampering. We use blockchain technology, which is a decentralized and tamper-proof system. Digital certificates will be stored on the blockchain, making them impossible to alter. Smart contracts will automate the verification process, making it faster and more reliable. By using a decentralized approach, we can eliminate single points of failure and make the whole system more trustworthy. In the future, we'll expand the system to handle different types of certificates and work on making it compatible with existing systems. We'll also research ways to make the blockchain system work for large-scale use. Continuous improvements will be made to keep up with changes in technology and security threats.

IAENG-IJCS, 2024

Rice category identification by image analysis is essential to ensure the quality and safety of r... more Rice category identification by image analysis is essential to ensure the quality and safety of rice production. In this study, we propose an intellectual approach to improve rice category identification using deep transfer learning features and machine learning classifiers. Specifically, we extracted features from three pre-trained models (Inception V3, VGG-19 and VGG-16) using transfer learning techniques. These were used as inputs to train MultiLayerPerceptron (MLP) and support vector machine (SVM) classifiers. The results of our experiments show that the proposed strategic results achieve high accuracy in identifying rice categories. The SVM (polynomial kernel) achieves the second-highest accuracy among all models and features, with an accuracy of 0.9948 using the VGG-19 and 0.9912 using Inception V3. The MLP classifier with (30 30) hidden layers achieve the first high accuracy, with an accuracy of 0.9972 (99.72%) using VGG-19 features. The results also show that the choice of deep transfer learning model and machine learning (ML) classifier can significantly affect the accuracy of rice category identification. Among the three pretrained models, VGG-19 features consistently perform the best, followed by Inception V3 and VGG-16. The choice of MLP hidden layer size also affects the accuracy, with 30 HL neurons achieving the best performance. Our proposed approach using deep TL features and ML classifiers shows promising results in improving rice category identification. Our study provides valuable insights into optimizing ML models for agricultural image analysis.

Indian Society for Education and Environment (iSee), 2024

Objectives: The research aims to enhance breast cancer detection accuracy and eﬀectiveness using ... more Objectives: The research aims to enhance breast cancer detection accuracy
and eﬀectiveness using deep transfer learning and pre-trained neural net-
works. It analyses breast ultrasound images and identiﬁes important charac-
teristics using pre-trained networks. The goal is to create a more eﬃcient and
accurate automated system for breast cancer detection. Methods: The study
uses breast ultrasound cancer image data from the Kaggle Data Repository to
extract informative features, identify cancer-related characteristics, and clas-
sify them into benign, malignant, and normal tissue. Pre-trained Deep Neural
Networks (DNNs) extract these features and feed them into a 10-fold cross-
validation SVM classiﬁer. The SVM is evaluated using various kernel functions
to identify the best kernel for separating data points. This methodology aims
to achieve accurate classiﬁcation of breast cancer in ultrasound images. Find-
ings: The study conﬁrms the eﬀectiveness of deep transfer learning for breast
cancer detection in ultrasound images, with Inception V3 outperforming VGG-
16 and VGG-19 in extracting relevant features. The combination of Inception
V3 and the SVM classiﬁer with a polynomial kernel achieved the highest clas-
siﬁcation accuracy, indicating its ability to model complex relationships. The
study demonstrated an AUC of 0.944 and a classiﬁcation accuracy of 87.44%
using the Inception V3 + SVM polynomial. Novelty: This research demonstrates
the potential of deep transfer learning and SVM classiﬁers for accurate breast
cancer detection in ultrasound images. It integrates Inception V3, VGG-16, and
VGG-19 for breast cancer detection, demonstrating improved classiﬁcation
accuracy. The combination of Inception V3 and SVM (polynomial) achieved a sig-
niﬁcant AUC (0.944) and classiﬁcation accuracy (87.44%), outperforming other
models tested. This research underscores the potential of these technologies
for accurate breast cancer detection in ultrasound images.

Indian Society for Education and Environment (iSee), 2024

Bulletin of Electrical Engineering and Informatics, 2023

Procedia Computer Science, 2023

International Journal of Computing and Digital Systems, 2025

IEEE ACCESS, 2024

Institute of Advanced Engineering and Science (IAES), 2024

Drinkers Voice Recognition Intelligent System: An Ensemble Stacking Machine Learning Approach

Annals of data science, Jul 7, 2024

Indian journal of science and technology, May 14, 2024

International journal of computing and digital system/International Journal of Computing and Digital Systems, May 1, 2024

Feature importance for software development effort estimation using multi level ensemble approaches

Bulletin of Electrical Engineering and Informatics, Apr 1, 2024

Optimizing Chronic Kidney Disease Diagnosis in Uddanam: A Smart Fusion of GA-MLP Hybrid and PCA Dimensionality Reduction

Procedia Computer Science, Dec 31, 2022

IEEE access, 2024

Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET 2024), Advances in Computer Science Research 112, 2024

Journal of Emerging Technologies and Innovative Research (JETIR), 2024

IAENG-IJCS, 2024

Indian Society for Education and Environment (iSee), 2024

Bulletin of Electrical Engineering and Informatics, 2023

Procedia Computer Science, 2023