Nithiya Baskaran

@citchennai.edu.in

Associate Professor, Department of Computer Science and Engineering
Chennai Institute of Technology

Nithiya Baskaran


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

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Engineering, Artificial Intelligence, Human-Computer Interaction
11

Scopus Publications

49

Scholar Citations

4

Scholar h-index

1

Scholar i10-index

Scopus Publications

  • Enhancing American Sign Language Detection Through Histogram of Gradient Features Using Machine Learning Techniques
    Nithiya Baskaran, P Aurchana, K. Manasa, B. Jogeswara Rao
    2026 6th International Conference on Advances in Electrical Computing Communications and Sustainable Technologies Icaect 2026, 2026
    The predominant means of communication is speech; however, there are persons whose speaking or hearing abilities are impaired. Communication presents a significant barrier for persons with such disabilities. The primary objective of this work is to develop an efficient solution with machine learning for the speech and hearing impairment community with the help of AI. This study helps in the development of future technologies in the enhancement of communication accessibility. In the proposed work, a machine learning approach for the American Sign Language alphabet using image detection and processing techniques are utilized. The experiment was conducted using an American Sign Language recognition dataset obtained from Kaggle, which provides images of hand Gestures of the alphabet. The dataset was divided into training and testing subsets in the ratio of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$80: 20$</tex>, where 80 % of the data was used in training the model and the remaining 20 % was designated for testing the model. The dataset was normalized using denoising filters and proceeded with Canny edge detection to highlight sharp edge features. From the edge detected image, Histogram of Oriented Gradient features were extracted to detect the letter and shape information. These extracted features are then fed into the classifiers: Support Vector Machine, Random Forest, and KNearest Neighbors. The study was analyzed by experimental results, which show that the Support Vector Machine classifier achieved the highest accuracy of 93.33 %, closely followed by Random Forest and K-Nearest Neighbors. The analysis confirms that applying edge-based preprocessing along with the Histogram of Oriented Gradients feature extraction enables an effective classification of American Sign Language alphabets. This Comparative study provides a solid foundation for developing low-complexity, high-accuracy sign language recognition systems.
  • AI-Powered CNN Model for Automated Lung Cancer Diagnosis in Medical Imaging
    Walid Ayadi, Yasser Farhat, Saeed Ali Althabahi, Nithiya Baskaran, Showkat A. Dar, R. Indhumathi, Madhuri Prashant Pant, Showkat A. Bhat, Aafaq A. Rather
    International Journal of Statistics in Medical Research, 2025
    Lung cancer remains a critical health concern in the entire world, which has been a major cause of high rates of cancer-related mortalities that affect individuals in every part of the world. The findings emphasize the notable potential of deep learning procedures to assist radiologists in diagnosing cases of lung-related abnormalities appropriately. Such methods are also leading to the improvement of AI-based healthcare products. The enhancements to the suggested model [16, 17, 18, 21] in the future will be aimed at tuning hyperparameters, 3D CNN [16, 17, 18] architectures, and the integration of patient clinical data, with the aim of further increasing the accuracy [16, 17, 19] of diagnosis as well as system performance. This paper uses the IQ-OTHNCCD dataset, a publicly available and highly annotated set of CT imaging that has been annotated by experts in the medical field. The preprocessing techniques applied will involve changing the images to Grayscale, normalizing the pixel values, ensuring consistency in the images, and converting them to a standard size of 128x128 pixels, which is the ideal size to feed the images into the CNN [16, 17, 18]. In the proposed work, the model [16, 17, 18, 21] integrates multi-scale convolutional layers with adaptive dropout (rate=0.5) and ReLU activations, yielding 95% accuracy [16, 17, 19] and 0.95 F1-score (95% CI: 93.8–96.2%) on a 70/15/15 train/validation/test split— a 4% improvement in F1-score. Preprocessing includes grayscale conversion, pixel normalization to [0,1], and resizing to 128x128 pixels. The architecture comprises three convolutional blocks (32/64/128 filters, 3x3 kernels), max-pooling (2x2), flattening, a 512-unit dense layer, and a 3-unit softmax output. Future enhancements include hyperparameter tuning, 3D CNN [16, 17, 18] integration, and clinical data fusion to exceed 97% accuracy [16, 17, 19].
  • Evaluating AI-Driven Recommendation Systems Efficacy in Delivering Personalized Content on Internet Streaming Platforms
    Nithiya Baskaran, U. Shivani, S. Varsha
    Lecture Notes in Networks and Systems, 2025
  • Soil Detection And Crop Recommendation Using Convolutional Neural Networks
    P Aurchana, Kothakonda Manasa, Nithiya Baskaran, B. Jogeswara Rao
    Conference Proceedings 2025 IEEE Silchar Subsection Conference IEEE Silcon 2025, 2025
    India is a land of agriculture, ranking among the top three global producers of numerous crops. The Indian farmer is at the center of the agricultural sector, but most Indian farmers remain at the bottom of the social hierarchy. Furthermore, farmers struggle to determine which crop is best suited and profitable for their soil, despite the few technical options available today, due to the variance in soil types between geographical regions. In the proposed work, a comprehensive framework for soil type classification and crop suggestion is presented to classify soil based on Convolutional Neural Networks, specifically VGG19, MobileNet V2, and ResNet50, utilizing image characteristics of soil types (color and texture). For this, Pre-Trained Convolutional Neural Networks, namely VGG19, MobileNet V2, and ResNet50, are utilized as feature extractors. The extracted features were fed into the Support Vector Machine, in which Mobilenetv2 with Support Vector Machine gives the highest accuracy of 99%. Besides the nitrogen, phosphorus, potassium, pH, temperature, and moisture can also be given as inputs to predict the fertility. It is a system developed to assist farmers and agriculture-related experts with information for making informed decisions for improved crop yield as well as soil management. The model delivered high accuracy and F1 scores for various types of soils, which is crucial for successful application in practice.
  • Sustainable Farming: Web Platform Empowering Indian Farmers through AI Insights
    Nithiya Baskaran, S. Amutha, Pachaivannan Partheeban, P. Aurchana, S. Mahalakshmi
    Proceedings IEEE 10th International Conference on Smart Structures and Systems Icsss 2025, 2025
  • The Role of AI in Vehicle-to-Infrastructure Communication for Tolling Systems
    Mahalakshmi S, Nithiya Baskaran, Pachaivanna Partheeban, Kowsalyadevi Krishnara, Arjun S
    Proceedings of 2024 IEEE International Women in Engineering Wie Conference on Electrical and Computer Engineering Wiecon Ece 2024, 2024
    The number of vehicles people use is substantial; thus, surveillance must be accurate. License plates are detected and recognized by detecting static and real-time images captured by the camera using OCR technology, Computer vision, and AI in the tool booth and other important roads. This model can be used for many things, such as automated services, increased security, determining whether a car is registered, and identifying and validating vehicles connected to traffic violations. Real-time investigations are part of the automated recognition of license plates and their application to other traffic control and highway surveillance aspects. To detect plates from the images and in real-time video, Use Easy OCR with the license plates to read the plate number, utilize tensor flow object detection to detect and recognize, and save the detected license plate for further analysis and search.
  • Efficient VM selection strategies in cloud datacenter using fuzzy soft set
    Nithiya Baskaran, Eswari R.
    Journal of Organizational and End User Computing, 2021
    A cloud data center is established to meet the storage demand due to the rate of growth of data. The inefficient use of resources causes an enormous amount of power consumption in data centers. In this paper, a fuzzy soft set-based virtual machine (FSS_VM) consolidation algorithm is proposed to address this problem. The algorithm uses four thresholds to detect overloaded hosts and applies fuzzy soft set approach to select appropriate VM for migration. It considers all factors: CPU utilization, memory usage, RAM usage, and correlation values. The algorithm is experimentally tested for 11 different combinations of choice parameters where each combination is considered as fuzzy soft set and compared with existing algorithms for various metrics. The experimental results show that proposed FSS_VM algorithm achieves significant improvement in optimizing the objectives such as power consumption, service level agreement violation rate, and VM migrations compared to all existing algorithms. Moreover, performance comparison among the fuzzy soft set-based VM selection methods are made, and Pareto-optimal fuzzy soft sets are identified. The results show that the Pareto-based VM selection improves the QoS. The time complexity of the proposed algorithm increases when it finds best VM for migration. The future work will reduce the time complexity and will concentrate on developing an efficient VM placement strategy for VM migration since it has the greater impact on improving QoS in VM placement.
  • An efficient threshold-fuzzy-based algorithm for VM consolidation in cloud datacenter
    Nithiya Baskaran, R. Eswari
    International Journal of Grid and High Performance Computing, 2021
    Cloud computing has grown exponentially in the recent years. Data growth is increasing day by day, which increases the demand for cloud storage, which leads to setting up cloud data centers. But they consume enormous amounts of power, use the resources inefficiently, and also violate service-level agreements. In this paper, an adaptive fuzzy-based VM selection algorithm (AFT_FS) is proposed to address these problems. The proposed algorithm uses four thresholds to detect overloaded host and fuzzy-based approach to select VM for migration. The algorithm is experimentally tested for real-world data, and the performance is compared with existing algorithms for various metrics. The simulation results testify to the proposed AFT_FS method is the utmost energy efficient and minimizes the SLA rate compared to other algorithms.
  • CPU-memory aware VM consolidation for cloud data centers
    Scalable Computing, 2020
  • Fuzzy softset based VM selection in cloud datacenter
    Eswari Rajagopal, Nithiya Baskaran
    2019 International Conference on Intelligent Computing and Control Systems Iccs 2019, 2019
    The data is increasing day to day exponentially, which increases demand for cloud storage that leads to the establishment of an enormous number of cloud data centers. But usage of cloud data center employs huge amounts of electrical power and end up costly. Virtual machine consolidation decreases Service Level Agreement (SLA) Violations and energy consumption. In order to achieve that optimal selection of VM for migration is needed. Existing approaches concentrate on one selection factor at a time for VM selection. But the proposed algorithm considers all four selection parameters at a time using fuzzy soft set theory. The experimental results show that the proposed FSS-RCMCor algorithm efficiently selects VM for migration than the existing algorithms with minimum SLA violation rate and less number of Virtual machine migrations, and maximum energy efficiency.
  • Adaptive Threshold-Based Algorithm for Multi-objective VM Placement in Cloud Data Centers
    Nithiya Baskaran, R. Eswari
    Lecture Notes in Electrical Engineering, 2019

RECENT SCHOLAR PUBLICATIONS

  • Brain Computer Interface Using EEG Signals for Emotional State Detection
    PA Kumar, OGA Balaji, N Baskaran, P Aurchana
    2026 International Conference on Data Science, Agents and Artificial … , 2026
    2026
  • Enhancing American Sign Language Detection Through Histogram of Gradient Features Using Machine Learning Techniques
    N Baskaran, P Aurchana, K Manasa, BJ Rao
    2026 Sixth International Conference on Advances in Electrical, Computing … , 2026
    2026
  • SeaSense: An AI-Powered Conversational Interface for Oceanographic Data Analysis Using Retrieval Augmented Generation Architecture
    N Baskaran, S Srinithi, V Samyuktha
    2026 International Conference on Data Science, Agents and Artificial … , 2026
    2026
  • Android Espion: Integrated Remote Control for Android Device
    S Mahalakshmi, N Baskaran, RR Hemamalini, K Kowsalyadevi, ...
    2025 IEEE 9th International Conference on Information and Communication … , 2025
    2025
  • Sustainable Farming: Web Platform Empowering Indian Farmers through AI Insights
    N Baskaran, S Amutha, P Partheeban, P Aurchana, S Mahalakshmi
    2025 10th International Conference on Smart Structures and Systems (ICSSS), 1-6 , 2025
    2025
  • Soil Detection And Crop Recommendation Using Convolutional Neural Networks
    P Aurchana, K Manasa, N Baskaran, BJ Rao
    2025 IEEE Silchar Subsection Conference (SILCON), 1-6 , 2025
    2025
  • EFFICIENT RESOURCE ALLOCATION IN CLOUD DATA CENTER USING HYBRID MULTI-OBJECTIVE OPTIMIZATION ALGORITHMS
    N Baskaran
    International Journal of Applied Mathematics 38 (10s), 266-280 , 2025
    2025
  • Evaluating AI-Driven Recommendation Systems Efficacy in Delivering Personalized Content on Internet Streaming Platforms
    N Baskaran, U Shivani, S Varsha
    International Conference on Soft Computing: Theories and Applications, 209-219 , 2024
    2024
  • The Role of AI in Vehicle-to-Infrastructure Communication for Tolling Systems
    S Mahalakshmi, N Baskaran, P Partheeban, K Krishnara, S Arjun
    2024 IEEE International Women in Engineering (WIE) Conference on Electrical … , 2024
    2024
  • Efficient VM Selection Strategies in Cloud Datacenter Using Fuzzy Soft Set
    N Baskaran, R Eswari
    Journal of Organizational and End User Computing (JOEUC) 33 (5), 27 , 2021
    2021
    Citations: 29
  • An Efficient Threshold-Fuzzy-Based Algorithm for VM Consolidation in Cloud Datacenter
    N Baskaran, R Eswari
    International Journal of Grid and High Performance Computing (IJGHPC) 13 (1 … , 2021
    2021
    Citations: 2
  • CPU-MEMORY AWARE VM CONSOLIDATION FOR CLOUD DATA CENTERS
    B NITHIYA, R ESWARI
    Journal of Scalable Computing: Practice and Experience 21 (2), 159-172 , 2020
    2020
    Citations: 7
  • Adaptive Threshold-Based Algorithm for Multi-objective VM Placement
    N Baskaran, R Eswari
    Frontier Computing: Theory, Technologies and Applications (FC 2018), 118 , 2019
    2019
  • Fuzzy Softset based VM Selection in Cloud Datacenter
    R Eswari, B Nithiya
    2019 International Conference on Intelligent Computing and Control Systems … , 2019
    2019
    Citations: 7
  • Adaptive Threshold-Based Algorithm for Multi-objective VM Placement in Cloud Data Centers
    B Nithiya, R Eswari
    International Conference on Frontier Computing, 118-129 , 2018
    2018
    Citations: 4

MOST CITED SCHOLAR PUBLICATIONS

  • Efficient VM Selection Strategies in Cloud Datacenter Using Fuzzy Soft Set
    N Baskaran, R Eswari
    Journal of Organizational and End User Computing (JOEUC) 33 (5), 27 , 2021
    2021
    Citations: 29
  • CPU-MEMORY AWARE VM CONSOLIDATION FOR CLOUD DATA CENTERS
    B NITHIYA, R ESWARI
    Journal of Scalable Computing: Practice and Experience 21 (2), 159-172 , 2020
    2020
    Citations: 7
  • Fuzzy Softset based VM Selection in Cloud Datacenter
    R Eswari, B Nithiya
    2019 International Conference on Intelligent Computing and Control Systems … , 2019
    2019
    Citations: 7
  • Adaptive Threshold-Based Algorithm for Multi-objective VM Placement in Cloud Data Centers
    B Nithiya, R Eswari
    International Conference on Frontier Computing, 118-129 , 2018
    2018
    Citations: 4
  • An Efficient Threshold-Fuzzy-Based Algorithm for VM Consolidation in Cloud Datacenter
    N Baskaran, R Eswari
    International Journal of Grid and High Performance Computing (IJGHPC) 13 (1 … , 2021
    2021
    Citations: 2
  • Brain Computer Interface Using EEG Signals for Emotional State Detection
    PA Kumar, OGA Balaji, N Baskaran, P Aurchana
    2026 International Conference on Data Science, Agents and Artificial … , 2026
    2026
  • Enhancing American Sign Language Detection Through Histogram of Gradient Features Using Machine Learning Techniques
    N Baskaran, P Aurchana, K Manasa, BJ Rao
    2026 Sixth International Conference on Advances in Electrical, Computing … , 2026
    2026
  • SeaSense: An AI-Powered Conversational Interface for Oceanographic Data Analysis Using Retrieval Augmented Generation Architecture
    N Baskaran, S Srinithi, V Samyuktha
    2026 International Conference on Data Science, Agents and Artificial … , 2026
    2026
  • Android Espion: Integrated Remote Control for Android Device
    S Mahalakshmi, N Baskaran, RR Hemamalini, K Kowsalyadevi, ...
    2025 IEEE 9th International Conference on Information and Communication … , 2025
    2025
  • Sustainable Farming: Web Platform Empowering Indian Farmers through AI Insights
    N Baskaran, S Amutha, P Partheeban, P Aurchana, S Mahalakshmi
    2025 10th International Conference on Smart Structures and Systems (ICSSS), 1-6 , 2025
    2025
  • Soil Detection And Crop Recommendation Using Convolutional Neural Networks
    P Aurchana, K Manasa, N Baskaran, BJ Rao
    2025 IEEE Silchar Subsection Conference (SILCON), 1-6 , 2025
    2025
  • EFFICIENT RESOURCE ALLOCATION IN CLOUD DATA CENTER USING HYBRID MULTI-OBJECTIVE OPTIMIZATION ALGORITHMS
    N Baskaran
    International Journal of Applied Mathematics 38 (10s), 266-280 , 2025
    2025
  • Evaluating AI-Driven Recommendation Systems Efficacy in Delivering Personalized Content on Internet Streaming Platforms
    N Baskaran, U Shivani, S Varsha
    International Conference on Soft Computing: Theories and Applications, 209-219 , 2024
    2024
  • The Role of AI in Vehicle-to-Infrastructure Communication for Tolling Systems
    S Mahalakshmi, N Baskaran, P Partheeban, K Krishnara, S Arjun
    2024 IEEE International Women in Engineering (WIE) Conference on Electrical … , 2024
    2024
  • Adaptive Threshold-Based Algorithm for Multi-objective VM Placement
    N Baskaran, R Eswari
    Frontier Computing: Theory, Technologies and Applications (FC 2018), 118 , 2019
    2019