N Toyaad Kumar Reddy

@nist.edu

N Toyaad Kumar Reddy


               Download Compact PDF  
https://researchid.co/toyaadnreddy
3

Scopus Publications

5

Scholar Citations

1

Scholar h-index

Scopus Publications

  • AI-Based Chatbot with Recommender System for Interactive Support
    N. Toyaad Kumar Reddy, Manas Ranjan Patra, Brojo Kishore Mishra
    Lecture Notes in Networks and Systems, 2026
  • An Integrated Multimodal Deep Learning Framework for Skin Disease Detection
    Shubhasri Pradhan, Sobhana Behera, N. Toyaad Kumar Reddy, Brojo Kishore Mishra, Manas Ranjan Patra
    2025 International Conference on Next Generation of Green Information and Emerging Technologies Giet 2025, 2025
    Accurate diagnosis of skin diseases from dermoscopic images and patient-reported symptoms remains a critical challenge in dermatology. In this work, we propose an integrated multimodal deep learning framework that fuses visual features from convolutional neural networks (CNNs) with textual features extracted from symptom descriptions to improve classification performance across nine common skin conditions. The visual branch employs a three-stage CNN with progressive convolution and pooling layers, followed by fully connected and dropout layers for robust feature extraction. The textual branch utilizes word embedding and a two-layer LSTM network to capture sequential symptom patterns. Features from both branches are concatenated and passed through dense layers with dropout for final classification. To enhance generalization, the image data are augmented via rotation, shifting, and horizontal flipping, while text inputs are tokenized and padded to uniform length. With early halting and model checkpointing to prevent overfitting, models get trained from beginning to end using an Adams optimiser and categorical cross-entropy loss. Analysis using a held-out sample of 10,000 dermoscopic pictures with standardised symptom data shows that the multi-modal framework outperforms the CNN-only a starting point by 6.3 % in accuracy with 0.07 in F1-score, with an overall accuracy of 88.7<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> and an approximate F1-score of 0.89. Confusion matrix analysis shows improved discrimination among visually similar conditions such as actinic keratosis and benign keratosis. These results indicate that integrating textual symptom information with image analysis substantially enhances diagnostic accuracy and reliability. The proposed system offers a promising tool for computer-aided skin disease screening and decision support in clinical settings.
  • Enhancing Agricultural Decision Making with an AI-Powered Crop Recommendation Engine
    Sobhana Behera, Shubhasri Pradhan, N. Toyaad Kumar Reddy, Brojo Kishore Mishra, Manas Ranjan Patra
    2025 International Conference on Next Generation of Green Information and Emerging Technologies Giet 2025, 2025
    The agricultural sector faces growing challenges in optimizing crop selection due to fluctuating climatic conditions, soil variability, and limited access to expert agronomic guidance. This paper presents an AI-powered crop recommendation engine designed to assist farmers and policymakers in making data-driven decisions tailored to local environmental conditions. The suggested system makes use of a collaborative machine learning technique that combines Random Forest, XGBoost, and support vector machine (SVM) classifiers to forecast the best crops based on input characteristics like rainfall, temperature, humidity, pH, nitrogen, phosphorus, and potassium content. A soft voting mechanism combines predictions from the individual models to improve accuracy and robustness. The system is further enhanced with a location-aware weather integration layer that dynamically fetches environmental parameters based on the user's geographic coordinates via a RESTful API interface. Extensive experimentation was conducted using a benchmark agricultural dataset, and the ensemble model achieved an accuracy of 98.5%, outperforming individual classifiers. The model is encapsulated within a Flask-based API, enabling real-time deployment for web and mobile applications. Results demonstrate that the proposed system offers high precision and reliability, making it a practical tool for precision agriculture. By providing farmers with intelligent crop recommendations, the engine aims to enhance yield potential, resource efficiency, and overall agricultural sustainability.

RECENT SCHOLAR PUBLICATIONS

  • Real-Time Multi-Label Emotion and Mental State Analysis using DistilBERT with Interactive Streamlit Visualization
    S Lokesh, D Yaswanth, R Dudhisty, K Koushik, NT Reddy
    Bhabani prasad, Real-Time Multi-Label Emotion and Mental State Analysis … , 2026
    2026.0
  • Development of an Intelligent Agricultural Decision Support System for Crop Recommendation Using Machine Learning Techniques
    S Behera, NTK Reddy, S Pradhan
    Next-Generation Computing Systems and Technologies 1 (1), 43-53 , 2025
    2025.0
  • An Integrated Multimodal Deep Learning Framework for Skin Disease Detection
    S Pradhan, S Behera, NTK Reddy, BK Mishra, MR Patra
    2025 International Conference on Next Generation of Green Information and … , 2025
    2025.0
  • Enhancing Agricultural Decision Making with an AI-Powered Crop Recommendation Engine
    S Behera, S Pradhan, NTK Reddy, BK Mishra, MR Patra
    2025 International Conference on Next Generation of Green Information and … , 2025
    2025.0
  • A Modular Retrieval-Augmented Conversational AI Chatbot System with Integrated Recommender Engine Using Local LLMs
    NTK Reddy, MR Patra, BK Mishra
    Cureus Journals 2 (1) , 2025
    2025.0
  • Design and implementation of an ai-based chatbot framework with retrieval-augmented generation and integrated recommender system for interactive user support
    N Reddy
    Available at SSRN 5250507 , 2025
    2025.0
    Citations: 5
  • AI-Based Chatbot with Recommender System for Interactive Support
    NTK Reddy, MR Patra, BK Mishra
    International Conference on Machine Learning, IoT and Big Data, 233-244 , 2025
    2025.0
  • Quantum Artificial Intelligence for Environmental Management
    M Nayak, NTK Reddy
    Quantum Artificial Intelligence, 136-147 , 0

MOST CITED SCHOLAR PUBLICATIONS

  • Design and implementation of an ai-based chatbot framework with retrieval-augmented generation and integrated recommender system for interactive user support
    N Reddy
    Available at SSRN 5250507 , 2025
    2025.0
    Citations: 5
  • Real-Time Multi-Label Emotion and Mental State Analysis using DistilBERT with Interactive Streamlit Visualization
    S Lokesh, D Yaswanth, R Dudhisty, K Koushik, NT Reddy
    Bhabani prasad, Real-Time Multi-Label Emotion and Mental State Analysis … , 2026
    2026.0
  • Development of an Intelligent Agricultural Decision Support System for Crop Recommendation Using Machine Learning Techniques
    S Behera, NTK Reddy, S Pradhan
    Next-Generation Computing Systems and Technologies 1 (1), 43-53 , 2025
    2025.0
  • An Integrated Multimodal Deep Learning Framework for Skin Disease Detection
    S Pradhan, S Behera, NTK Reddy, BK Mishra, MR Patra
    2025 International Conference on Next Generation of Green Information and … , 2025
    2025.0
  • Enhancing Agricultural Decision Making with an AI-Powered Crop Recommendation Engine
    S Behera, S Pradhan, NTK Reddy, BK Mishra, MR Patra
    2025 International Conference on Next Generation of Green Information and … , 2025
    2025.0
  • A Modular Retrieval-Augmented Conversational AI Chatbot System with Integrated Recommender Engine Using Local LLMs
    NTK Reddy, MR Patra, BK Mishra
    Cureus Journals 2 (1) , 2025
    2025.0
  • AI-Based Chatbot with Recommender System for Interactive Support
    NTK Reddy, MR Patra, BK Mishra
    International Conference on Machine Learning, IoT and Big Data, 233-244 , 2025
    2025.0
  • Quantum Artificial Intelligence for Environmental Management
    M Nayak, NTK Reddy
    Quantum Artificial Intelligence, 136-147 , 0