Computer Engineering, Computer Networks and Communications, Discrete Mathematics and Combinatorics, Artificial Intelligence
5
Scopus Publications
Scopus Publications
ADVERSARIAL ATTACKS ON HEALTHCARE DEEP LEARNING MODEL: VULNERABILITIES AND DEFENSES Mekala A, Shahnaz Fatima, Subba Rao BV, Rajendra Prasad J, Banupriya P.G, K venkataramana Informing Science, 2025 Aim/Purpose: The aim of this work is to introduce a novel technique for enhancing hospital network security using a Deep Autoencoder (DAE). Background: The digitization of healthcare institutions increases the risk of cybercrime due to network security vulnerabilities. Methodology: In this work, we model a novel technique called DeepHeal framework that uses a robust Deep Autoencoder (DAE) for improved cybersecurity and protection against intruders in healthcare networks. Using the deep learning techniques named DAE, DeepHeal identifies and stops cyber dangers, including various hostile attacks, illegal access, and data breaches. Healthcare networks are susceptible to various cyberattacks and intrusions. The DAE architecture enables training with high-level representations to detect anomalies and traffic patterns. By carefully analyzing datasets from real healthcare networks, we demonstrate that DeepHeal effectively identifies and counters various cyber threats. Contribution: The research using the proposed method offers a great potential in terms of accuracy, scalability, and real-time threat detection. Moreover, by highlighting the specific elements and network characteristics that are behind the anomalies it discovers, DeepHeal makes comprehension of them easier. Findings: The proposed DAE model combined with RNN achieves a higher accuracy and precision level of 98%. The proposed DAE model outperformed existing models, which offers its ability to identify and resolve cybersecurity problems in hospital networks. Recommendation for Researchers: The security threats on patient data is considered sensitive and it offers improved healthcare networks security and this is found essential for the reliability and confidentiality of patient data in healthcare networks. Future Research: The proposed method’s research offers great potential in terms of accuracy, scalability, and real-time threat detection.
Sugarcane yield prediction using NOA-based swin transformer model in IoT smart agriculture V. Gokula Krishnan, B. V. Subba Rao, J. Rajendra Prasad, P. Pushpa, S. Kumari Journal of Applied Biology and Biotechnology, 2024 The Internet of things (IoT) empowers precise organization and intelligent coordination for industrial facilities and smart farming, enhancing agricultural efficiency. Sugar production relies on various auxiliary elements, but in labor-intensive smart agriculture, creating accurate forecasts is a formidable challenge. Machine learning emerges as a potential solution, as current convolutional neural network-based phase recognition techniques struggle with long-range dependencies. To address this, a temporal-based swin transformer network (TSTN) is introduced, comprising a swin transformer and long short-term memory (LSTM). The swin transformer employs attention mechanisms for expressive representations, while LSTM excels at extracting temporal data with long-range dependencies. The nutcracker optimizer algorithm (NOA) fine-tunes LSTM weights. TSTN effectively blends these components, providing spatiotemporal data with enhanced context. This model outperforms competitors in accuracy, as demonstrated through testing with data from Uttar Pradesh. The integration of IoT and TSTN marks a significant advancement in optimizing agricultural operations for increased productivity and efficiency. In the comparative analysis, the proposed TSTN-NOA model achieves better performance and results than other existing
Segmentation of Paddy Fields from A Remote Sensing Images Using Ai Based Learning International Journal of Intelligent Systems and Applications in Engineering, 2024
CCEE: Clustering with chicken swarm based energy efficient algorithm with APUGR protocol for mobility awareness and energy saving in adhoc network T. Santhi Sri, J. Rajendra Prasad, R. Kiran Kumar International Journal of Knowledge Based and Intelligent Engineering Systems, 2018 Mobility awareness and energy efficiency have been considered as the two irreplaceable enhancement issues in Mobile Ad Hoc Networks (MANETs) where nodes navigate erratically toward any path with limited battery life, bringing about incessant change in topology. These constraints are generally conce ntrated to build the lifetime of such systems. The research work proposes Clustering with Chicken swarm based Energy Efficient (CCEE) algorithm to take care of the advanced issues for useful dynamic data transmission alongside Adaptive Position Update based Geographic Routing (APUGR) protocol. The race of cluster heads deals with mobility and remaining vitality and additionally, the level of availability for choosing nodes to fill in as cluster head set out toward longer duration of time. The cluster formation is exhibited by taking multi-objective fitness work utilizing chicken swarm enhancement. APUGR progressively changes the frequency of position refreshes, in light of the mobility dynamics of the nodes and the sending designs in the system. It depends on two basic standards: (i) nodes whose developments are harder to foresee refresh their positions all the more as often as possible and the other way round, (ii) nodes nearer to broadcasting ways refresh their positions all the more oftentimes and the other way round. In light of this routing, the data transmission from source to end user is extremely viable. The recreation comes about to demonstrate that the execution of CCEE achieves preferred outcomes, which are analyzed over existing energy efficient algorithms plot as far as load balancing factor, throughput, inter packet delivery ratio, average end-to-end delay and power consumption by using NS2 simulations.
An intuitive based clustering on document mining using dirichlet process mixture models and its kernels D. Ratnam, B. V. Subba Rao, J. Rajendra Prasad, S. Sai Kumar International Conference on Signal Processing Communication Power and Embedded System Scopes 2016 Proceedings, 2017 In machine learning and data mining tasks, Clustering is considered to be one of the most important techniques. The same sorts of documents are grouped by performing clustering techniques. Similarity measuring is used to determine transaction relationships. Hierarchical clustering model generates tree structured results. Partitioned based clustering produces the result in grid format. Text documents are unstructured data values with high dimensional attributes. Document clustering group transforms unlabeled text documents into meaningful clusters. In the event of document grouping process, traditional clustering methods require cluster count (K). Clustering accuracy degrades drastically with reference to the unsuitable cluster count. It is observed that document features are automatically partitioned into two groups namely — discriminative words and non-discriminative words. In particular, discriminative words are only useful for grouping documents. The involvement of non-discriminative words confuses the clustering process and leads to poor clustering solution in return. A variation inference algorithm is used to infer the document collection structure and partition of document words simultaneously. Dirichlet Process Mixture (DPM) model is used to partition documents in a way utilizing both the data likelihood and the clustering property of the Dirichlet Process (DP). Dirichlet Process Mixture Model for Feature Extraction (DPMFE) is used to discover the latent cluster structure based on the DPM model and it is performed without involving the number of clusters as input.
