Dinesh Chowdary Attota

Dinesh Chowdary Attota

Auto-generated profile, based on publicly available information

Dinesh Chowdary Attota

Kennesaw State University, Marietta, Georgia, USA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 August 2021 in IEEE Access
Verified
Journal Article
Open access
DADinesh Chowdary Attota
Dinesh Chowdary Attota
VMViraaji Mothukuri
Viraaji Mothukuri
RPReza Parizi
Reza Parizi
Seyedamin PouriyehSeyedamin Pouriyeh
Seyedamin Pouriyeh
The rise in popularity of Internet of Things (IoT) devices has attracted hackers to develop IoT-specific attacks. The microservice architecture of IoT devices relies on the Internet to provide their intended services. An unguarded IoT network makes inter-connected devices vulnerable to attacks. It will be a tedious and ineffective process to manually detect the attacks in the network, as the attackers frequently upgrade their attack strategies. Machine learning (ML)-assisted approaches have been proposed to build intrusion detection for cybersecurity automation in IoT networks. However, most such approaches focus on training an ML model using a single view of the dataset, which often fails to build insightful knowledge and understand each feature’s impact on the ML model’s decision-making ability. As such, the model training with a single view may result in an incomplete understanding of patterns in large feature-set datasets. Moreover, the current approaches are mainly designed in a centralized manner in which the raw data is transferred from the edge devices to the central server for training. This, in turn, may expose the data to all kinds of attacks without adhering to the privacy-preserving of data security. Multi-view learning has gained popularity for its ability to learn from different data views and deliver efficient performance with more distinguished predictions. This paper proposes a federated learning-based intrusion detection approach, called MV-FLID, that trains on multiple views of IoT network data in a decentralized format to detect, classify, and defend against attacks. The multi-view ensemble learning aspect helps in maximizing the learning efficiency of different classes of attacks. The Federated Learning (FL) aspect, wherein the device’s data is not shared to the server, performs profile aggregation efficiently with the benefit of peer learning. Our evaluation results show that our proposed approach has higher accuracy compared to the traditional non-FL centralized approach.

18 April 2024
Verified
Conference Paper
Open access
WSWilliam Stigall
William Stigall
AKAbdullah Al Hafiz Khan
Abdullah Al Hafiz Khan
DADinesh Attota
Dinesh Attota
FNFrancis Nweke
Francis Nweke
YPYong Pei
Yong Pei
The increasing application of artificial intelligence in daily life necessitates precise emotion classification for improved user interactions in areas like healthcare, marketing, and customer service. This work explores the development of emotion classification algorithms, focusing on text classification and providing low latency inferencing for seamless application into persistent-state systems. We use a parallel multi-task learning approach, to learn multiple tasks with a single loss function, allowing it to learn representations in both Emotion Classification and Sentiment Analysis simultaneously. We present a detailed analysis of a fine-tuned BERTTiny model EmoBERTTiny for emotion and sentiment classification tasks, comparing its performance against baseline models and state-of-the-art 7B parameter models. EmoBERTTiny is bench-marked against Llama-2-7B-chat and Mistral-7B-Instruct across Accuracy, F1-score, precision-recall curves and inference speed. EmoBERTTiny outperforms pre-trained and state-of-the-art models across all metrics and computational efficiency, achieving 93.14% accuracy in sentiment analysis and 85.48% accuracy on emotion classification. EmoBERTTiny processes a 256 token context window in 8.04ms on average post-tokenization, and 154.23ms on average in total processing speed.

23 March 2024 in Smart Health
Verified
Journal Article
Closed access
DADinesh Chowdary Attota
Dinesh Chowdary Attota
AAAbm Adnan Azmee
Abm Adnan Azmee
AKAbdullah Al Hafiz Khan
Abdullah Al Hafiz Khan
YPYong Pei
Yong Pei
DTDominic Thomas
Dominic Thomas
MNMonica Nandan
Monica Nandan