Tathagata Bhattacharya

Dr. Tathagata Bhattacharya

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
18 December 2024 in The Journal of Supercomputing
Verified
Journal Article
Closed access
PDF file
TBTathagata Bhattacharya
Tathagata Bhattacharya
APAdithya Vardhan Peddi
Adithya Vardhan Peddi
SPSrikanth Ponaganti
Srikanth Ponaganti
SVSai Teja Veeramalla
Sai Teja Veeramalla

20 January 2023 in Research Square
Verified
Preprint
Open access
PDF file
VTVivek Tanniru
Vivek Tanniru
TBTathagata Bhattacharya
Tathagata Bhattacharya
With the increasing prevalence of online fraud and the use of fake logos to deceive consumers, there is a need for effective methods to detect and prevent the use of fake logos on the internet. In this paper, we propose a method for detecting fake logos using machine learning techniques. Our approach involves extracting features from the logos and training a classifier to distinguish between real and fake logos. We evaluate the performance of our method on a dataset of real and fake logos and demonstrate its effectiveness in detecting fake logos with high accuracy. Every day, hundreds of domain names, websites and logos are being cloned by cyber criminals who want to gain your trust so they can steal your data. It is becoming a big issue in the online world and needs to be addressed. This article will discuss the initial project background of our new Online Fake Logo Detection System

16 September 2022 in Research Square
Verified
Preprint
Open access
PDF file
TBTathagata Bhattacharya
Tathagata Bhattacharya
JMJianzhou Mao
Jianzhou Mao
SBSutanu Bhattacharya
Sutanu Bhattacharya
TCTing Cao
Ting Cao
XPXiaopu Peng
Xiaopu Peng
XQXiao Qin
Xiao Qin
Power Management Strategies and the impact of carbon dioxide emissions from data centers across the globe have drawn significant attention worldwide. The rapid growth of energy consumed in data centers has led to 1) huge costs 2) depletion of non-renewable resources such as coal and petroleum, and 3) emission of greenhouse gases like CO2 in the atmosphere. These greenhouse gas significantly contributes to the climate change of the earth. To tackle this challenge, our research deals with modeling the energy resources of data centers, thereby offering insights to reduce the global carbon footprint and energy cost. In our model, we prioritize green energy consumption by eliminating brown energy resources. In this process, we devise an algorithm that can determine the amount of CO2 emission in the atmosphere per hour by different energy resources. We create an energy model for data centers by incorporating the support vector regression algorithm. Our model is adroit at projecting energy consumed in data centers powered by green energy. Our experimental results confirm that our model consistently delivers high prediction accuracy in terms of energy usage in data centers. The model is expected to facilitate data analytic venues to optimize energy efficiency and sustaininability for the development of future data centers.

13 August 2025
Verified
Book Chapter
Closed access
PDF file
TBTathagata Bhattacharya
Tathagata Bhattacharya
XPXiaopu Peng
Xiaopu Peng
ADAmit Das
Amit Das
PPPreet Patel
Preet Patel
JPJuhi Prajapati
Juhi Prajapati
SVSai Teja Veeramalla
Sai Teja Veeramalla

10 August 2025
Verified
Book Chapter
Closed access
PDF file
TBTathagata Bhattacharya
Tathagata Bhattacharya
VAVinod Alabana
Vinod Alabana
NVNithin Vudayagiri
Nithin Vudayagiri
ADAmith Das
Amith Das

23 January 2023 in Research Square
Verified
Preprint
Open access
PDF file
KRKarthik Kumar Reddy
Karthik Kumar Reddy
TBTathagata Bhattacharya
Tathagata Bhattacharya
SRShreevan Reddy
Shreevan Reddy
Memory forensics is used to implement and investigate malware that is executed or stored in RAM. Whether it is static malware analysis or dynamic malware analysis,each time the malware investigator retrieves the result, it is displayed in plaintext, and the investigator begins examining each result in the plaintext and triaging the malicious request. It's a labor-intensive process, and occasionally an investigator will upload malicious files to his or her computer to be analyzed for malware. These malicious files could contain worms or have the potential to infect the investigator's computer; if that happens, the attacker will keep an eye on all future investigations and the evidence they produce. With the help of this research and algorithm, whenever a malicious DLL or request is made, the algorithm will be able to identify it and flag it. This will save the investigator a lot of time because the investigator can upload files to his or her computer without worrying about whether they will be flagged as malicious behavior. We experimented wih multiple malicious files and our algorithm shows 98% efficacy.