Shalini Sivasamy

Followers

Following

Co-author

Public Views

With over 13 years of experience in IT, specializing in Aritifical Intelligence, Machine Learning and Natural Language Processing. Demonstrated success in leading and supporting the research, development, and delivery of large-scale Data Science and Natural Language Processing initiatives that guide business decisions and improve operational performance.

Throughout my career, I have excelled at motivating and leading diverse teams, consistently meeting and exceeding project objectives while fostering collaboration and innovation. My ability to build strong rapport with clients, stakeholders, and team members has been pivotal in delivering impactful solutions that align with organizational goals.

My technical skillset spans cutting-edge AI and deep learning methodologies, enabling me to design and implement sophisticated predictive models, natural language processing (NLP) systems, and computer vision solutions. By combining advanced technical expertise with strategic thinking, I have successfully turned raw data into actionable insights, driving efficiencies and unlocking new opportunities for growth.

less

Related Authors

University Malaysia of Computer Science and Engineering

IJETRM Journal

babak Safaei

Tsinghua University

Rekha Sugandhi

Savitribai Phule Pune University

Luca Romeo

University of Macerata

Ang Mei Choo

Universiti Kebangsaan Malaysia

Gustavo de Assis Costa

Instituto Federal de Educação, Ciência e Tecnologia de Goiás - Campus Goiânia

Uploads

Papers by Shalini Sivasamy

AI-Driven Predictive Maintenance in Smart Manufacturing: Enhancing Efficiency through Deep Learning Models

Journal of Information Systems Engineering and Management, 2025

Tool wear monitoring and predictive maintenance are critical in manufacturing, where traditional ... more Tool wear monitoring and predictive maintenance are critical in manufacturing, where traditional methods often struggle to adapt to changing conditions. This research presents an Adaptive Reinforcement Learning Framework for Real-Time Tool Wear Optimization and Predictive Maintenance (ARTOM). The framework integrates reinforcement learning with realtime sensor feedback to optimize machining parameters and maintenance schedules dynamically. Proximal Policy Optimization (PPO) is used to guide decision-making by balancing tool life, product quality, and operational costs. Multi-agent reinforcement learning divides tasks among agents to handle diverse machining scenarios, while sliding window techniques and dimensionality reduction ensure efficient data processing. The study has used the benchmark dataset, which include time-series sensor data and machining parameters. Metrics potential metrics have been used to evaluate prediction accuracy, while runtime and memory usage assess computational efficiency. Results has shown that ARTOM consistently achieves lower prediction errors and faster execution times than contemporary baseline models. These findings demonstrate ARTOM's ability to adapt to different tool conditions and improve operational decision-making.

DownloadEdit

AI-Driven Medical Chatbot for Predicting and Managing Infectious Diseases

IJARSCT, 2025

This paper explores the potential of chatbots to enhance the medical field by combating infectiou... more This paper explores the potential of chatbots to enhance the medical field by combating
infectious diseases, including respiratory illnesses, zoonotic diseases, and viral outbreaks such as COVID-
19. By increasing awareness among users, these chatbots can help individuals discover medical solutions to
prevent and manage these conditions. We have developed a training model that facilitates better human
interaction with databases through natural language processing, tailored to characterize users effectively.
Our proposed AI chatbot model employs a recurrent neural network to ensure efficient interaction and
prediction, addressing current deficiencies in guidelines for improving lifestyle programs. The model
achieves a minimum loss of 0.112 and a maximum accuracy of 93. Additionally, this paper investigates the
feasibility of implementing chatbots to offer 24/7 support, broaden healthcare services, and deliver
personalized, real-time responses. Our conclusion emphasizes the capabilities, benefits, and challenges
faced by healthcare chatbots during pandemics. This research aims to guide the development of chatbot
technology, ensuring it remains innovative and effective in preventing infectious diseases, ranging from
viral infections like influenza to emerging global health threats..

DownloadEdit

Neural Architecture Search Based Deepfake Detection Model using YOLO

IJAR SCT, 2025

Deepfakes are intentionally created to disseminate false information or serve malicious purposes.... more Deepfakes are intentionally created to disseminate false information or serve malicious purposes. Detecting deepfakes has become increasingly difficult due to the advancing technology involved in their creation. This paper introduces a deep learning model based on Neural Architecture Search (NAS) that incorporates the You Only Look Once (YOLO) model for image segmentation and employs data augmentation to enhance the diversity of the dataset. The goal is to improve deepfake detection accuracy compared to current models. The study utilized the CelebDF v2 dataset, which includes 590 genuine videos and 5,639 deepfake videos. From this dataset, 100 deepfake and 100 real videos were chosen, and frames were extracted. After augmentation, the resulting dataset comprised 2,000 real and 2,000 deepfake images. The proposed model attained a testing accuracy of 99.04% and performed exceptionally well across other evaluation metrics such as F1 score, precision, and recall.

DownloadEdit

Neural Architecture Search Based Deepfake Detection Model using YOLO

by Somnath Banerjee and Shalini Sivasamy

IJAR SCT, 2025

Deepfakes are intentionally created to disseminate false information or serve malicious purposes... more Deepfakes are intentionally created to disseminate false information or serve malicious
purposes. Detecting deepfakes has become increasingly difficult due to the advancing technology involved
in their creation. This paper introduces a deep learning model
based on Neural Architecture Search (NAS)
that incorporates the You Only Look Once (YOLO) model for image segmentation and employs data
augmentation to enhance the diversity of the dataset. The goal is to improve deepfake detection accuracy
compared to cur
rent models. The study utilized the CelebDF v2 dataset, which includes 590 genuine videos
and 5,639 deepfake videos. From this dataset, 100 deepfake and 100 real videos were chosen, and frames
were extracted. After augmentation, the resulting dataset compr
ised 2,000 real and 2,000 deepfake images.
The proposed model attained a testing accuracy of 99.04% and performed exceptionally well across other
evaluation metrics such as F1 score, precision, and recall
.

DownloadEdit

Shalini Sivasamy

Uploads

Papers by Shalini Sivasamy

Log In