Research

Research & publications

Exploring applied machine learning, intelligent systems, sensor data, and software-led research workflows.

Publications

Selected work

June 2025Research Work

Smart Phone Sensor Data Fusion: A Joint Learning Approach to Activity Recognition

A deep learning system for smartphone-based human activity recognition using accelerometer and gyroscope data, comparing a Full Transformer model with a proposed Joint Learning fusion architecture.

Best accuracy
98%
Best F1 score
98%
Dataset
UCI-HAR
Human Activity RecognitionSensor FusionCNN-LSTMTransformers
View research98%
June 2025Research Work

Early Prediction of Down Syndrome Using Deep Transfer Learning-Based Approaches

A transfer learning study for early Down Syndrome screening from facial images using multiple pre-trained CNN architectures and targeted fine-tuning.

Dataset size
~3,000 images
Best model
Xception
Split
75 / 25
Deep LearningTransfer LearningMedical ImagingCNN
View research~3,000 images
July 2024Research Work

Non-contact Inspection of Electrically Discharged Materials Using Machine Learning

A machine learning approach for predicting surface roughness in EDM-machined materials using process parameters instead of contact-based inspection.

Best model
KNN
R2 score
~0.999
Augmented samples
9,300
Machine LearningManufacturingEDMRegression
View researchKNN