Credit where it was earned
More than thirty student names appear in my author lists. This page credits them by project, and shows the next cohort what working with me actually looks like.
Projects that became papers
10 projects, 31 people credited alongside me. Names marked with a dot are senior collaborators rather than project students. Every title links to the paper it produced.
AI-Driven Metadata Extraction from Scientific Texts for Enhanced Biomedical Data Reusability
V Thangaraju
AI-Driven Solar Panel Optimization during Cloudy times Using Real-Time Image Processing
SK Radhakrishnan, V Mohandoss, K Moorthy, A Zaahrah
Eco-Transformers: Carbon-Efficient and Accelerated Inference through Self-Pruning and Activation Freezing in Large Language Models
E Shriaarthy ·, RS Niranjana, R Rohini, B Sivanesan and a co-author the record does not name
Automated real-time infection detection in oyster mushroom using smart farming system
Best Paper Award, Data Learning Track, IConSCEPT 2024, NIT Puducherry
N Kumar, T Nisanth, S Sibi
Extractive Text Summarization Using Two-Layered RBM
P Kavya, N Dhaksana, A Harshini, A Maria Nivetha
Sentiment analysis on movie reviews: a comparative analysis
B Sangeetha ·, DT Goutham, N Vaibhav Ram
Hand Gesture recognition using Deep Learning
B Srivathsan, R Gunaseelan
Fake Review Detection using Deep Learning
Sangeetha B ·, Raja Kumar I, Shevannth R
Fake news detection using supervised learning techniques
B Sangeetha ·, D Sasikumar, M Gayathri, G Sriram, K Nelson and a co-author the record does not name
On-Device Application Categorization using Application Resources
R Inturi, A Srikanth, T Hariprasath, P Abishavarthana and a co-author the record does not name
Funded work
Still to come: the team on this one by name, and where the alumni above went next. I would rather ask each of them first than publish it from memory.
Data we released
The part of a student project that outlives it. If no public dataset exists for your problem, building one is the contribution.
Awards this work won
If you want to be on this page
Read the guide on what a student team can actually build in a semester. It is the method behind every project above: pick a problem where you can get the first hundred examples yourself, baseline before you reach for the biggest model, and measure something beyond accuracy.
Then write to me through the student path on work with me. Name a paper here and the question it left you with.