Kaggle Projects

---

CMI - Detect Behavior with Sensor Data

(Predicting Body Focused Repetitive Behaviors from a Wrist-Worn Device)

• Objective: To develop a predictive model capable of distinguishing (1) BFRB-like gestures from non-BFRB-like gestures and (2) the specific type of BFRB-like gesture.
• Data: Dataset contains sensor recordings taken while participants performed 8 BFRB-like gestures and 10 non-BFRB-like gestures while wearing the Helios device on the wrist of their dominant arm.
• Model:
• Result: Achieved an accuracy of 85.27 % on test data.

NeurIPS - Open Polymer Prediction 2025

(Predicting polymer properties with machine learning to accelerate sustainable materials research.)

• Objective: To predict five key chemical properties derived from molecular dynamics simulation: glass transition temperature (Tg), fractional free volume (FFV), thermal conductivity (Tc), polymer density, and radius of gyration (Rg).
• Data: provided with a polymer’s structure as a simple text string (SMILES)
• Model:
• Result: