• Abhigynaa.S
• Prakruthi.VS
• Ananya.R
• Nithya shree.KS
• Srinidhi Gowda

This project develops a human activity recognition (HAR) system using a Hidden Markov model (HMM). It classifies activities such as walking, running, sitting, and jumping using accelerometer and gyroscope data. The Baum-welch algorithm trains the model, and the Viterbi algorithm performs activity prediction. Applications include healthcare monitoring, fitness tracking, and smart home automation.

1. *Hidden Markov Model (HMM)* is a statistical model used to represent time series data with hidden states and observed events — ideal for modeling human activities. 2. *Sensor data* (e.g., from accelerometers, gyroscopes) is collected from wearable devices to recognize physical movements. 3. *Training phase* involves learning the probabilities of transitions between activities and the likelihood of observed signals. 4. *Inference/Prediction phase* uses algorithms like *Viterbi* to decode the most probable sequence of activities from observed data.

" "Hidden macro model based on activity recognition system"", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.a375-a379, July-2025, Available :http://www.jetir.org/papers/JETIR2507039.pdf

" "Hidden macro model based on activity recognition system"", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. ppa375-a379, July-2025, Available at : http://www.jetir.org/papers/JETIR2507039.pdf