Dear Padmaja Buddi

These blogs and articles might be helpful, have a look:

1. https://machinelearningmastery.com/how-to-develop-rnn-models-for-human-activity-recognition-time-series-classification/

2. https://machinelearningmastery.com/how-to-load-and-explore-a-standard-human-activity-recognition-problem/

3. Article Feature learning for Human Activity Recognition using Convol...

4. https://link.springer.com/article/10.1007/s11042-020-09004-3

Kind Regards

Qamar Ul Islam

It happens when a model learns the detail and noise in the training data to the extent that it negatively impacts the performance of the model on new data.

You may encounter over-fitting which may impede accurate identification of unseen features.