This project was done under the supervision of Prof. Dr. Veena Bansal, Department of Industrial & Management Engineering, IIT Kanpur.
This project titled ”Yoga Asana Library.” covers the basic understanding of Machine learning (ML) for natural language processing (NLP) and text analytics to understand the meaning of text documents. Machine learning for NLP and text analytics involves techniques for identifying parts of speech, entities, sentiment, and other aspects of a text.
The main goal was to build a model which can mainly suggest suitable Yoga poses that the user can primarily adopt in his/her daily routine to provide the required set of benefits the user wants. The model must predict the Asanas taking care of user-defined constraints, including benefits, age, contraindication, and Level of Pose.
Datasets from suggested books were collected and implemented pre-processing steps like
Tokenization, Stop words removal, and Regular Expressions using the python gensim
library. For computers to use language for predictive models and understand words, the
words must be translated into numbers. We analyzed the importance of famous word
vectorization techniques like One hot encoding, Word2Vec, and TF-IDF to get the word
embeddings. At last, Artificial Neural Network was built to meet the project purpose.
This Project was done in collaboration with one of my group partner named Kshitij Kaithal.
