Personality matching and Recommendation
The repo holds the code for personality matching and clustering based on user-group characteristics. This project was done as part of LivSpaceHack.
Input data is expected to be the user-replies for a set of question ranging in number from 1 to 21. Based on the replies for few questions, we predict the answers to other questions based on similarity index mapping of users. Once, we get the characteristics of the user, the next part is clustering and identifying the user-group of any given user. Then, comes the recommendation part, where images with certain tags are recommended to the user based on the user-group he is presented in.
Cheers :)