PROJECT-I

SONG RECOMMENDER SYSTEM

TEAM MEMBERS: NIHAL GUPTA

20SCSE1180130
SHAMBHAVI PAL
20SCSE1180083

MENTOR: MOHD ANAS

 Objective
• In this project, we would be combining multiple services and open-source tools to make a
Chatbot that recommends songs based on the tone of the conversation which the user is
having with choosing options.

In this project, we would be building an extensive option service, to which you can talk to.
And talking to a chatbot wouldn't be business-driven. It would just be casual conversations.
Further, on top of it, the chatbot would also be recommending songs to the user based on the
tone of the user. This song recommendation feature employs the use of Last.fm API, very much
similar to the popular Spotify API. Also for tone/emotion analysis of the conversation we will be
using the IBM Tone Analyzer API.
Product Architecture
 Task 1

Create a web page for styling with CSS on notepad or VS Code

Task 2
Product Architecture

High-Level Approach
• User starts the conversation
• Select the language in which you want to listen the song
• Select the singer from the options given
• Based on your selection, the recommender system transfer
you to Youtube to that song
• The system will also take picture of your emotion
after all the selections done
Primary goals
• Setting up an open-source project locally and handling the errors being faced
• Using multiple services to build up a new service over them.
• Having a real-world recommender, which gives result based on your selection person and enjoying the music
recommended by the system.
Outcome
By the end of this milestone, you would have a working song recommender system with a good UI that
recommends songs to the user along with choosing option.
User based collaborative filtering
Item based collaborative filtering)
 Decision Tree

Decision tree is commonly and easy to implement algorithm used all along. The Decision
trees can be used both for the classification as well as regression of the problems. The
Decision trees frequently copies and then follows the human standard of thinking so that it is
simple to understand and make good interpretations of the given set of data . A decision tree
effectively and efficiently makes us understand the concept behind for the data so as to
interpret it and develop and produce optimum output. This means that a decision tree is
defined as a tree in which each node will represent an attribute or a feature, also each of the
links will depict a decision and each leaf will represent an outcome.
 Result

We have tested all the models (used algorithms) with the help of confusion matrix .Graph for
real positive rate against the real negative rate are prepared and model prepared with the help
of random forest was having the highest true positive against true negative rate means its true
positive rate was highest .This was the most effective model of all in predicting songs of
users taste and learning rate of random forest was highest for music recommender system.
Hybrid approach of collaborative filtering and content based filtering also had great learning
rate and accuracy
Thank you