A WebApp having Music Player and Music Recommender System
- The Music Recommender System suggests the songs related to those searched and played by a user as well as those which are played by selecting from the user's hard disk.
- It is made using Node.js as the backend language with Python used for recommendation system
- The songs are recommended using the Python Machine Learning code. Python version 2.7 is used with Panda library used for dataframes
- IT MAKES USE OF THE API OF last.fm TO SHOW THE RESULTS OF A SEARCHED QUERY
- It contains an Offline Music Player which can be used to play songs by selecting them from the client's system
- Python version 2.7 for Machine Learning code
- Node.js installed with some core modules : body-parser, ejs, express, express-session, fs, mongoose, python-shell
- MongoDb for data storage of the users. In this probject MongoLab is used.
- Change the PythonPath variable in the process/index.js (Line 133 and Line 170) file to the python executable to run the Python Machine Learning code
- Change the Mongo connection URL in the config/connectMongo.js (Line 9) file to the url of the MongoDb connection
- Change the Mongo username and password in the config/database.js file for the mongo connection
- The Music Recommender System needs the data for the song for which suggestions are to given. The data is stored in the files process/online.csv and process/offline.csv
- The names of a same song stored on different users' disks may have different names. So the recommendations may not be efficient.