Music Recommendation System in Machine Learning

Music Recommendation System in Machine Learning

HTML

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.neighbors import NearestNeighbors
 
# Load the dataset
data = pd.read_csv('music_ratings.csv')
 
# Split the dataset into training and testing sets
train_data, test_data = train_test_split(data, test_size=0.2)
 
# Create a Nearest Neighbors model
knn_model = NearestNeighbors(metric='cosine', algorithm='brute')
knn_model.fit(train_data.iloc[:, 1:])  # Assuming the first column contains user IDs
 
# Function to recommend songs based on user input
def recommend_songs(input_songs, k=5):
    distances, indices = knn_model.kneighbors(input_songs)
    recommended_songs = []
    for index in indices.flatten():
        recommended_songs.append(data.iloc[index]['song_name'])
    return recommended_songs[:k]
 
# Example usage
user_input = [[3, 4, 5, 2, 1]]  # Example user input with ratings for each song
recommended = recommend_songs(user_input, k=3)
print("Recommended songs:")
for song in recommended:
    print(song)
 

In this example, we start by loading the music ratings dataset from a CSV file. We then split the data into training and testing sets using train_test_split from scikit-learn. Next, we create a NearestNeighbors model with cosine similarity as the distance metric. We fit the model with the training data, excluding the user IDs. The recommend_songs function takes user input in the form of ratings for each song and returns a list of recommended songs. It uses the trained model to find the nearest neighbors based on the user input and retrieves the song names associated with those neighbors. Finally, we provide an example usage where we pass a user's input ratings and specify the number of songs to recommend (k). The recommended songs are printed as output. Note that this is a basic example, and you can further enhance the recommendation system by using more advanced techniques, such as collaborative filtering or matrix factorization. Additionally, you may want to consider incorporating user feedback and continuously updating the model to improve recommendations over time.