HarmonyHub
所属分类:HarmonyOS
开发工具:Jupyter Notebook
文件大小:0KB
下载次数:0
上传日期:2024-02-21 20:05:39
上 传 者:
sh-1993
说明: 和谐中心(HarmonyHub)
(HarmonyHub)
文件列表:
ModelTraining.ipynb
Recommend.py
dataframebinaryfile.pkl
similaritybinaryfile.pkl
# Spotify Song Recommendation System
## Overview
This project aims to build a song recommendation system using natural language processing techniques and cosine similarity on song lyrics. The system is built using Python, pandas, scikit-learn, and NLTK libraries.
## Usage
1. **Data Preparation**:Load the dataset containing song information, including artist, song title, and lyrics.
2. **Data Cleaning**: Clean the text data by converting it to lowercase and removing unwanted characters.
3. **Tokenization and Stemming**: Tokenize the lyrics and apply stemming using NLTK.
4. **TF-IDF Vectorization**: Convert the tokenized lyrics into TF-IDF features.
5. **Cosine Similarity**: Calculate cosine similarity between songs based on their TF-IDF vectors.
6. **Song Recommendation**: Provide recommendations for similar songs based on a given song.
## Serialization
The similarity matrix and DataFrame are serialized into binary files for later use.
## Instructions on how to execute the files:
1. Download the dataset from this link [https://www.kaggle.com/datasets/notshrirang/spotify-million-song-dataset/].
2. Run the Commands that are in ModelTraining.ipynb in Anaconda Navigator and it will generate two binary files.
3. Run the MRS.py file using command ‘streamlit run Recommend.py’ in terminal.
### GitHub Repository
The Python script used for analysis can be found in this [GitHub repository](https://github.com/VedaSamhitha2810/HarmonyHub/blob/master/ModelTraining.ipynb). Feel free to explore the script and dataset for further insights.
近期下载者:
相关文件:
收藏者: