https://github.com/bibeklakra91/recommendation-system

https://github.com/bibeklakra91/recommendation-system

Last synced: 11 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/bibeklakra91/recommendation-system
• Owner: BibekLakra91
• License: mit
• Created: 2024-01-24T08:32:35.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2024-01-24T10:14:49.000Z (10 months ago)
• Last Synced: 2024-01-24T11:04:50.851Z (10 months ago)
• Language: Jupyter Notebook
• Size: 97.7 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        

# Personalised Recommendation System for a Content Feed

This repository contains code for building a recommendation system after finding similarity measurements using cosine similarity. The model undergoes training using 70% of the provided dataset.

## Data

The dataset used for this recommendation system is provided in the "Assignment - ML - Supporting Data.xlsx" file. It includes information about items, user interactions, and other relevant features.

## Requirements

Make sure you have the required libraries installed. You can install them using the following command:

```bash

pip install pandas numpy scikit-learn

```

**OR**

```bash

conda install pandas numpy scikit-learn

```

## Running the Code

1. Open the Jupyter Notebook `Recommendation_System.ipynb`.

2. Run the cells in the notebook to load the dataset, preprocess features, and train the recommendation system.

3. The system calculates cosine similarity between items and predicts interactions based on a specified threshold.

4. The accuracy of the recommendation system is evaluated using the test set.

## Code Structure

- `Recommendation_System.ipynb`: Jupyter Notebook containing the code for building and testing the recommendation system.

- `README.md`: Documentation file providing information about the project, data, requirements, and how to run the code.

## Important Functions

- `important_features(dataset)`: A function to preprocess features in the dataset by combining 'tags' and 'hashtags' into a single 'feature' column.

## Results

The recommendation system achieved an accuracy of 98% on the test set. Adjustments to the similarity threshold and interaction rates can be made based on specific requirements.