Picasso Library - But For Books

View the streamlit app here

Tools used:

1. StreamLit
2. Pandas
3. Python
4. Matplotlib
5. Seaborn

What are Recommendation Systems

Recommendation System is an algorithm that can suggest products based on user or item activities.

Types:

• Content-based - These systems are based on the characteristics of the items themselves
• Collaborative Filtering - These systems use a collection of user ratings of items to make recommendations. Users with similar interests will receive recommendations based on previous users viewing.
• Hybrid - These systems use the content-based and collaborative filtering to form a more powerful recommendation system.

Stages of this project

1. Obtain the dataset
2. Analysis
3. Preprocessing of the data
4. Modelling
5. Deployment

Data Source

The data was sourced from Kaggle

Modelling

Using the NearestNeighbors algorithm, a cluster was formed that tagged each book or author to a particular cluster that enabled suggestion of books or authors according to the distance.