Skip to content

Akshatha9108/Recommend_bot

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 

Repository files navigation

Recommend_bot

Steps before running the model In order to run this model Open the link below and download the two pickle (.pkl) files https://drive.google.com/drive/folders/1LsoiC_6zGXetBy8oPrbz0X2xZ9ik3I9T?usp=share_link so that the rest of the program can work as intended Steps for running the model Once downloaded place it in the root folder when running pog.py when you run the code in pog.py it should give a https link which once used in a browser should lead you to the CreAItive library There in the recommend tab you can use the recommendation model for books that are already in the dataset (which can be checked in the final_DB in the dataset_fixing folder)

The Dataset We have used 4 datasets in total for both the popular books and The recommendation bot . books_1.Best_Books_Ever.csv (https://zenodo.org/record/4265096) Users.csv (https://www.kaggle.com/datasets/arashnic/book-recommendation-dataset) Ratings.csv (https://www.kaggle.com/datasets/arashnic/book-recommendation-dataset) Books.csv (https://www.kaggle.com/datasets/arashnic/book-recommendation-dataset)

For the Popular books dataset we have combined both Books and ratings together and then eliminated books that have less than 200 ratings (referring to the number of ratings) And out of those we have chosen the top 12 highest books based on user ratings

For the recommendation bot we have combined both Books dataset and the books_1.Best_Books_Ever as we needed both the image of the books as well as the genre for a content based recommender After removing all unnecessary data we combined the names of authors , the description and the genre to create a column called tags which will be used for the comparison

Then we vectorize the data present in the tags using CountVectorizer so that they can be compared to each other and later the similarity will be judged based on the cosine similarity technique

Cosine similarity After comparing the books with each other based on the tags column we finally have a x dimentional space (where x is determined by the number of books) Based on the angle between the different books from the point of origin We select the books that are closest to each other and at the end output that

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 95.7%
  • HTML 2.2%
  • Python 2.1%