recommend-py 0.1.5

A library to do recommendations based on matrix

Recommend_py

Description

this is a library to do recommendations based on matrix

Functions

• Function 1: is_utility(df) input the matrix and return if it is a utility matrix
• Function 2: to_utility(df, id_col_num) input the matrix and the column number of the id and return the utility matrix
• Function 3: utility_normalize(df) input the utility matrix and return the normalized utility matrix
• Function 4: utility_standardize(df) input the utility matrix and return the standardized utility matrix
• Function 5: similarity_calculation(df, mean_same_item) input the utility matrix and the minimum number of same items and return the similarity matrix
• Function 6: recommend_calculation(df, top_n, mean_score) input the utility matrix, the number of top items to recommend and the minimum score a similar user must give for an item to be recommended and return the recommendation matrix

Installation

To install this project, clone the repository and install the required packages:

pip install pandas
pip install numpy
pip install scikit-learn

Usage

To use this project, import the library and call the functions:

import recommend_py

recommend_py.is_utility(df)
recommend_py.to_utility(df, id_col_num)
recommend_py.utility_normalize(df)
recommend_py.utility_standardize(df)
recommend_py.similarity_calculation(df, mean_same_item)
recommend_py.recommend_calculation(df, top_n, mean_score)