Skip to main content

A library that provides an easy ready-to-use Recommendation engine.

Project description

Introduction

This library provides a pure Python interface for a Generalized Recommendation system.

Recommender Systems

Recommender Systems are algorithms aimed at suggesting relevant items to users (items being products to buy, movies to watch, text to read or anything else depending on industries). There exist two major categories of methods

  1. Content based methods
  2. Collaborative filtering methods

Content Based Recommendation

Content based approaches use additional information about users and/or items. The major idea of content based methods is to try to build a model, based on the available “features”, that explain the observed user-item interactions.

Collaborative Filtering Methods

Collaborative methods for recommender systems are methods that are based solely on the past interactions recorded between users and items in order to produce new recommendations.

Installing

pip install py-recommendation

Usage

Module Import

import py_recommendation as prc

Fetch the Input Data

Generate input data object which can be used for various types of recommendations available

Items Data

itemsInitialise = prc.ItemData()

Configure Data

# Reffer from sample data given below:
sample_items_data = {"title":['Place-A','Place-B','Place-C','Place-D','Place-E','Place-F','Place-G'],
            "tags":[
                'party & dance, nightlife, food, bar & restaurant, enjoy family',
                'Historic old monument, landmarks & enjoy$ beautiful',
                'Beach party, bar & nightlife @water, Beautiful',
                'Old Historic Temple, peaceful & rest, Family vacation',
                'Nature waterfall & green, beauty water',
                'Beautiful Himalaya hills, nature snow, water & peace',
                'Sights & Landmarks, Monuments & Statues'
            ]
           }
itemData = itemsInitialise.setData(
    data = sample_items_data,
    itemNameField = 'title',
    itemTagField = 'tags',
    )

Or else set data from file,

itemData = itemsInitialise.setData_file(
    filePath = 'path to the data file',
    itemName_col = 'column name carrying the item name/id',
    itemTag_col = 'column name carrying the item features',
    fileType = 'type of file (xlsx/csv)'
    )

Users Data

usersInitialise = prc.UsersData()

Configure Data

# Reffer from sample data given below:
sample_users_data = {"names":['Professor','Lisbon','Berlin','Nairobi','Helsinki','Rio'],
            "visited":[
                ['Place-B','Place-D'],
                ['Place-D','Place-E'],
                ['Place-B'],
                ['Place-C','Place-E'],
                ['Place-F'],
                ['Place-A', 'Place-F']
            ]
           }

usersData = usersInitialise.setData(
    data = sample_users_data,
    userNameField = 'names',
    triedItemField = 'visited',
    )

Recommeder based on Similar Items

This is used to find other items similar to the given item based on the features.

similarItemInit = prc.SimilarItem(itemData)
recommendation = similarItemInit.similarItem('Place-D')

Content-based Recommeder for specific users based on items tried

contentRecommendInit = prc.UserContent(itemData=itemData, usersData=usersData)
recommendation = contentRecommendInit.contentRecomend('Lisbon')
  • recommendation carries a list with each recommended items and their recommendation score on 100.
  • recommendation might have items with negative score which implies the item is already tried. This scenario occurs only if you have less number of items in 'ItemData'

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

py_recommendation-0.1.5.tar.gz (7.6 kB view hashes)

Uploaded Source

Built Distribution

py_recommendation-0.1.5-py3-none-any.whl (9.8 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page