tigo / recommendation
collaborative filtering recommender systems
Fund package maintenance!
tigoCaval
www.paypal.com/donate?hosted_button_id=PPDESEV98R8KS
www.paypal.com/donate?hosted_button_id=S7FBV5N6ZTRXQ
Installs: 7 775
Dependents: 0
Suggesters: 0
Security: 0
Stars: 78
Watchers: 3
Forks: 14
Open Issues: 2
Requires
- php: >=7.0
Requires (Dev)
- phpunit/phpunit: ^9.5
README
Collaborative filtering recommendation system
- Ranking algorithm using likes / dislikes or star-based rating
- This package can be used in any PHP application or with any framework.
- Download package:
composer require tigo/recommendation
- MIT license. Feel free to use this project. Leave a star ⭐ or make a fork !
If you found this project useful, consider making a donation to support the developer.
Getting started
Starting with composer
- Install composer
- Download package:
composer require tigo/recommendation
- PHP >= 7.0; Versions that have been tested: 7.2.25, 7.3.23 e 8.0.1.
//Somewhere in your project, you may need to use autoload include __DIR__ ."/vendor/autoload.php";
Algorithms
- ranking
- euclidean
- slope one
Introduction
Recommend a product using collaborative filtering
/** $table gets the array from the database. $user is the foreign key that represents the user who will receive the recommendation. **/ use Tigo\Recommendation\Recommend; // import class $client = new Recommend(); $client->ranking($table,$user); //optional third parameter refers to the score not accepted $client->euclidean($table,$user); //optional third parameter refers to the minimum accepted score $client->slopeOne($table, $user); //optional third parameter refers to the minimum accepted score
Configuration
Sometimes, it may be necessary to rename the value of the constants (According to your database table).
- Configure: standard key (Directory:
./src/configuration/StandardKey.php
)
const SCORE = 'score'; //score const PRODUCT_ID = 'product_id'; //Foreign key const USER_ID = 'user_id'; //Foreign key
Example
A simple didactic demonstration of the algorithm
/** Example using "rating: liked and disliked" like: score = 1; dislike: score = 0 **/ $table = [ ['product_id'=> 'A', 'score'=> 1, 'user_id'=> 'Pedro' ], ['product_id'=> 'B', 'score'=> 1, 'user_id'=> 'Pedro' ], ['product_id'=> 'A', 'score'=> 1, 'user_id'=> 'João' ], ['product_id'=> 'B', 'score'=> 1, 'user_id'=> 'João' ], ['product_id'=> 'C', 'score'=> 1, 'user_id'=> 'João' ] ]; use Tigo\Recommendation\Recommend; // import class $client = new Recommend(); print_r($client->ranking($table,"Pedro")); // result = ['C' => 2] print_r($client->ranking($table,"Pedro",1)); // result = []; print_r($client->euclidean($table,"Pedro")); // result = ['C' => 1] print_r($client->euclidean($table,"Pedro", 2)); // result = [] ; print_r($client->slopeOne($table,'Pedro')); // result = ['C' => 1] print_r($client->slopeOne($table,'Pedro', 2)); // result = []
Supporting this project
If you are interested in supporting this project, you can help in many ways. Leave a star ⭐ or make a donation of any value.
Sponsor supporting this project
- []
Contributors
License
MIT license. See the archive License