Apriori Algorithm implementation in TypeScript / JavaScript.
This implementation does not generate k-candidates as efficiently as it possibly could, as it adopts a brute-force approach: Every k-itemset is considered as a potential candidate, and an additional step is required to prune the unnecessary ones. More information about this question here.
The Apriori Algorithm is a great, easy-to-understand algorithm for frequent-itemset mining. However, faster and more memory efficient algorithms such as the FPGrowth Algorithm have been proposed since it was released.
If you need a more efficient frequent-itemset mining algorithm, consider checking out my implementation of the FPGrowth Algorithm.
This is a Node.js module available through the npm registry.
Installation is done using the npm install
command:
$ npm install --save node-apriori
In your TypeScript project, import and use Apriori
as follows. Same example with a JavaScript syntax is available here.
import { Apriori, Itemset, IAprioriResults } from 'node-apriori';
let transactions: number[][] = [
[1,3,4],
[2,3,5],
[1,2,3,5],
[2,5],
[1,2,3,5]
];
// Execute Apriori with a minimum support of 40%. Algorithm is generic.
let apriori: Apriori<number> = new Apriori<number>(.4);
// Returns itemsets 'as soon as possible' through events.
apriori.on('data', (itemset: Itemset<number>) => {
// Do something with the frequent itemset.
let support: number = itemset.support;
let items: number[] = itemset.items;
});
// Execute Apriori on a given set of transactions.
apriori.exec(transactions)
.then( (result: IAprioriResults<number>) => {
// Returns both the collection of frequent itemsets and execution time in millisecond.
let frequentItemsets: Itemset<number>[] = result.itemsets;
let executionTime: number = result.executionTime;
});
- Clone this repository:
git clone https://github.com/alexisfacques/Node-Apriori.git cd Node-Apriori
- Install the project's dependencies with
npm
:npm install
- Compile the project's sources to NodeJS executable JavaScript:
npm run tsc
- This should allow you to execute the given example as follows:
npm test
This project is licensed under the MIT License - see the LICENSE file for details