minhashjs
v1.0.1
Published
MinHash and LSH in JavaScript (Datasketch implementation)
Downloads
19
Readme
MinHash and LSH in JavaScript (Datasketch implementation)
Install
npm i minhashjs
MinHash Usage
const MH = require('./MinHash.js');
first_sentence = ['minhash', 'is', 'a', 'data', 'structure', 'for',
'estimating', 'the', 'similarity', 'between']
second_sentence = ['minhash', 'is', 'a', 'data', 'structure', 'for',
'estimating', 'the', 'similarity', 'between']
// First, instanciate the MinHash object
m1 = new MH.MinHash()
m2 = new MH.MinHash()
// Then, update each MinHash
first_sentence.forEach((d, i) => m1.update(d));
second_sentence.forEach((d, i) => m2.update(d));
// You can now compute the jaccard distance between these two hashes
m1.jaccard(m2)
LSH Usage
const F = require('./ForestLSH.js');
const MH = require("./MinHash");
var data1 = ['minhash', 'is', 'a', 'probabilistic', 'data', 'structure', 'for',
'estimating', 'the', 'similarity', 'between', 'datasets']
var data2 = ['minhash', 'dog', 'data', 'structure', 'for',
'fork', 'twice', 'similarity', 'food', 'money']
var data3 = ['minhash', 'is', 'probability', 'data', 'structure', 'for',
'estimating', 'the', 'similarity', 'between', 'documents']
// Create MinHash objects
var m1 = new MH.MinHash(num_perm=128)
var m2 = new MH.MinHash(num_perm=128)
var m3 = new MH.MinHash(num_perm=128)
// Update the hashes
data1.forEach((d, i) => m1.update(d));
data2.forEach((d, i) => m2.update(d));
data3.forEach((d, i) => m3.update(d));
// Create a MinHash LSH Forest with the same num_perm parameter
var forest = new F.MinHashLSHForest(num_perm=128)
// Add m2 and m3 into the index
forest.add("m2", m2)
forest.add("m3", m3)
//IMPORTANT: must call index() otherwise the keys won't be searchable
forest.index()
// Using m1 as the query, retrieve top X keys that have the higest Jaccard
var X = 2
let r = forest.query(m1, X)
Run tests
npm run test
Run speed tests
npm run speed
🤝 Contributing
Contributions, issues and feature requests are welcome!Feel free to check issues page.