Evaluating English word difficulty using word frequency data

Installation

cd ~/my-project/
npm install difficulty --save

Basic Usage

Async/Await

Welcome to ES7! ⎝( OωO)⎠

import { create } from 'difficulty';

(async () => {
  try {
    const difficulty = await create();
    const a = difficulty.getLevel('apple');
    const b = difficulty.getLevel('cappuccino');

    console.log(`apple is level ${a}, easy!`);
    console.log(`cappuccino is level ${b}, too hard!`);
  } catch(err) {
    console.error(err);
  }
})();

Output

apple is level 0, easy!
cappuccino is level 3, so hard!

Old schooler?

const Difficulty = require('difficulty');

// Using Promise
Difficulty.create().then((difficulty) => {
  ...
});

// Sync is available too!
const difficulty = Difficulty.createSync();

API

Difficulty.create(options)

filepath

Path to your csv file, in following format.

"Word","Freq_HAL"
"a","10610626"
...

Default: ./node_modules/difficulty/wordlist/wordlist.csv

levelsThreshold

Define your own levels with frequency threshold in descending order.

Default: [20000, 10000, 5000]

- Level 0: freq > 20000
- Level 1: 10001 to 20000
- Level 2: 5001 to 10000
- Level 3: < 5000 or **non exist words**

License

ISC

Important

The default word frequency data is generated from English Lexicon Project, which is for non-commercial research purposes only and may not be used in the development of speech technology. You should consider using your own wordlist for other purpose.

Author

github/auphone

Good luck! σ`∀´)σ