README.md
# Keyword Extractor
A package to extract keywords from text
[](https://packagist.org/packages/keyword-extractor/keyword-extractor)
[](https://travis-ci.org/iranianpep/keyword-extractor)
[](https://scrutinizer-ci.com/g/iranianpep/keyword-extractor/build-status/master)
[](https://codeclimate.com/github/iranianpep/keyword-extractor)
[](https://codeclimate.com/github/iranianpep/keyword-extractor/coverage)
[](https://scrutinizer-ci.com/g/iranianpep/keyword-extractor/?branch=master)
[](https://scrutinizer-ci.com/g/iranianpep/keyword-extractor/?branch=master)
[](https://codeclimate.com/github/iranianpep/keyword-extractor)
[](https://packagist.org/packages/keyword-extractor/keyword-extractor)
[](https://styleci.io/repos/152369408)
[](https://www.codacy.com/app/iranianpep/keyword-extractor?utm_source=github.com&utm_medium=referral&utm_content=iranianpep/keyword-extractor&utm_campaign=Badge_Grade)
[](https://packagist.org/packages/keyword-extractor/keyword-extractor)
[](https://raw.githubusercontent.com/iranianpep/keyword-extractor/master/LICENSE)
## Server Requirements
- PHP >= 7.3
## Usage
- To install ths package:
```
composer require keyword-extractor/keyword-extractor
```
- Extract the keywords:
```
$keywordExtractor = new KeywordExtractor();
$text = 'This is a simple sentence.';
$result = $keywordExtractor->run($text);
```
The result with the default modifiers and no sorting values will be:
```
Array
(
[simpl] => Array
(
[frequency] => 1
[occurrences] => Array
(
[0] => Array
(
[ngram] => simple
[indexes] => Array
(
[0] => 3
)
)
)
)
[sentenc] => Array
(
[frequency] => 1
[occurrences] => Array
(
[0] => Array
(
[ngram] => sentence.
[indexes] => Array
(
[0] => 4
)
)
)
)
)
```
Currently, the default modifiers are as follow (they will be applied to the tokens in order):
```
[
new EmailFilter(),
new PunctuationFilter(),
new WhitelistFilter($this->getWhitelist()),
new BlacklistFilter($this->getBlacklist()),
new StopWordFilter(),
new NumberFilter(),
new StemFilter(),
// run the blacklist even after stemming too
new BlacklistFilter($this->getBlacklist()),
]
```
Obviously, you can set your own modifiers:
```
$keywordExtractor->setModifiers([new PunctuationFilter()]);
```
Also, whitelist can be used as follow:
```
$keywordExtractor = new KeywordExtractor();
$text = 'This is a simple sentence and simple sentence.';
$keywordExtractor->setWhitelist(['simple']);
$result = $keywordExtractor->run($text);
```
Which results in:
```
Array
(
[simple] => Array
(
[frequency] => 2
[occurrences] => Array
(
[0] => Array
(
[ngram] => simple
[indexes] => Array
(
[0] => 3
)
)
[1] => Array
(
[ngram] => simple
[indexes] => Array
(
[0] => 6
)
)
)
)
[sentenc] => Array
(
[frequency] => 2
[occurrences] => Array
(
[0] => Array
(
[ngram] => sentence
[indexes] => Array
(
[0] => 4
)
)
[1] => Array
(
[ngram] => sentence.
[indexes] => Array
(
[0] => 7
)
)
)
)
)
```
Blacklist can also be used in the same way as whitelist:
```
$keywordExtractor = new KeywordExtractor();
$text = 'This is a simple sentence.';
$keywordExtractor->setBlacklist(['simple']);
$result = $keywordExtractor->run($text);
```
The result is:
```
Array
(
[sentenc] => Array
(
[frequency] => 1
[occurrences] => Array
(
[0] => Array
(
[ngram] => sentence.
[indexes] => Array
(
[0] => 4
)
)
)
)
)
```
To sort by frequency in descending order:
```
$keywordExtractor->run($text, Sorter::SORT_BY_FREQUENCY, Sorter::SORT_DIR_DESC);
```
To sort by min occurrences distance:
```
$text = 'sentence and sentence';
$result = $this->keywordExtractor->run($text, Sorter::SORT_BY_MIN_OCCURRENCE_DISTANCE);
Array
(
[sentenc] => Array
(
[frequency] => 2
[occurrences] => Array
(
[0] => Array
(
[ngram] => sentence
[indexes] => Array
(
[0] => 0
)
)
[1] => Array
(
[ngram] => sentence
[indexes] => Array
(
[0] => 2
)
)
)
[minOccurrencesDistance] => 1
)
)
```