|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace Onoi\Tesa\Tests\Integration; |
|
4
|
|
|
|
|
5
|
|
|
use Onoi\Tesa\SanitizerFactory; |
|
6
|
|
|
|
|
7
|
|
|
/** |
|
8
|
|
|
* @group onoi-tesa |
|
9
|
|
|
* |
|
10
|
|
|
* @license GNU GPL v2+ |
|
11
|
|
|
* @since 0.1 |
|
12
|
|
|
* |
|
13
|
|
|
* @author mwjames |
|
14
|
|
|
*/ |
|
15
|
|
|
class CombinedSanitizerTextStopwordTest extends \PHPUnit_Framework_TestCase { |
|
16
|
|
|
|
|
17
|
|
|
/** |
|
18
|
|
|
* @dataProvider textByLanguageProvider |
|
19
|
|
|
*/ |
|
20
|
|
|
public function testByLanguage( $languageCode, $text, $expected ) { |
|
21
|
|
|
|
|
22
|
|
|
$sanitizerFactory = new SanitizerFactory(); |
|
23
|
|
|
|
|
24
|
|
|
$sanitizer = $sanitizerFactory->newSanitizer( $text ); |
|
25
|
|
|
$sanitizer->toLowercase(); |
|
26
|
|
|
|
|
27
|
|
|
$text = $sanitizer->sanitizeWith( |
|
28
|
|
|
$sanitizerFactory->newGenericRegExTokenizer(), |
|
29
|
|
|
$sanitizerFactory->newCdbStopwordAnalyzer( $languageCode ), |
|
30
|
|
|
$sanitizerFactory->newNullSynonymizer() |
|
31
|
|
|
); |
|
32
|
|
|
|
|
33
|
|
|
$this->assertEquals( |
|
34
|
|
|
$expected, |
|
35
|
|
|
$text |
|
36
|
|
|
); |
|
37
|
|
|
} |
|
38
|
|
|
|
|
39
|
|
|
public function textByLanguageProvider() { |
|
40
|
|
|
|
|
41
|
|
|
// https://en.wikipedia.org/wiki/Stop_words |
|
42
|
|
|
$provider[] = array( |
|
|
|
|
|
|
43
|
|
|
'en', |
|
44
|
|
|
// |
|
45
|
|
|
'In computing, stop words are words which are filtered out before or after processing of ' . |
|
46
|
|
|
'natural language data (text).[1] Though stop words usually refer to the most common words ' . |
|
47
|
|
|
'in a language, there is no single universal list of stop words used by all natural language ' . |
|
48
|
|
|
'processing tools, and indeed not all tools even use such a list. Some tools specifically avoid '. |
|
49
|
|
|
'removing these stop words to support phrase search.', |
|
50
|
|
|
// |
|
51
|
|
|
'computing stop words filtered processing natural language data text stop words refer common ' . |
|
52
|
|
|
'words language single universal list stop words natural language processing tools list tools ' . |
|
53
|
|
|
'specifically avoid removing stop words support phrase search' |
|
54
|
|
|
); |
|
55
|
|
|
|
|
56
|
|
|
// https://es.wikipedia.org/wiki/Palabra_vac%C3%ADa |
|
57
|
|
|
$provider[] = array( |
|
58
|
|
|
'es', |
|
59
|
|
|
// |
|
60
|
|
|
'Palabras vacías es el nombre que reciben las palabras sin significado como artículos, pronombres, ' . |
|
61
|
|
|
'preposiciones, etc. que son filtradas antes o después del procesamiento de datos en lenguaje natural ' . |
|
62
|
|
|
'(texto). A Hans Peter Luhn, uno de los pioneros en recuperación de información, se le atribuye la ' . |
|
63
|
|
|
'acuñación de la locución inglesa stop words y el uso del concepto en su diseño. Está controlada por ' . |
|
64
|
|
|
'introducción humana y no automática.', |
|
65
|
|
|
// |
|
66
|
|
|
'palabras vacías nombre que reciben palabras significado artículos pronombres preposiciones etc que ' . |
|
67
|
|
|
'son filtradas después del procesamiento datos lenguaje natural texto hans peter luhn pioneros ' . |
|
68
|
|
|
'recuperación información atribuye acuñación locución inglesa stop words del concepto diseño está ' . |
|
69
|
|
|
'controlada introducción humana automática' |
|
70
|
|
|
); |
|
71
|
|
|
|
|
72
|
|
|
// https://de.wikipedia.org/wiki/Stoppwort |
|
73
|
|
|
$provider[] = array( |
|
74
|
|
|
'de', |
|
75
|
|
|
// |
|
76
|
|
|
'Stoppwörter nennt man im Information Retrieval Wörter, die bei einer Volltextindexierung nicht beachtet ' . |
|
77
|
|
|
'werden, da sie sehr häufig auftreten und gewöhnlich keine Relevanz für die Erfassung des Dokumentinhalts ' . |
|
78
|
|
|
'besitzen.', |
|
79
|
|
|
// |
|
80
|
|
|
'stoppwörter nennt information retrieval wörter volltextindexierung beachtet häufig auftreten gewöhnlich ' . |
|
81
|
|
|
'relevanz erfassung dokumentinhalts besitzen' |
|
82
|
|
|
); |
|
83
|
|
|
|
|
84
|
|
|
// https://en.wikipedia.org/wiki/Query_expansion |
|
85
|
|
|
$provider[] = array( |
|
86
|
|
|
'en', |
|
87
|
|
|
// |
|
88
|
|
|
"The goal of query expansion in this regard is by increasing recall, precision can potentially increase " . |
|
89
|
|
|
"(rather than decrease as mathematically equated), by including in the result set pages which are more " . |
|
90
|
|
|
"relevant (of higher quality), or at least equally relevant. Pages which would not be included in the " . |
|
91
|
|
|
"result set, which have the potential to be more relevant to the user's desired query, are included, and " . |
|
92
|
|
|
"without query expansion would not have, regardless of relevance.", |
|
93
|
|
|
// |
|
94
|
|
|
"goal query expansion regard increasing recall precision potentially increase decrease mathematically " . |
|
95
|
|
|
"equated including result set pages relevant quality equally relevant pages included result set potential " . |
|
96
|
|
|
"relevant user desired query included query expansion relevance" |
|
97
|
|
|
); |
|
98
|
|
|
|
|
99
|
|
|
return $provider; |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
} |
|
103
|
|
|
|
onoi /
tesa
Adding an explicit array definition is generally preferable to implicit array definition as it guarantees a stable state of the code.
Let’s take a look at an example:
As you can see in this example, the array
$myArrayis initialized the first time when the foreach loop is entered. You can also see that the value of thebarkey is only written conditionally; thus, its value might result from a previous iteration.This might or might not be intended. To make your intention clear, your code more readible and to avoid accidental bugs, we recommend to add an explicit initialization $myArray = array() either outside or inside the foreach loop.