Arabic
[ class tree: Arabic ] [ index: Arabic ] [ all elements ]
Packages:
Arabic


Files:
Arabic.php
ArAutoSummarize.class.php
ArCharsetC.class.php
ArCharsetD.class.php
ArCompressStr.class.php
ArDate.class.php
ArGender.class.php
ArGlyphs.class.php
ArIdentifier.class.php
ArKeySwap.class.php
ArMktime.class.php
ArNormalise.class.php
ArNumbers.class.php
ArQuery.class.php
ArSoundex.class.php
ArStandard.class.php
ArStemmer.class.php
ArStrToTime.class.php
ArTransliteration.class.php
ArWordTag.class.php
EnTransliteration.class.php
Hiero.class.php
Salat.class.php

Classes:
Arabic
ArabicException
ArAutoSummarize
ArCharsetC
ArCharsetD
ArCompressStr
ArDate
ArGender
ArGlyphs
ArIdentifier
ArKeySwap
ArMktime
ArNormalise
ArNumbers
ArQuery
ArSoundex
ArStandard
ArStemmer
ArStrToTime
ArTransliteration
ArWordTag
EnTransliteration
Hiero
Salat

Class: ArAutoSummarize

Source Location: /sub/ArAutoSummarize.class.php

Class Overview




This PHP class do automatic keyphrase extraction to provide a quick  mini-summary for a long Arabic document




Author(s):






          
          


Copyright:


    
  
  • 2009 Khaled Al-Shamaa
    • 


        




Variables







Methods
























Class Details



[line 144]

This PHP class do automatic keyphrase extraction to provide a quick  mini-summary for a long Arabic document





Tags:





  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    copyright:  2009 Khaled Al-Shamaa
  

  

    link:  http://www.ar-php.org
  

  

    license:  LGPL
  










[ Top ]







Class Variables



	
	

	
$_commonChars = array('É','å','í','ä','æ','Ê','á','Ç','Ó','ã', 'e', 't', 'a', 'o', 'i', 'n', 's')

	
[line 149]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]


	
	

	
$_commonWords = array()

	
[line 153]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]


	
	

	
$_importantWords = array()

	
[line 154]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]


	
	

	
$_normalizeAlef = array('Ã','Å','Â')

	
[line 146]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]


	
	

	
$_normalizeDiacritics = array('ó','ð','õ','ñ','ö','ò','ú','ø')

	
[line 147]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]


	
	

	
$_separators = array('.',"\n",'¡','º','(','[','{',')',']','}',',',';')

	
[line 151]

  



Tags:





  

    access:  protected
  






  

	

  

    

      Type:  
      mixed
    

      

	



	
[ Top ]










Class Methods




  

	
	
constructor __construct [line 159]

	

    

    

		ArAutoSummarize __construct(
)
    

    


	
		Loads initialize values





Tags:





  

    access:  public
  









	
        
[ Top ]

  

  

	
	
method cleanCommon [line 480]

	

    

    

		string cleanCommon(
string
$str, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Extracting common Arabic words (roughly) from input Arabic string (document content)





Tags:





  

    return:  Arabic document as a string free of common words (roughly)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input normalized Arabic document as a string
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method doRateSummarize [line 311]

	

    

    

		string doRateSummarize(
string
$str, integer
$rate, string
$keywords, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Summarize percentage of the input Arabic string (document content) into output





Tags:





  

    return:  Output summary requested
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        integer  
        $rate  
        Rate of output summary sentence number as percentage                                of the input Arabic string (document content)
      

          

        string  
        $keywords  
        List of keywords higlited by search process
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method doSummarize [line 288]

	

    

    

		string doSummarize(
string
$str, integer
$int, string
$keywords, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Summarize input Arabic string (document content) into specific number of  sentences in the output





Tags:





  

    return:  Output summary requested
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        integer  
        $int  
        Number of sentences required in output summary
      

          

        string  
        $keywords  
        List of keywords higlited by search process
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method getMetaKeywords [line 384]

	

    

    

		string getMetaKeywords(
string
$str, integer
$int, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Extract keywords from a given Arabic string (document content)





Tags:





  

    return:  List of the keywords extracting from input Arabic string                (document content)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        integer  
        $int  
        Number of keywords required to be extracting from                                input string (document content)
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method highlightRateSummary [line 361]

	

    

    

		string highlightRateSummary(
string
$str, integer
$rate, string
$keywords, string
$style, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Highlight key sentences (summary) as percentage of the input string  (document content) using CSS and send the result back as an output.





Tags:





  

    return:  Output highlighted key sentences summary (using CSS)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        integer  
        $rate  
        Rate of highlighted key sentences summary number as                                percentage of the input Arabic string (document content)
      

          

        string  
        $keywords  
        List of keywords higlited by search process
      

          

        string  
        $style  
        Name of the CSS class you would like to apply
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method highlightSummary [line 336]

	

    

    

		string highlightSummary(
string
$str, integer
$int, string
$keywords, string
$style, [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Highlight key sentences (summary) of the input string (document content)  using CSS and send the result back as an output





Tags:





  

    return:  Output highlighted key sentences summary (using CSS)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        integer  
        $int  
        Number of key sentences required to be highlighted in                                the input string (document content)
      

          

        string  
        $keywords  
        List of keywords higlited by search process
      

          

        string  
        $style  
        Name of the CSS class you would like to apply
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

        

    


        
[ Top ]

  

  

	
	
method loadExtra [line 182]

	

    

    

		void loadExtra(
)
    

    


	
		Load enhanced Arabic stop words list





Tags:





  

    access:  public
  









	
        
[ Top ]

  

  

	
	
method _acceptedWord [line 661]

	

    

    

		boolean _acceptedWord(
string
$word)
    

    


	
		Check some conditions to know if a given string is a formal valid word or not





Tags:





  

    return:  True if passed string is accepted as a valid word else                  it will return False
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        string  
        $word  
        String to be checked if it is a valid word or not
      

        

    


        
[ Top ]

  

  

	
	
method _doNormalize [line 458]

	

    

    

		string _doNormalize(
string
$str)
    

    


	
		Normalized Arabic document





Tags:





  

    return:  Normalized Arabic document
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

        

    


        
[ Top ]

  

  

	
	
method _draftStem [line 507]

	

    

    

		string _draftStem(
string
$str)
    

    


	
		Remove less significant Arabic letter from given string (document content).

Please note that output will not be human readable.





Tags:





  

    return:  Output string after removing less significant Arabic letter                 (not human readable output)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

        

    


        
[ Top ]

  

  

	
	
method _minAcceptedRank [line 639]

	

    

    

		integer _minAcceptedRank(
array
$arr, integer
$int)
    

    


	
		Calculate minimum rank for sentences which will be including in the summary





Tags:





  

    return:  Minimum accepted sentence rank (sentences with rank more                  than this will be listed in the document summary)
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        array  
        $arr  
        Sentences ranks
      

          

        integer  
        $int  
        Number of sentences you need to include in your summary
      

        

    


        
[ Top ]

  

  

	
	
method _rankSentences [line 564]

	

    

    

		array _rankSentences(
array
$sentences, 
$stemmedSentences, array
$arr, array
$stemedSentences)
    

    


	
		Ranks sentences in a given Arabic string (document content).





Tags:





  

    return:  Two dimension array, first item is an array of document                sentences, second item is an array of ranks of document                sentences.
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        array  
        $sentences  
        Sentences of the input Arabic document as an array
      

          

        array  
        $stemedSentences  
        Stemmed sentences of the input Arabic document as an array
      

          

        array  
        $arr  
        Words ranks array (word as an index and value refer to                                the word frequency)
      

          

          
        $stemmedSentences  
        
      

        

    


        
[ Top ]

  

  

	
	
method _rankWords [line 523]

	

    

    

		hash _rankWords(
string
$str)
    

    


	
		Ranks words in a given Arabic string (document content). That rank refers  to the frequency of that word appears in that given document.





Tags:





  

    return:  Associated array where document words referred by index and               those words ranks referred by values of those array items.
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  protected
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

        

    


        
[ Top ]

  

  

	
	
method _summarize [line 206]

	

    

    

		string _summarize(
string
$str, string
$keywords, integer
$int, string
$mode, string
$output, [
$style = null], [string
$inputCharset = null], [string
$outputCharset = null], [object
$main = null])
    

    


	
		Core summarize function that implement required steps in the algorithm





Tags:





  

    return:  Output summary requested
  

  

    author:  Khaled Al-Shamaa <khaled.alshamaa@gmail.com>
  

  

    access:  public
  









	
        
Parameters:

    

    

          

        string  
        $str  
        Input Arabic document as a string
      

          

        string  
        $keywords  
        List of keywords higlited by search process
      

          

        integer  
        $int  
        Sentences value (see $mode effect also)
      

          

        string  
        $mode  
        Mode of sentences count [number|rate]
      

          

        string  
        $output  
        Output mode [summary|highlight]
      

          

        string  
        $inputCharset  
        (optional) Input charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set input charset)
      

          

        string  
        $outputCharset  
        (optional) Output charset [utf-8|windows-1256|iso-8859-6]                                default value is NULL (use set output charset)
      

          

        object  
        $main  
        Main Ar-PHP object to access charset converter options
      

          

          
        $style  
        
      

        

    


        
[ Top ]

  






        

		    

		    Documentation generated on Fri, 12 Mar 2010 01:01:42 +0300 by phpDocumentor 1.4.0