Nlp-python Natural Language Processing 01

1  #-*-coding:utf-8-*-2 """3 Created on Wed Sep 6 22:21:094 5 @author: Administrator6 """7 ImportNLTK8  fromNltk.bookImport*9 #Search for WordsTenText1.concordance ("Monstrous")#Find Keywords One  A #search for similar words -Text1.similar ('Monstrous') -  the #Search the common context -Text2.common_contexts (['Monstrous','very']) -  -  + #the distribution of words -Text4.dispersion_plot (['moustrous','very']) +  A #the length of the vocabulary at Len (TEXT3) -  - #Repetition word density -Len (TEXT3)/Len (Set (TEXT3)) -  - #keyword density inText3.count ('smote') -100*text4.count ('a')/Len (TEXT4) to  + deflexical_diversity (text): -     returnLen (text)/Len (set (text)) the  * defpercentage (count, total): $     return* Count/ TotalPanax Notoginseng  -  the  +sent1=['Pager','Me','Ishmael','.'] A  the #Gets the text word index, noting the length of the index, starting from zero +text3[172] -  $Text3.index (' Love') $  - #frequency distribution, the judgment of common words - #simple statistics, frequency distribution theFdist1 =freqdist (Text1) - WuyiVocabulary1 =Fdist1.keys () thefdist1['Whale'] -Fdist1.plot (cumulative=True) Wu  - #Low Frequency words About fdist1.hapaxes () $  - #Fine-grained word selection -V =Set (Text1) -Long_words = [w forWinchVifLen (W) >15] A Sorted (long_words) +  the #Word frequency plus the length of words is decided at the same time -FDIST5 =freqdist (TEXT5) $Sorted ([w forWinchSet (TEXT5)ifLen (W) > 7 andFDIST5[W] > 7]) the  the #common words collocation, double-element word collocation the  fromNltk.utilImportBigrams theList (Bigrams ([' More',' is','said','than',' Done'])) -  in  the #the common collocation of double-element words the text4.collocations () About  the #the length of each word in the text the[Len (W) forWinchText1] the  + #the distribution of each length word, the output is a dictionary -fdist = Freqdist ([Len (W) forWinchText1]) the BayiFdist.keys ()#Index Value the Fdist.items () theFdist.max ()#The index of the word with the most words appearing -  -FDIST[3]#where the index value is 3

Nlp-python Natural Language Processing 01