NLTP app-Analysis Buyer Reviews ratings-high frequency words: two-dimensional relationship

W

#-*-coding:utf-8-*- fromNltkImport*#To fix:no such file or directoryOs.chdir (R'E:\zpy') F= Open ('Reviews_text_lt_3.txt','R') F_r=f.read () strlist= F_r.split (' ') Fdist1=freqdist (strlist)#total number of wordsPrintFdist1#The expression keys () gives us all the different types of linked lists in the textVocabulary1 =Fdist1.keys ()#look at the top 50 items in this list by slicingRes0_50 =vocabulary1[:50]PrintRes0_50

C:\>python E:\zpy\wltp.py<freqdist with 16789 samples and180043 outcomes>["','Raining','disappointing. It','uncomfortable ...',"Lot ' s",'Uv.\nso,','Yellow','Seller',' Four','vaporizers. I','Does','completely!!','Hanging','Monday,','asap!! this','Until','instead. the','malfunctioned.','Lately','looking',' Last','Eligible','Electricity','disappointed','Oneworks','Powdery','unanswered','also.','Refun'sooooo','Foul','On\nafter','fingers.','advice:','Fingers,','advice?','Each ),','month. I']c:\>

SELECT     amz_review_textfrom    amz_reviews_grab_us WHERE      <3;

For buyers through Amazon US USA station, in the time period y-m-d of the first 3,000 records of the database, regardless of the category, price, the relative value of the score and other factors,

The following assumptions are drawn:
0-Sell property is yellow, other conditions in the same circumstances, may not be popular, the score is relatively low;
1-Monday may give buyers a bad purchase experience, Monday promotional activities must be combined with other factors, such as cultural customs, news events cautious;
Note: Dev's current perspective

NLTP app-Analysis Buyer Reviews ratings-high frequency words: two-dimensional relationship