问情感分析python 3亚马逊

EN

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
from bs4 import BeautifulSoup

import nltk
from nltk.corpus import stopwords 
from nltk.stem.porter import PorterStemmer
from nltk.stem import SnowballStemmer

data = pd.read_csv('Reviews.csv') 
print(data.head(4))

data['pos'] = np.where(data['Score'] > 3, 1, 0) 
#suddivion the file that analyzed by the score ( from 1 to 5)
X_train, X_test, y_train, y_test = train_test_split(data['Text'],data['pos'], test_size = 0.2, random_state=0) 


nltk.download()
# now start the problems
def pulitoretesto (prim_testo, rmv_stpwrds=False, stemming=False, split_testo=False):

    testo = BeautifulSoup(prim_testo, 'lxml').get_text() 
    lett = re.sub("[^a-zA-Z]", " ",testo)  
    pr = lett.lower().split() 

    if rmv_stpwrds: 
        stop = set(stopwords.words("english"))
        pr= [a for a in pr if not w in stop]

    if stemming==True: 
        eliminsuf = SnowballStemmer('english') 
        pr = [eliminsuf.stem(a) for a in pr]  

    if split_testo==True:  
        return (pr)

    return( " ".join(pr))   
X_train_nuovo = [] 

for x in X_train:
       X_train_nuovo.append(pulitoretesto(x))   
print(' nuova X_train :\n',  X_train_nuovo)

X_test_nuovo = []

 for x in X_test:
     X_test_nuovo.append(pulitoretesto(x)) 

  tokenizer=nltk.data.load('tokenizers/punkt/english.pickle')

def parsfrasi(revi,tokenizer, rmv_stpwrds=False):
    prmtv_frs=tokenizer.tokenize(revi.strip()) 
    frasi = []
    for prmtv_frs1 in prmtv_frs : 
        if len(prmtv_frs1) > 0 :
            frasi.pulitoretesto(prmtv_frs1, rmv_stpwrds, split_testo=True )
    return frasi 

frasi = []
for revi in X_train_nuovo:
    frasi += parsfrasi(revi, tokenizer)