Keras下的文本情感分析简介。与MLP,RNN,LSTM模型下的文本情感测试

 # coding: utf-8
 
 # In[1]:
 
 import urllib.request
 import os
 import tarfile
 
 # In[2]:
 
 url="http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz"
 filepath="example/data/aclImdb_v1.tar.gz"
 if not os.path.isfile(filepath):
     result=urllib.request.urlretrieve(url,filepath)
     print('downloaded:',result)
 if not os.path.exists("example/data/aclImdb_v1/aclImdb"):
     tfile = tarfile.open("data/aclImdb_v1.tar.gz", 'r:gz')
     result=tfile.extractall('data/')
 
 # In[3]:
 
 from keras.datasets import imdb
 from keras.preprocessing import sequence
 from keras.preprocessing.text import Tokenizer
 
 # In[4]:
 
 import re
 def rm_tags(text):
     re_tag = re.compile(r'<[^>]+>')
     return re_tag.sub('', text)
 
 # In[5]:
 
 import os
 def read_files(filetype):
     path = "example/data/aclImdb_v1/aclImdb/"
     file_list=[]
 
     positive_path=path + filetype+"/pos/"
     for f in os.listdir(positive_path):
         file_list+=[positive_path+f]
 
     negative_path=path + filetype+"/neg/"
     for f in os.listdir(negative_path):
         file_list+=[negative_path+f]
 
     print('read',filetype, 'files:',len(file_list))
     all_labels = ([1] * 12500 + [0] * 12500) 
 
     all_texts  = []
     for fi in file_list:
         with open(fi,encoding='utf8') as file_input:
             all_texts += [rm_tags(" ".join(file_input.readlines()))]
 
     return all_labels,all_texts
 
 # In[6]:
 
 y_train,train_text=read_files("train")
 
 # In[7]:
 
 y_test,test_text=read_files("test")
 
 # In[8]:
 
 train_text[0]
 
 # In[9]:
 
 y_train[0]
 
 # In[10]:
 
 train_text[12500]
 
 # In[11]:
 
 y_train[12500]
 
 # In[12]:
 
 token = Tokenizer(num_words=2000)
 token.fit_on_texts(train_text)
 
 # In[13]:
 
 print(token.document_count)
 print(token.word_index)
 
 # In[14]:
 
 x_train_seq = token.texts_to_sequences(train_text)
 x_test_seq  = token.texts_to_sequences(test_text)
 
 # In[15]:
 
 print(x_train_seq[0])
 
 # In[16]:
 
 x_train = sequence.pad_sequences(x_train_seq, maxlen=100)
 x_test  = sequence.pad_sequences(x_test_seq,  maxlen=100)
 
 # In[17]:
 
 x_train[0]
 
 # In[18]:
 
 from keras.models import Sequential
 from keras.layers.core import Dense, Dropout, Activation,Flatten
 from keras.layers.embeddings import Embedding
 model = Sequential()
 model.add(Embedding(output_dim=32,
                     input_dim=2000,
                     input_length=100))
 model.add(Dropout(0.2))
 model.add(Flatten())
 model.add(Dense(units=256,
                 activation='relu' ))
 model.add(Dropout(0.2))
 model.add(Dense(units=1,
                 activation='sigmoid' ))
 model.summary()
 
 # In[19]:
 
 model.compile(loss='binary_crossentropy',
               optimizer='adam',
               metrics=['accuracy'])
 train_history =model.fit(x_train, y_train,batch_size=100,
                          epochs=10,verbose=2,
                          validation_split=0.2)
 
 # In[20]:
 
 get_ipython().magic('pylab inline')
 import matplotlib.pyplot as plt
 def show_train_history(train_history,train,validation):
     plt.plot(train_history.history[train])
     plt.plot(train_history.history[validation])
     plt.title('Train History')
     plt.ylabel(train)
     plt.xlabel('Epoch')
     plt.legend(['train', 'validation'], loc='upper left')
     plt.show()
 
 # In[21]:
 
 show_train_history(train_history,'acc','val_acc')
 show_train_history(train_history,'loss','val_loss')
 
 # In[22]:
 
 scores = model.evaluate(x_test, y_test, verbose=1)
 scores[1]
 
 # In[23]:
 
 probility=model.predict(x_test)
 
 # In[24]:
 
 probility[:10]
 
 # In[25]:
 
 probility[12500:12510]
 
 # In[26]:
 
 predict=model.predict_classes(x_test)
 
 # In[27]:
 
 predict_classes=predict.reshape(-1)
 
 # In[28]:
 
 SentimentDict={1:'正面的',0:'负面的'}
 def display_test_Sentiment(i):
     print(test_text[i])
     print('标签label:',SentimentDict[y_test[i]],
           '预测结果:',SentimentDict[predict_classes[i]])
 
 # In[29]:
 
 display_test_Sentiment(2)
 
 # In[30]:
 
 display_test_Sentiment(12505)
 
 # In[31]:
 
 from keras.models import Sequential
 from keras.layers.core import Dense, Dropout, Activation
 from keras.layers.embeddings import Embedding
 from keras.layers.recurrent import SimpleRNN
 model = Sequential()
 model.add(Embedding(output_dim=32,
                     input_dim=2000,
                     input_length=100))
 model.add(Dropout(0.35))
 model.add(SimpleRNN(units=16))
 model.add(Dense(units=256,activation='relu' ))
 model.add(Dropout(0.35))
 model.add(Dense(units=1,activation='sigmoid' ))
 model.summary()
 
 # In[32]:
 
 model.compile(loss='binary_crossentropy',
               optimizer='adam',
               metrics=['accuracy'])
 train_history =model.fit(x_train, y_train,batch_size=100,
                          epochs=10,verbose=2,
                          validation_split=0.2)
 
 # In[33]:
 
 scores = model.evaluate(x_test, y_test, verbose=1)
 scores[1]
 
 # In[34]:
 
 from keras.models import Sequential
 from keras.layers.core import Dense, Dropout, Activation,Flatten
 from keras.layers.embeddings import Embedding
 from keras.layers.recurrent import LSTM
 model = Sequential()
 model.add(Embedding(output_dim=32,
                     input_dim=2000,
                     input_length=100))
 model.add(Dropout(0.2))
 model.add(LSTM(32))
 model.add(Dense(units=256,
                 activation='relu' ))
 model.add(Dropout(0.2))
 model.add(Dense(units=1,
                 activation='sigmoid' ))
 model.summary()
 
 # In[35]:
 
 model.compile(loss='binary_crossentropy',
               #optimizer='rmsprop',
               optimizer='adam',
               metrics=['accuracy'])
 train_history =model.fit(x_train, y_train,batch_size=100,
                          epochs=10,verbose=2,
                          validation_split=0.2)
 
 # In[36]:
 
 show_train_history(train_history,'acc','val_acc')
 show_train_history(train_history,'loss','val_loss')
 scores = model.evaluate(x_test, y_test, verbose=1)
 scores[1]
 
 # In[ ]:

文本来源于IMDb网络电影数据集。下载，放到合适的路径下，然后，开始。

过滤掉HTML标签。因为数据集中有相关标签。：

之后读取所有数据和目标标签，然后建立字典：

将文本转化为数字串：

格式化数字串长度为100

建立MLP模型，其中嵌入层将每个长度为100的数字串转为100个32维的向量，将文字映射成多维的几何空间向量，让每一个文字有上下的关联性。

编译，训练，绘图，评估后的准确率：

建立RNN模型，有关RNN模型的介绍：https://www.cnblogs.com/bai2018/p/10466418.html

测试评估：

建立LSTM模型，相关介绍：https://www.cnblogs.com/bai2018/p/10466497.html

准确率：