樸素貝葉斯演算法——實現新聞分類(Sklearn實現)
阿新 • • 發佈:2018-12-30
1、樸素貝葉斯實現新聞分類的步驟
(1)提供文字檔案,即資料集下載
(2)準備資料
將資料集劃分為訓練集和測試集;使用jieba模組進行分詞,詞頻統計,停用詞過濾,文字特徵提取,將文字資料向量化
(3)分析資料:使用matplotlib模組分析
(4)訓練演算法:使用sklearn.naive_bayes 的MultinomialNB進行訓練
在scikit-learn中,一共有3個樸素貝葉斯的分類演算法類。分別是GaussianNB,MultinomialNB和BernoulliNB。
其中GaussianNB就是先驗為高斯分佈的樸素貝葉斯
(5)測試演算法:使用測試集對貝葉斯分類器進行測試
2、程式碼實現
# -*- coding: UTF-8 -*- import os import random import jieba from sklearn.naive_bayes import MultinomialNB import matplotlib.pyplot as plt """ 函式說明:中文文字處理 Parameters: folder_path - 文字存放的路徑 test_size - 測試集佔比,預設佔所有資料集的百分之20 Returns: all_words_list - 按詞頻降序排序的訓練集列表 train_data_list - 訓練集列表 test_data_list - 測試集列表 train_class_list - 訓練集標籤列表 test_class_list - 測試集標籤列表 """ def TextProcessing(folder_path, test_size=0.2): folder_list = os.listdir(folder_path) # 檢視folder_path下的檔案 data_list = [] # 資料集資料 class_list = [] # 資料集類別 # 遍歷每個子資料夾 for folder in folder_list: new_folder_path = os.path.join(folder_path, folder) # 根據子資料夾,生成新的路徑 files = os.listdir(new_folder_path) # 存放子資料夾下的txt檔案的列表 j = 1 # 遍歷每個txt檔案 for file in files: if j > 100: # 每類txt樣本數最多100個 break with open(os.path.join(new_folder_path, file), 'r', encoding='utf-8') as f: # 開啟txt檔案 raw = f.read() word_cut = jieba.cut(raw, cut_all=False) # 精簡模式,返回一個可迭代的generator word_list = list(word_cut) # generator轉換為list data_list.append(word_list) # 新增資料集資料 class_list.append(folder) # 新增資料集類別 j += 1 data_class_list = list(zip(data_list, class_list)) # zip壓縮合並,將資料與標籤對應壓縮 random.shuffle(data_class_list) # 將data_class_list亂序 index = int(len(data_class_list) * test_size) + 1 # 訓練集和測試集切分的索引值 train_list = data_class_list[index:] # 訓練集 test_list = data_class_list[:index] # 測試集 train_data_list, train_class_list = zip(*train_list) # 訓練集解壓縮 test_data_list, test_class_list = zip(*test_list) # 測試集解壓縮 all_words_dict = {} # 統計訓練集詞頻 for word_list in train_data_list: for word in word_list: if word in all_words_dict.keys(): all_words_dict[word] += 1 else: all_words_dict[word] = 1 # 根據鍵的值倒序排序 all_words_tuple_list = sorted(all_words_dict.items(), key=lambda f: f[1], reverse=True) all_words_list, all_words_nums = zip(*all_words_tuple_list) # 解壓縮 all_words_list = list(all_words_list) # 轉換成列表 return all_words_list, train_data_list, test_data_list, train_class_list, test_class_list """ 函式說明:讀取檔案裡的內容,並去重 Parameters: words_file - 檔案路徑 Returns: words_set - 讀取的內容的set集合 """ def MakeWordsSet(words_file): words_set = set() # 建立set集合 with open(words_file, 'r', encoding='utf-8') as f: # 開啟檔案 for line in f.readlines(): # 一行一行讀取 word = line.strip() # 去回車 if len(word) > 0: # 有文字,則新增到words_set中 words_set.add(word) return words_set # 返回處理結果 """ 函式說明:文字特徵選取 Parameters: all_words_list - 訓練集所有文字列表 deleteN - 刪除詞頻最高的deleteN個詞 stopwords_set - 指定的結束語 Returns: feature_words - 特徵集 """ def words_dict(all_words_list, deleteN, stopwords_set=set()): feature_words = [] # 特徵列表 n = 1 for t in range(deleteN, len(all_words_list), 1): if n > 1000: # feature_words的維度為1000 break # 如果這個詞不是數字,並且不是指定的結束語,並且單詞長度大於1小於5,那麼這個詞就可以作為特徵詞 if not all_words_list[t].isdigit() and all_words_list[t] not in stopwords_set and 1 < len(all_words_list[t]) < 5: feature_words.append(all_words_list[t]) n += 1 return feature_words """ 函式說明:根據feature_words將文字向量化 Parameters: train_data_list - 訓練集 test_data_list - 測試集 feature_words - 特徵集 Returns: train_feature_list - 訓練集向量化列表 test_feature_list - 測試集向量化列表 """ def TextFeatures(train_data_list, test_data_list, feature_words): def text_features(text, feature_words): # 出現在特徵集中,則置1 text_words = set(text) features = [1 if word in text_words else 0 for word in feature_words] return features train_feature_list = [text_features(text, feature_words) for text in train_data_list] test_feature_list = [text_features(text, feature_words) for text in test_data_list] return train_feature_list, test_feature_list # 返回結果 """ 函式說明:新聞分類器 Parameters: train_feature_list - 訓練集向量化的特徵文字 test_feature_list - 測試集向量化的特徵文字 train_class_list - 訓練集分類標籤 test_class_list - 測試集分類標籤 Returns: test_accuracy - 分類器精度 """ def TextClassifier(train_feature_list, test_feature_list, train_class_list, test_class_list): classifier = MultinomialNB().fit(train_feature_list, train_class_list) test_accuracy = classifier.score(test_feature_list, test_class_list) return test_accuracy if __name__ == '__main__': # 文字預處理 folder_path = './SogouC/Sample' # 訓練集存放地址 all_words_list, train_data_list, test_data_list, train_class_list, test_class_list = TextProcessing(folder_path,test_size=0.2) # 生成stopwords_set stopwords_file = './stopwords_cn.txt' stopwords_set = MakeWordsSet(stopwords_file) test_accuracy_list = [] """ deleteNs = range(0, 1000, 20) # 0 20 40 60 ... 980 for deleteN in deleteNs: feature_words = words_dict(all_words_list, deleteN, stopwords_set) train_feature_list, test_feature_list = TextFeatures(train_data_list, test_data_list, feature_words) test_accuracy = TextClassifier(train_feature_list, test_feature_list, train_class_list, test_class_list) test_accuracy_list.append(test_accuracy) plt.figure() plt.plot(deleteNs, test_accuracy_list) plt.title('Relationship of deleteNs and test_accuracy') plt.xlabel('deleteNs') plt.ylabel('test_accuracy') plt.show() """ feature_words = words_dict(all_words_list, 450, stopwords_set) train_feature_list, test_feature_list = TextFeatures(train_data_list, test_data_list, feature_words) test_accuracy = TextClassifier(train_feature_list, test_feature_list, train_class_list, test_class_list) test_accuracy_list.append(test_accuracy) ave = lambda c: sum(c) / len(c) print(ave(test_accuracy_list))
結果為:
注:
結巴分詞詞性標註常用詞性表示