python基于情感词典进行情感分析的代码

以下是一段基于情感词典的Python情感分析代码。使用情感词典可以将文本中的每个单词与情感极性相关联，然后计算每个单词的情感极性得分并对得分进行加权平均，从而得出整个文本的情感极性。

首先，需要下载并导入情感词典。这里使用的是NTUSD情感词典，可以从以下链接下载并解压缩：

http://www.nlg.csie.ntu.edu.tw/nlpresource/NTUSD-Fin/NTUSD-Fin.zip

然后读取情感词典文件并将每个单词与情感极性相关联：

def read_sentiment_dict(filename):
    sentiment_dict = {}
    with open(filename, 'r', encoding='utf-8') as f:
        for line in f:
            tokens = line.strip().split('\t')
            word = tokens[0]
            sentiment = int(tokens[1])
            sentiment_dict[word] = sentiment
    return sentiment_dict

filename = 'NTUSD-Fin/NTUSD_positive_negative_simplified.txt'
sentiment_dict = read_sentiment_dict(filename)

接下来，可以定义一个函数来计算文本的情感极性得分。该函数将文本分成单词，并对每个单词计算情感极性得分，然后将得分进行加权平均：

def calculate_sentiment(text, sentiment_dict):
    words = text.split()
    sentiment_scores = []
    for word in words:
        if word in sentiment_dict:
            sentiment_scores.append(sentiment_dict[word])
    if len(sentiment_scores) > 0:
        sentiment_score = sum(sentiment_scores) / len(sentiment_scores)
    else:
        sentiment_score = 0
    return sentiment_score

最后，可以使用该函数来计算文本的情感极性得分，并根据得分来判断情感：

text = "I love this product, it's amazing!"
sentiment_score = calculate_sentiment(text, sentiment_dict)
if sentiment_score > 0:
    sentiment = 'Positive'
elif sentiment_score == 0:
    sentiment = 'Neutral'
else:
    sentiment = 'Negative'
print(sentiment)

完整代码：

def read_sentiment_dict(filename):
    sentiment_dict = {}
    with open(filename, 'r', encoding='utf-8') as f:
        for line in f:
            tokens = line.strip().split('\t')
            word = tokens[0]
            sentiment = int(tokens[1])
            sentiment_dict[word] = sentiment
    return sentiment_dict

def calculate_sentiment(text, sentiment_dict):
    words = text.split()
    sentiment_scores = []
    for word in words:
        if word in sentiment_dict:
            sentiment_scores.append(sentiment_dict[word])
    if len(sentiment_scores) > 0:
        sentiment_score = sum(sentiment_scores) / len(sentiment_scores)
    else:
        sentiment_score = 0
    return sentiment_score

filename = 'NTUSD-Fin/NTUSD_positive_negative_simplified.txt'
sentiment_dict = read_sentiment_dict(filename)

text = "I love this product, it's amazing!"
sentiment_score = calculate_sentiment(text, sentiment_dict)
if sentiment_score > 0:
    sentiment = 'Positive'
elif sentiment_score == 0:
    sentiment = 'Neutral'
else:
    sentiment = 'Negative'
print(sentiment)