计算文件夹中文本文件复合情感值平均值时遇技术问题

Hey there! Let's figure out how to calculate the average compound sentiment score for each text file in your folder—building right off the code you already have for sentence-level analysis. Here's a step-by-step solution with fully working code and explanations:

Full Working Code #

import glob
import os
import nltk.data
from nltk.sentiment.vader import SentimentIntensityAnalyzer
from nltk import sentiment
from nltk import word_tokenize

# Initialize VADER and sentence tokenizer (your original setup!)
sid = SentimentIntensityAnalyzer()
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')

# Replace this with your actual folder path containing text files
folder_path = "/path/to/your/text/files"

# Loop through every .txt file in the target folder
for file_path in glob.glob(os.path.join(folder_path, "*.txt")):
    # Grab the filename to keep track of which file we're processing
    filename = os.path.basename(file_path)
    
    # Read the full content of the text file
    with open(file_path, 'r', encoding='utf-8') as file:
        text_content = file.read()
    
    # Split the text into individual sentences (using your tokenizer)
    sentences = tokenizer.tokenize(text_content)
    
    # Collect compound sentiment scores for all sentences in the file
    compound_scores = []
    for sentence in sentences:
        # Get VADER's sentiment scores for the sentence
        sentiment_scores = sid.polarity_scores(sentence)
        # Extract the compound score (the overall sentiment value: -1 to 1)
        compound_scores.append(sentiment_scores['compound'])
    
    # Calculate and print the average compound score for the file
    if compound_scores:
        average_compound = sum(compound_scores) / len(compound_scores)
        print(f"✅ File: {filename} | Average Compound Sentiment: {average_compound:.4f}")
    else:
        print(f"⚠️ File: {filename} | No valid sentences found (empty or unreadable file?)")

Key Details & Troubleshooting Tips #

• Folder Path Setup: Make sure to replace /path/to/your/text/files with the actual path to your folder. Using os.path.join ensures this works across Windows, Mac, and Linux without path separator issues.
• Encoding Fixes: If you get encoding errors when reading files, try swapping encoding='utf-8' for encoding='latin-1' or adding errors='ignore' to the open() function.
• Empty File Handling: The check for compound_scores prevents division-by-zero errors if a file is empty or has no recognizable sentences.
• Optional: Save Results to CSV: If you want to store the results instead of just printing them, add this code (place it right before the file loop):

  import csv
  
  # Create a CSV file to save results
  with open("sentiment_results.csv", 'w', newline='', encoding='utf-8') as csvfile:
      writer = csv.writer(csvfile)
      writer.writerow(["Filename", "Average Compound Sentiment"])
      
      # Then inside your file loop, replace the print statement with:
      if compound_scores:
          writer.writerow([filename, round(average_compound, 4)])
      else:
          writer.writerow([filename, "N/A"])
  

内容的提问来源于stack exchange，提问作者K. SOT