Customer Sentiment Analysis Report Generator

Overview

This Python application automates the process of generating a comprehensive customer sentiment analysis report. It leverages the Webz.io eCommerce API and ChatGPT to create detailed reports by analyzing customer reviews of a specified product.

Features

• Automated Report Generation: Generates reports by summarizing up to 50 reviews into positive and negative categories using ChatGPT.
• Customer Review Analysis: Splits product reviews into positive and negative feedback for a detailed analysis.
• Image Generation: Utilizes DALL-E 3 to create professional cover images for the reports.
• Sentiment-Based Summarization: Summarizes feedback based on sentiment (positive/negative) and includes this in the report.
• Document Formatting: Creates Word documents with formatted text, hyperlinks, and images.
• Report Sections: Includes sections like Analysis of Feedback, Recommendations, and Conclusions, differentiated by sentiment.

Installation

1. Install required Python packages:

   • json
   • glob
   • docx
   • requests
   • openai
   • bs4

2. Set up an environment variable OPENAI_API_KEY with your OpenAI API key.

Usage

1. Place the product reviews in the reviews folder, following the ndjson format.
2. Run the main() function to start the report generation process.
3. The application will process the reviews, generate summaries, and compile them into a Word document titled customer sentiment analysis report.docx.

Functions

• add_image_from_base64(doc, image_url): Adds an image to the Word document from a base64 encoded string.
• html_to_word(doc, html_content): Converts HTML content to formatted text in a Word document.
• add_hyperlink(paragraph, url, text): Adds a hyperlink to a Word paragraph.
• generate_article_image(name): Generates an image for the report using DALL-E 3.
• call_gpt_completion(prompt): Calls GPT-4 to generate text based on a given prompt.
• extract_points(reviews, sentiment): Extracts key points from reviews based on sentiment.
• generate_intro(product_name, product_description): Generates an introduction for the report.
• generate_title(product_name): Generates a title for the report.
• create_negative_report(feedback, product_name): Creates a negative sentiment analysis report.
• create_positive_report(feedback, product_name): Creates a positive sentiment analysis report.
• create_word_doc(...): Compiles the entire report into a Word document.
• read_ndjson_file(file_path): Reads ndjson files and returns their content.

Example

if __name__ == "__main__":
    main()