日本語版

Complaint Busters - Idea Generation Tool to Address Complaints from X (Formerly Twitter) Trends

Overview

Complaint Busters is a tool that extracts negative content from X (formerly Twitter) trends and posts, analyzes it with AI, and generates concrete service ideas to address the complaints. The generated ideas are output in the following formats:

• Summary within 140 characters (for X posts)
• Campfire-style project proposal
• Qiita-style technical article

Additionally, the tool can automatically post the generated ideas to X.

Key Features

1. Trend Retrieval

   Fetch trends from X and extract negative trends (mock data can also be used).

2. Sentiment Analysis

   Analyze the content of trends and posts using AI to determine whether they are positive or negative.

3. Idea Generation

   Generate concrete service ideas to address complaints based on negative trends.

4. Format Conversion

   Convert the generated ideas into the following formats:

   • X post within 140 characters
   • Campfire-style project proposal
   • Qiita-style technical article

5. Automatic Posting

   Automatically post the generated content to X.

6. File Output

   Save the generated ideas to a specified directory.

Technologies Used

• Python

  Used for the overall implementation of the program.

• Tweepy

  Utilized for accessing the X API to fetch trends and post updates.

• Google GenAI

  Uses the Gemini model for prompt-based generation.

• dotenv

  Manages environment variables.

Requirements

• Python 3
• Python libraries:
  • tweepy
  • google-genai
  • python-dotenv

Installation

1. Install the required libraries:

   pip install -r requirements.txt

2. Set up environment variables. Create a .env file with the following content:

   X_API_KEY="YOUR_X_API_KEY"
   X_API_KEY_SECRET="YOUR_X_API_KEY_SECRET"
   X_ACCESS_TOKEN="YOUR_X_ACCESS_TOKEN"
   X_ACCESS_TOKEN_SECRET="YOUR_X_ACCESS_TOKEN_SECRET"
   X_BEARER_TOKEN="YOUR_X_BEARER_TOKEN"
   GEMINI_KEY="YOUR_GEMINI_API_KEY"

How to Run

1. To run with mock data:

   python src/main.py --mock

2. To run with the actual X API:

   python src/main.py

File Structure

• src/main.py

  Main script that performs trend retrieval, sentiment analysis, idea generation, and posting.

• src/lib/

  Modules containing APIs and utility functions.

• mock/

  Stores mock data (trends and post content).

• prompt/

  Stores AI prompt templates.

• output/

  Directory for saving generated ideas.

Notes

• Please comply with the X API terms of service.
• The generated ideas are not guaranteed to be realistic or appropriate. Review carefully before implementation.

Challenges & Future Plans

• X API Costs

  Currently, the cost of using the X API is high, making sufficient data collection difficult. This issue will be addressed once funding is secured through crowdfunding.

• Trend Categorization

  Currently, each tweet is analyzed individually, but aggregating similar tweets could improve trend analysis accuracy. In the future, prompts will be refined for proper categorization to enhance classification accuracy.