BibAI Filter - AI-Powered Academic Publication Analyzer

Advanced AI analysis for academic publications

🆕 What's New (April 2026 Release)

This release is a major refresh that upgrades every supported provider to its April 2026 flagship family and streamlines the provider roster:

• OpenAI upgraded to the GPT-5.4 family (gpt-5.4-2026-03-05, -mini, -nano). max_tokens has been migrated to max_completion_tokens everywhere, as required by the new reasoning-capable models.
• Anthropic upgraded to Claude 4 family (claude-opus-4-7, claude-sonnet-4-6, claude-haiku-4-5-20251001).
• Google upgraded to Gemini 3.x family (gemini-3.1-pro-preview, gemini-3-flash-preview, gemini-3.1-flash-lite-preview) via the new google-genai SDK.
• DeepSeek added as a fourth first-class provider (deepseek-reasoner and deepseek-chat on V3.2). DeepSeek is accessed through the OpenAI-compatible endpoint, so no extra SDK is required.
• Removed providers: Mistral AI, Cohere, and Azure OpenAI were retired from the UI and ai_processor to reduce maintenance surface. They can be reintroduced via a PR if requested.
• Hygiene: added .gitignore, untracked compiled __pycache__/ artifacts that were accidentally committed earlier, and sanitized config/API_Settings.json so it only contains placeholder keys.

📖 Overview

BibAI Filter is a sophisticated desktop application designed for researchers and academics who need to efficiently filter large volumes of scholarly publications. Using state-of-the-art AI models, this tool analyzes titles, abstracts, and keywords from your Excel-based publication lists to identify the most relevant papers for your research topics.

✨ Key Features

• Seamless Data Import: Easily load Excel files (.xlsx or .xls) containing your publication databases

• Flexible Column Selection: Define which columns contain titles, abstracts, and keywords

• AI-Powered Analysis: Score publications based on relevance to your specified research topic using advanced AI models

  • Supported AI Providers (April 2026):

    Provider	Premium	Mid	Fast
    OpenAI	gpt-5.4-2026-03-05	gpt-5.4-mini-2026-03-17	gpt-5.4-nano-2026-03-17
    Anthropic	claude-opus-4-7	claude-sonnet-4-6	claude-haiku-4-5-20251001
    Google	gemini-3.1-pro-preview	gemini-3-flash-preview	gemini-3.1-flash-lite-preview
    DeepSeek	deepseek-reasoner (V3.2 Thinking)	deepseek-chat (V3.2)	—

  Note: DeepSeek currently ships only two official API models; no dedicated "fast" tier is offered.

• Smart Filtering: Filter publications based on a customizable relevance threshold

• Comprehensive Results: Export filtered publications to a new Excel file with original data and AI relevance scores

• Real-Time Progress Tracking: Monitor the filtering process with an intuitive progress indicator

• User-Friendly Interface: Clean and intuitive PyQt5-based interface for a smooth user experience

🚀 Installation

1. Clone the Repository

   git clone https://github.com/bcankara/BibAIFilter.git

   cd BibAIFilter

2. Create a Virtual Environment (Recommended)

   python -m venv .venv
   
   # On Linux/macOS
   source .venv/bin/activate
   
   # On Windows
   .venv\Scripts\activate

3. Install Dependencies

   pip install -r requirements.txt

4. Launch the Application

   python main.py

🔍 Usage Guide

1. Start the application

2. Configure AI Settings

   • Navigate to the "Settings" tab
   • Select your preferred AI Provider (OpenAI, Anthropic, Google, or DeepSeek)
   • Enter your API Key for the selected provider
   • Choose an appropriate AI Model (premium / mid / fast tier depending on provider)
   • Click "Test Connection" to verify the key before running large batches

3. Load and Filter Publications

   • Switch to the "Input & Filtering" tab
   • Click "Select Excel File" to load your publication database
   • Specify which columns contain Titles, Abstracts, and Keywords
   • Enter your Research Topic in the text field (e.g., "Quantum Computing in Cryptography")
   • Adjust the "Relevance Threshold" slider to set filtering sensitivity (value between 0 and 1)
   • Select an output location using "Choose Output File"
   • Start the process by clicking "Begin Filtering"

4. Review Results

   • When processing completes, the filtered results will be saved to your specified output file
   • The log area will show a summary of the operation

📋 Requirements

All dependencies are listed in the requirements.txt file. Key requirements include:

• Python 3.8+
• PyQt5
• pandas
• openpyxl
• xlrd
• openai (also used for DeepSeek via OpenAI-compatible API)
• anthropic
• google-genai
• requests

🤝 Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Please adhere to coding standards and clearly describe your changes.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔒 Security Note

API keys are sensitive information and should be handled securely. The application stores keys locally in the config/API_Settings.json file. The version tracked in this repository contains only placeholder values (sk-YOUR_..._API_KEY_HERE); your real keys stay on your machine.

Before committing any changes, always verify that config/API_Settings.json does not contain your real API keys. A standard .gitignore is provided for __pycache__/, virtual environments, and other local files.