Zwounds/boolean-search-model

Boolean Search Query Model

Convert natural language queries into proper boolean search expressions for academic databases. This model helps researchers and librarians create properly formatted boolean search queries from natural language descriptions.

Features

• Converts natural language to boolean search expressions
• (MOSTLY!) Handles multi-word terms correctly with quotes
• Removes meta-terms (articles, papers, research, etc.)
• Groups OR clauses appropriately
• Minimal, clean formatting

Installation

pip install transformers torch unsloth

from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    "Zwounds/boolean-search-model",
    max_seq_length=2048,
    dtype=None,  # Auto-detect
    load_in_4bit=True
)
FastLanguageModel.for_inference(model)

Quick Start

# Format your query
query = "Find papers about climate change and renewable energy"
prompt = f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
Convert this natural language query into a boolean search query by following these rules:

1. FIRST: Remove all meta-terms from this list (they should NEVER appear in output):
   - articles, papers, research, studies
   - examining, investigating, analyzing
   - findings, documents, literature
   - publications, journals, reviews
   Example: "Research examining X" → just "X"

2. SECOND: Remove generic implied terms that don't add search value:
   - Remove words like "practices," "techniques," "methods," "approaches," "strategies"
   - Remove words like "impacts," "effects," "influences," "role," "applications"
   - For example: "sustainable agriculture practices" → "sustainable agriculture"
   - For example: "teaching methodologies" → "teaching"
   - For example: "leadership styles" → "leadership"

3. THEN: Format the remaining terms:
   CRITICAL QUOTING RULES:
   - Multi-word phrases MUST ALWAYS be in quotes - NO EXCEPTIONS
   - Examples of correct quoting:
     - Wrong: machine learning AND deep learning
     - Right: "machine learning" AND "deep learning"
     - Wrong: natural language processing
     - Right: "natural language processing"
   - Single words must NEVER have quotes (e.g., science, research, learning)
   - Use AND to connect required concepts
   - Use OR with parentheses for alternatives (e.g., ("soil health" OR biodiversity))

Example conversions showing proper quoting:
"Research on machine learning for natural language processing"
→ "machine learning" AND "natural language processing"

"Studies examining anxiety depression stress in workplace"
→ (anxiety OR depression OR stress) AND workplace

"Articles about deep learning impact on computer vision"
→ "deep learning" AND "computer vision"

"Research on sustainable agriculture practices and their impact on soil health or biodiversity"
→ "sustainable agriculture" AND ("soil health" OR biodiversity)

"Articles about effective teaching methods for second language acquisition"
→ teaching AND "second language acquisition"

### Input:
{query}

### Response:
"""

# Generate boolean query
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(result)  # "climate change" AND "renewable energy"

Examples

Input queries and their boolean translations:

1. Natural: "Studies about anxiety depression stress in workplace"

   • Boolean: (anxiety OR depression OR stress) AND workplace

2. Natural: "Articles about artificial intelligence ethics and regulation or policy"

   • Boolean: "artificial intelligence" AND (ethics OR regulation OR policy)

3. Natural: "Research on quantum computing applications in cryptography or optimization"

   • Boolean: "quantum computing" AND (cryptography OR optimization)

Rules

The model follows these formatting rules:

1. Meta-terms are removed:

   • "articles", "papers", "research", "studies"
   • Focus on actual search concepts

2. Quotes only for multi-word terms:

   • "artificial intelligence" AND ethics ✓
   • NOT: "ethics" AND "ai" ✗

3. Logical grouping:

   • Use parentheses for OR groups
   • (x OR y) AND z

4. Minimal formatting:

   • No unnecessary parentheses
   • No repeated terms

Local Development

# Clone repo
git clone https://github.com/your-username/boolean-search-model.git
cd boolean-search-model

# Install dependencies
pip install -r requirements.txt

# Run tests
python test_boolean_model.py

Contributing

1. Fork the repository
2. Create your feature branch
3. Add tests for any new functionality
4. Submit a pull request

Model Card

See MODEL_CARD.md for detailed model information including:

• Training data details
• Performance metrics
• Limitations
• Intended use cases

License

This model is subject to the Llama 2 license. See the LICENSE file for details.

Citation

If you use this model in your research, please cite:

@misc{boolean-search-llm,
  title={Boolean Search Query LLM},
  author={Stephen Zweibel},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/Zwounds/boolean-search-model}
}

Contact

Stephen Zweibel - @szweibel