> For the complete documentation index, see [llms.txt](https://docs.raga.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.raga.ai/ragaai-catalyst/ragaai-metric-library/information-extraction/minea.md).

# MINEA

Apply MINEA (Multiple Infused Needle Extraction Accuracy) to test subjective question correction. Identify biases and improve fairness in LLM responses.

**Objective:** 

MINEA assesses information extraction models by measuring the accuracy of identifying rare but critical information (the "needles") from large datasets (the "haystack"). It uses a combination of rarity detection and contextual extraction to evaluate how well a model pulls specific, infrequent data points. A high MINEA score implies the model successfully extracts crucial, rare information without being overwhelmed by more common or irrelevant data.

**Required Columns in Dataset:**

`Labeled Text`, `Source Document`

**Interpretation:**

* **High MINEA**: Indicates that the model excels at identifying rare and crucial pieces of information, even when they are hidden within a large volume of irrelevant or common data.
* **Low MINEA**: Suggests the model struggles to detect infrequent yet important details, leading to potential information gaps in extraction.

**Execution via UI:**

<figure><img src="https://1811327582-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FYbIiNdp1QbG4avl7VShw%2Fuploads%2FLboxuR4Tul53OJPgsCDR%2Fimage.png?alt=media&amp;token=ced2a4a4-c7d3-4446-b18e-bf09420008b5" alt=""><figcaption></figcaption></figure>

**Execution via SDK:**

```python
metrics=[
    {"name": "MINEA", "config": {"model": "gpt-4o-mini", "provider": "openai"}, "column_name": "your-text", "schema_mapping": schema_mapping}
]
```