# Evaluating Entities

## Case 1

* Dealing with entities of the same type

### Case 1a: # of predicted entities > # of actual entities

**Predicted**: `{name: [Laura, Will]}` **Actual**: `{name: [Laura]}`

**Solution:**

```python
# This will be the score for that individual entity key (name)
intersection(predicted, actual) / set(predicted, actual)
```

### Case 1b: # of predictd entities < # of actual entities

**Predicted**: `{name: [Laura]}` **Actual**: `{name: [Laura, Will]}`

**Solution**: *Same as Case 1a*

## Case 2

* Dealing with missing entity types

### Case 2a: # of predicted entity types > # of actual entity types

**Predicted**:

```
{
  name: [Laura, Will]
  date: [today]
}
```

**Actual**:

```
{
  date: [today]
}
```

**Solution**:

* Give 0 to the extra entity (name)

### Case 2b: # of predicted entity types < # of actual entity types

**Predicted**:

```
{
  date: [today]
}
```

**Actual**:

```
{
  name: [Laura, Will]
  date: [today]
}
```

**Solution**: *Same solution as Case 2b*

## Overall Entity Score

In the end, we will have an average score for each type of entity (eg. name will have score 90%, date will have score 80%, etc.)

Take the average of all these scores and that will be the overall entity score.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://lauradang.gitbook.io/notes/machine-learning/rasa/evaluating-entities.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.