Character filters in Elasticsearch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a character filter in Elasticsearch?

A character filter is a preprocessing step that modifies the text before tokenization. It changes or removes characters to prepare the text for analysis.

beginner

Name two built-in character filters in Elasticsearch.

Two built-in character filters are html_strip which removes HTML tags, and mapping which replaces characters or strings based on a mapping.

intermediate

How does the mapping character filter work?

The mapping filter replaces specified characters or strings with other characters or strings before tokenization. For example, it can replace accented letters with plain letters.

beginner

Why use character filters before tokenization?

Character filters clean or normalize text by removing unwanted characters or replacing them. This helps tokenizers produce consistent tokens and improves search accuracy.

intermediate

Show a simple example of a custom character filter using mapping in Elasticsearch.

Example:

{
  "analysis": {
    "char_filter": {
      "my_mapping": {
        "type": "mapping",
        "mappings": ["æ=>ae", "œ=>oe"]
      }
    }
  }
}

This replaces 'æ' with 'ae' and 'œ' with 'oe' before tokenization.

What is the main purpose of a character filter in Elasticsearch?

A Modify text before tokenization

B Create tokens from text

C Store indexed documents

D Run search queries

Character filters change or clean text before it is split into tokens.

Which character filter removes HTML tags from text?

A html_strip

B mapping

C lowercase

D stop

The html_strip filter removes HTML tags before tokenization.

What does the mapping character filter do?

A Removes stop words

B Converts text to lowercase

C Replaces characters or strings based on rules

D Splits text into tokens

Mapping replaces specified characters or strings with others before tokenization.

When are character filters applied in the analysis process?

A After tokenization

B Only during search queries

C During indexing only

D Before tokenization

Character filters run before tokenization to prepare the text.

Which of these is NOT a function of character filters?

A Replacing characters

B Splitting text into tokens

C Removing unwanted characters

D Normalizing text

Splitting text into tokens is done by tokenizers, not character filters.

Explain what character filters do in Elasticsearch and why they are important.

Think about how text is prepared before breaking it into words.

Describe how you would use a mapping character filter to replace special characters in text.

Consider replacing accented letters with plain letters.