Standard Analyzer

Public Preview

The standard analyzer is the default analyzer in Zilliz Cloud, which is automatically applied to text fields if no analyzer is specified. It uses grammar-based tokenization, making it effective for most languages.

Definition

The standard analyzer consists of:

• Tokenizer: Uses the standard tokenizer to split text into discrete word units based on grammar rules. For more information, refer to Standard.

• Filter: Uses the lowercase filter to convert all tokens to lowercase, enabling case-insensitive searches. For more information, refer to

The functionality of the standard analyzer is equivalent to the following custom analyzer configuration:

Python

analyzer_params = {
    "tokenizer": "standard",
    "filter": ["lowercase"]
}

Java

Map<String, Object> analyzerParams = new HashMap<>();
analyzerParams.put("type", "english");
analyzerParams.put("filter", Collections.singletonList("lowercase"));

Configuration

To apply the standard analyzer to a field, simply set type to standard in analyzer_params, and include optional parameters as needed.

Python

analyzer_params = {
    "type": "standard", # Specifies the standard analyzer type
}

Java

Map<String, Object> analyzerParams = new HashMap<>();
analyzerParams.put("type", "standard");

The standard analyzer accepts the following optional parameters:

Parameter	Description
stop_words	An array containing a list of stop words, which will be removed from tokenization. Defaults to english, a built-in set of common English stop words.

Example configuration of custom stop words:

Python

analyzer_params = {
    "type": "standard", # Specifies the standard analyzer type
    "stop_words", ["of"] # Optional: List of words to exclude from tokenization
}

Java

Map<String, Object> analyzerParams = new HashMap<>();
analyzerParams.put("type", "standard");
analyzerParams.put("stop_words", Collections.singletonList("of"));

After defining analyzer_params, you can apply them to a VARCHAR field when defining a collection schema. This allows Zilliz Cloud to process the text in that field using the specified analyzer for efficient tokenization and filtering. For more information, refer to Example use.

Example output

Here’s how the standard analyzer processes text.

Original text:

"The Milvus vector database is built for scale!"

Expected output:

["the", "milvus", "vector", "database", "is", "built", "for", "scale"]