Decompounder

The decompounder filter splits compound words into individual components based on a specified dictionary, making it easier to search for parts of compound terms. This filter is particularly useful for languages that frequently use compound words, such as German.

Configuration

The decompounder filter accepts its component dictionary either inline via the word_list parameter or from a registered file resource via the word_list_file parameter.

Inline word list

The decompounder filter is a custom filter in Zilliz Cloud. To use it, specify "type": "decompounder" in the filter configuration, along with a word_list parameter that provides the dictionary of word components to recognize.

Python

python

analyzer_params = {
    "tokenizer": "standard",
    "filter":[{
        "type": "decompounder", # Specifies the filter type as decompounder
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"],
    }],
}

Java

java

import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;

Map<String, Object> analyzerParams = new HashMap<>();
analyzerParams.put("tokenizer", "standard");
analyzerParams.put("filter",
        Collections.singletonList(
                new HashMap<String, Object>() {{
                    put("type", "decompounder");
                    put("word_list", Arrays.asList("dampf", "schiff", "fahrt", "brot", "backen", "automat"));
                }}
        )
);

Go

go

analyzerParams = map[string]any{"tokenizer": "standard",
    "filter": []any{map[string]any{
        "type":       "decompounder",
        "word_list": []string{"dampf", "schiff", "fahrt", "brot", "backen", "automat"},
    }}}

Rust

rust

let analyzer_params = serde_json::json!({
    "tokenizer": "standard",
    "filter": [{
        "type": "decompounder",
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"]
    }]
});

C++

c++

nlohmann::json analyzer_params = {
    {"tokenizer", "standard"},
    {"filter", {
        {
            {"type", "decompounder"},
            {"word_list", {"dampf", "schiff", "fahrt", "brot", "backen", "automat"}}
        }
    }}
};

NodeJS

javascript

// javascript
const analyzer_params = {
    "tokenizer": "standard",
    "filter":[{
        "type": "decompounder", // Specifies the filter type as decompounder
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"],
    }],
};

cURL

bash

# restful
analyzerParams='{
  "tokenizer": "standard",
  "filter": [
    {
      "type": "decompounder",
      "word_list": [
        "dampf",
        "schiff",
        "fahrt",
        "brot",
        "backen",
        "automat"
      ]
    }
  ]
}'

The decompounder filter accepts the following configurable parameters.

Parameter	Description
word_list	A list of word components used to split compound terms. This dictionary determines how compound words are decomposed into individual terms.

The decompounder filter operates on the terms generated by the tokenizer, so it must be used in combination with a tokenizer. For a list of tokenizers available in Zilliz Cloud, refer to Standard Tokenizer and its sibling pages.

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 details, refer to Example use.

Examples

Before applying the analyzer configuration to your collection schema, verify its behavior using the run_analyzer method.

Analyzer configuration

Python

python

analyzer_params = {
    "tokenizer": "standard",
    "filter":[{
        "type": "decompounder", # Specifies the filter type as decompounder
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"],
    }],
}

Java

java

import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;

Map<String, Object> analyzerParams = new HashMap<>();
analyzerParams.put("tokenizer", "standard");
analyzerParams.put("filter",
        Collections.singletonList(
                new HashMap<String, Object>() {{
                    put("type", "decompounder");
                    put("word_list", Arrays.asList("dampf", "schiff", "fahrt", "brot", "backen", "automat"));
                }}
        )
);

Go

go

analyzerParams = map[string]any{"tokenizer": "standard",
    "filter": []any{map[string]any{
        "type":       "decompounder",
        "word_list": []string{"dampf", "schiff", "fahrt", "brot", "backen", "automat"},
    }}}

Rust

rust

let analyzer_params = serde_json::json!({
    "tokenizer": "standard",
    "filter": [{
        "type": "decompounder",
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"]
    }]
});

C++

c++

nlohmann::json analyzer_params = {
    {"tokenizer", "standard"},
    {"filter", {
        {
            {"type", "decompounder"},
            {"word_list", {"dampf", "schiff", "fahrt", "brot", "backen", "automat"}}
        }
    }}
};

NodeJS

javascript

// javascript
const analyzer_params = {
    "tokenizer": "standard",
    "filter":[{
        "type": "decompounder", // Specifies the filter type as decompounder
        "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"],
    }],
};

cURL

bash

# restful
analyzerParams='{
  "tokenizer": "standard",
  "filter": [
    {
      "type": "decompounder",
      "word_list": [
        "dampf",
        "schiff",
        "fahrt",
        "brot",
        "backen",
        "automat"
      ]
    }
  ]
}'

Verification using run_analyzer

Python

python

from pymilvus import (
    MilvusClient,
)

client = MilvusClient(uri="YOUR_CLUSTER_ENDPOINT")

# Sample text to analyze
sample_text = "dampfschifffahrt brotbackautomat"

# Run the standard analyzer with the defined configuration
result = client.run_analyzer(sample_text, analyzer_params)
print("Standard analyzer output:", result)

Java

java

import io.milvus.v2.client.ConnectConfig;
import io.milvus.v2.client.MilvusClientV2;
import io.milvus.v2.service.vector.request.RunAnalyzerReq;
import io.milvus.v2.service.vector.response.RunAnalyzerResp;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

ConnectConfig config = ConnectConfig.builder()
        .uri("YOUR_CLUSTER_ENDPOINT")
        .build();
MilvusClientV2 client = new MilvusClientV2(config);

List<String> texts = new ArrayList<>();
texts.add("dampfschifffahrt brotbackautomat");

RunAnalyzerResp resp = client.runAnalyzer(RunAnalyzerReq.builder()
        .texts(texts)
        .analyzerParams(analyzerParams)
        .build());
List<RunAnalyzerResp.AnalyzerResult> results = resp.getResults();

Go

go

import (
    "context"
    "encoding/json"
    "fmt"

    "github.com/milvus-io/milvus/client/v3/milvusclient"
)

client, err := milvusclient.New(ctx, &milvusclient.ClientConfig{
    Address: "YOUR_CLUSTER_ENDPOINT",
    APIKey:  "YOUR_CLUSTER_TOKEN",
})
if err != nil {
    fmt.Println(err.Error())
    // handle error
}

bs, _ := json.Marshal(analyzerParams)
texts := []string{"dampfschifffahrt brotbackautomat"}
option := milvusclient.NewRunAnalyzerOption(texts...).
    WithAnalyzerParamsStr(string(bs))

result, err := client.RunAnalyzer(ctx, option)
if err != nil {
    fmt.Println(err.Error())
    // handle error
}
fmt.Println(result)

Rust

rust

use milvus::v2::error::Result;
use milvus::v2::prelude::*;
use serde_json::json;

#[tokio::main]
async fn main() -> Result<()> {
    let config = ConnectConfig::new().uri("YOUR_CLUSTER_ENDPOINT").token("YOUR_CLUSTER_TOKEN");
    let client = ClientV2::new(&config).await?;

    let analyzer_params = json!({
        "tokenizer": "standard",
        "filter": [{
            "type": "decompounder",
            "word_list": ["dampf", "schiff", "fahrt", "brot", "backen", "automat"]
        }]
    });

    let result = client.run_analyzer(
        RunAnalyzerRequest::builder()
            .texts(["dampfschifffahrt brotbackautomat".to_string()])
            .analyzer_params(analyzer_params)
            .build()?,
    ).await?;
    println!("{:?}", result.results());
    Ok(())
}

C++

c++

#include "milvus/MilvusClientV2.h"
#include <iostream>
#include <string>

auto client = milvus::MilvusClientV2::Create();

milvus::ConnectParam connect_param{"YOUR_CLUSTER_ENDPOINT"};
auto status = client->Connect(connect_param);
if (!status.IsOk()) {
    std::cout << status.Message() << std::endl;
}

std::string text = "dampfschifffahrt brotbackautomat";
auto request = milvus::RunAnalyzerRequest()
                       .AddText(text)
                       .WithAnalyzerParams(analyzer_params);

milvus::RunAnalyzerResponse response;
status = client->RunAnalyzer(request, response);
if (!status.IsOk()) {
    std::cout << status.Message() << std::endl;
}

for (const auto& result : response.Results()) {
    for (const auto& token : result.Tokens()) {
        std::cout << token.token_ << " ";
    }
    std::cout << std::endl;
}

NodeJS

javascript

// javascript
import { MilvusClient } from '@zilliz/milvus2-sdk-node';

const client = new MilvusClient({
  address: 'YOUR_CLUSTER_ENDPOINT',
  token: 'YOUR_CLUSTER_TOKEN'
});

const res = await client.runAnalyzer({
  text: 'dampfschifffahrt brotbackautomat',
  analyzer_params: analyzer_params
});

console.log(res);

cURL

bash

# restful
curl --request POST \
  --url "${CLUSTER_ENDPOINT}/v2/vectordb/common/run_analyzer" \
  --header "Authorization: Bearer ${TOKEN}" \
  --header "Content-Type: application/json" \
  --data '{
    "analyzerParams": "{\"tokenizer\":\"standard\",\"filter\":[{\"type\":\"decompounder\",\"word_list\":[\"dampf\",\"schiff\",\"fahrt\",\"brot\",\"backen\",\"automat\"]}]}",
    "text": ["dampfschifffahrt brotbackautomat"]
  }'

Expected output

python

['dampf', 'schiff', 'fahrt', 'brotbackautomat']