Keyword Grouping — extract topic clusters from article URLs

Analyse Keywords API

Hosted on ApyHub

What it does

Keyword Grouping analyzes one to ten publicly accessible article URLs and returns ranked keyword groups for the content it finds. Send article URLs, optionally include an article_length hint, and set return_results_top if you want to limit how many groups come back.

Use Keyword Grouping when you need to compare articles by topic, cluster related terms across a content library, or pull out the main themes from a batch of pages. The service is built for article analysis, so it works well for editorial workflows, content research, and SEO tooling where you want grouped terms instead of a flat keyword list.

The response contains data.keyword_groups, an array of objects with a numeric score and a keywords array. Each group represents a ranked set of related keywords extracted from the provided articles, which makes it easy to sort by relevance or feed the results into downstream classification, search, or reporting jobs.

Because the input is limited to public article URLs, this is a good fit for pipelines that analyze published pages without uploading files or raw HTML. Send the URLs, get structured keyword groups back, and use them to identify repeated topics, build content clusters, or surface the terms that matter most across a set of articles.

Quickstart

Send a list of article URLs to extract and group their keywords.

curl -X POST "http://localhost:8080/apyhub/extract-and-group-keywords-from-article" \
  -H "apy-token: $APY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "urls": [
      "https://assets.apyhub.com/samples/sample.html"
    ]
  }'

What you'll get back

Returns a JSON object with a data object. Inside data, keyword_groups is an array of ranked keyword groups, and each group contains a numeric score and a keywords array of strings.

{
  "data": {
    "individual_results": [
      {
        "url": "https://assets.apyhub.com/samples/sample.html",
        "latent_semantic_keywords": [
          "eligendi",
          "expedita",
          "eos",
          "alias",
          "aliquid",
          "dignissimos",
          "ea",
          "eaque",
          "aliquam",
          "dicta"
        ],
        "high_frequency_keywords": [
          {
            "keyword": "et",
            "frequency": 56
          },
          {
            "keyword": "ut",
            "frequency": 35
          }
        ],
        "high_frequency_keyphrases": [
          {
            "keyphrase": "consequatur et",
            "frequency": 3
          },
          {
            "keyphrase": "mollitia et",
            "frequency": 3
          }
        ],
        "preview": "Lorem ipsum dolor sit amet...."
      }
    ],
    "aggregated_results": null
  }
}

About this endpoint

What it does

Extracts and groups keywords from one or more publicly accessible article URLs. The request sends a list of URLs and optional tuning fields, and the response returns ranked keyword groups.

Request Body

Parameter	Type	Mandatory	Description
urls	String Array	Yes	Publicly accessible article URLs to analyse. Maximum 10; extras are ignored.
article_length	Integer	No	Expected article character length hint (optional; helps the parser tune extraction).
return_results_top	Integer	No	Maximum number of keyword groups to return (5–50). Omit to return all groups.

Response

Returns a JSON object with a data field containing keyword extraction results. The response includes an individual_results array with keyword analysis for each processed URL. If aggregation is requested, the aggregated_results field contains the combined analysis; otherwise, it is null.

Parameter	Type	Mandatory	Description
data	Object	Yes	Wrapper object containing the response payload.
data.individual_results	Object Array	Yes	Keyword extraction results for each processed URL.
data.individual_results[].url	String	Yes	URL of the processed article or webpage.
data.individual_results[].latent_semantic_keywords	String Array	Yes	List of latent semantic (contextually related) keywords extracted from the content.
data.individual_results[].high_frequency_keywords	Object Array	Yes	Most frequently occurring keywords with their occurrence counts.
data.individual_results[].high_frequency_keywords[].keyword	String	Yes	Extracted keyword.
data.individual_results[].high_frequency_keywords[].frequency	Integer	Yes	Number of times the keyword appears in the content.
data.individual_results[].high_frequency_keyphrases	Object Array	Yes	Most frequently occurring keyphrases with their occurrence counts.
data.individual_results[].high_frequency_keyphrases[].keyphrase	String	Yes	Extracted keyphrase.
data.individual_results[].high_frequency_keyphrases[].frequency	Integer	Yes	Number of times the keyphrase appears in the content.
data.individual_results[].preview	String	Yes	Short preview of the processed content.
data.aggregated_results	Object \| Null	Yes	Aggregated keyword extraction results across all processed URLs, or `null` if aggregation is not available.

Body

Name

Type

Description

bodyREQUIRED

object

—

Quickstart

What you'll get back

About this endpoint

What it does

Request Body

Response

Body

Errors any endpoint can return