By using Japanese Keyword Extraction API, you can create an application which extracts "keywords" from an input document.
Japanese Keyword Extraction API extracts "keywords", such as person names, location names, organization names, and so on, from an input document consisting of a title and a body. The "keywords" are scored by the position of appearance in the title and the body, the number of occurrences in the document, and the type of named entity.
The "keywords" are useful to tag news articles to make it easier to search.
Acceptable POST requests: application/x-www-form-urlencoded or application/json
※ Please be note that GET requests are not acceptable.
|app_id (Required)||Application ID（Required item）|
If omitted, then request_id is assumed to be “form|pos|read”
|title (Required)||Japanese title string to analyze（Required item）|
|body (Required)||Japanese body string to analyze(Required item)|
|max_num||Maximum number of keywords to output
If omitted, then max_num is 10.
|focus||Type of named entity to focus attention on
Select a type of the named entity to focus attention on from "ORG" (organization name), "PSN" (person name), and "LOC" (location name). The "focused" type gets more score. If omitted or set a type other than the above, then all types are scored equally.
API response body: JSON format
API content-type: application/json
|request_id||Same value as request.|
|focus||Same value as request.|
|keywords||The list of keywords
The array of keywords with a score.