Skip to content

Relation extraction

Relation extraction is a type of document analysis that labels concepts expressed in the text with their semantic role.

Relation extraction also performs knowledge linking: Knowledge Graph information and open data—Wikidata, DBpedia and GeoNames references—are returned for relation items corresponding to syncons of the expert.ai Knowledge Graph. In the case of actual places, geographic coordinates are also provided.

Full analysis includes relation extraction, but if you are not interested in the other analyses, you can use specific resources having paths like this:

analyze/context name/language code/relations

Boxed parts are placeholders, so for example:

https://nlapi.expert.ai/v2/analyze/standard/en/relations

is the URL of the standard context resource performing relation extraction on an English text.
These resources must be requested with the POST method.

In the reference section of this manual you will find all the information you need to perform relation extraction using the API's RESTful interface, specifically:

Note

Even if you consume the API through a ready-to-use client that hides low-level requests and responses, knowing the output format helps you understand and navigate the results.

Here is an example of performing relation extraction on a short English text:

This example code uses expertai-nlapi, the open-source Python client corresponding to the nlapi-python GitHub project.

The client gets user credentials from two environment variables:

EAI_USERNAME
EAI_PASSWORD

Set those variables with your account credentials before running the sample program below.

The program prints a shallow (no nesting) representation of relations.

from expertai.nlapi.cloud.client import ExpertAiClient
client = ExpertAiClient()

text = "Michael Jordan was one of the best basketball players of all time. Scoring was Jordan's stand-out skill, but he still holds a defensive NBA record, with eight steals in a half." 
language= 'en'

output = client.specific_resource_analysis(
    body={"document": {"text": text}}, 
    params={'language': language, 'resource': 'relations'
})

print("Relations:");

for relation in output.relations:
    print(relation.verb.lemma, ":");
    for related in relation.related:
        print("\t", "(", related.relation, ")", related.lemma);

This example code uses @expertai/nlapi, the open-source NodeJS client corresponding to the nlapi-nodejs GitHub project.

The client gets user credentials from two environment variables:

EAI_USERNAME
EAI_PASSWORD

Set those variables with your account credentials before running the sample program below.

The program prints a shallow (no nesting) representation of relations.

import {NLClient} from "@expertai/nlapi";
import {Language} from "@expertai/nlapi";
import {Analysis} from "@expertai/nlapi";

var nlClient = new NLClient();

var text = "Michael Jordan was one of the best basketball players of all time. Scoring was Jordan's stand-out skill, but he still holds a defensive NBA record, with eight steals in a half.";

nlClient.analyze(text, {
  language: Language.EN,
  context: "standard",
  analysis: Analysis.Relations
}).then((result) => {
    console.log("Relations:");

    for (const relation of result.data.relations) {
        console.log(relation.verb.lemma + ":")
        for (const related of relation.related) {
            console.log("\t(" + related.relation + ") " + related.lemma);
        }
    }
})

This example code uses nlapi-java-sdk, the open-source Java client corresponding to the nlapi-java GitHub project.

The client gets user credentials from two environment variables:

EAI_USERNAME
EAI_PASSWORD

Set those variables with your account credentials before running the sample program below.

The program prints the JSON response.

import ai.expert.nlapi.security.Authentication;
import ai.expert.nlapi.security.Authenticator;
import ai.expert.nlapi.security.BasicAuthenticator;
import ai.expert.nlapi.security.DefaultCredentialsProvider;
import ai.expert.nlapi.v2.API;
import ai.expert.nlapi.v2.cloud.Analyzer;
import ai.expert.nlapi.v2.cloud.AnalyzerConfig;
import ai.expert.nlapi.v2.message.AnalyzeResponse;

public class Main {

    public static Authentication createAuthentication() throws Exception {
        DefaultCredentialsProvider credentialsProvider = new DefaultCredentialsProvider();
        Authenticator authenticator = new BasicAuthenticator(credentialsProvider);
        return new Authentication(authenticator);
    }

    public static Analyzer createAnalyzer() throws Exception {
        return new Analyzer(AnalyzerConfig.builder()
                .withVersion(API.Versions.V2)
                .withContext("standard")
                .withLanguage(API.Languages.en)
                .withAuthentication(createAuthentication())
                .build());
    }

    public static void main(String[] args) {
        try {
            String text = "Michael Jordan was one of the best basketball players of all time. Scoring was Jordan's stand-out skill, but he still holds a defensive NBA record, with eight steals in a half.";

            Analyzer analyzer = createAnalyzer();

            AnalyzeResponse relations = analyzer.relations(text);


            // Output JSON representation

            System.out.println("JSON representation:");
            relations.prettyPrint();
        }
        catch(Exception ex) {
            ex.printStackTrace();
        }
    }
}

The following curl command posts a document to the named entity recognition resource of the API's REST interface.
Run the command from a shell after replacing token with the actual authorization token.

curl -X POST https://nlapi.expert.ai/v2/analyze/standard/en/relations \
    -H 'Authorization: Bearer token' \
    -H 'Content-Type: application/json; charset=utf-8' \
    -d '{
  "document": {
    "text": "Michael Jordan was one of the best basketball players of all time. Scoring was Jordan'\''s stand-out skill, but he still holds a defensive NBA record, with eight steals in a half."
  }
}'

The server returns a JSON object.

The following curl command posts a document to the named entity recognition resource of the API's REST interface.
Open a command prompt in the folder where you installed curl and run the command after replacing token with the actual authorization token.

curl -X POST https://nlapi.expert.ai/v2/analyze/standard/en/relations -H "Authorization: Bearer token" -H "Content-Type: application/json; charset=utf-8" -d "{\"document\": {\"text\": \"Michael Jordan was one of the best basketball players of all time. Scoring was Jordan's stand-out skill, but he still holds a defensive NBA record, with eight steals in a half.\"}}"

The server returns a JSON object.