Abstract Meaning Representation


Abstract Meaning Representation

38/300
loading

Introduction

Abstract Meaning Representation captures “who is doing what to whom” in a sentence. Each sentence is represented as a rooted, directed, acyclic graph with labels on edges (relations) and leaves (concepts).

How to Use

Apply for Auth

We are hosting a non-commercial API service and you are welcome to apply for an auth keyopen in new window. An auth key is a password which gives you access to our API and protects our server from being abused.

Create RESTful Client

Create a HanLPClientopen in new window:

      from hanlp_restful import HanLPClient
# Support en: English, zh: Chinese, ja: Japanese, mul: Multilingual
HanLP = HanLPClient('https://www.hanlp.com/api', auth=None, language='mul')

    

Parse

      
HanLP.abstract_meaning_representation('男孩希望女孩相信他。')

    

The return value is the Meaning Representation (MR) format of the corresponding AMR graph for each sentence. Note that the term "boy" has two anchors, corresponding to "boy" and "he," respectively. In other words, the MR format actually includes the result of coreference resolution.

Visualization

Set visualization='svg'for visualization。

      
from IPython.display import SVG, display
def show_svg(g):
    display(SVG(data=g['svg']))
graph = HanLP.abstract_meaning_representation('The boy wants the girl to believe him.', 
                                               language='en', visualization='svg')[0]
show_svg(graph)

    

Native APIs

Please refer to docsopen in new window.

Last update: 12/21/2024, 7:01:30 PM
Contributors: hankcs