Method and system for identifying Chinese event sequential relationship

Abstract

The invention provides a method and system for identifying a Chinese event sequential relationship. Each document of an original text, the event sequential relationship of which needs to identify, is subjected to word segmentation, entity identification, syntactic analysis, dependence relationship analysis and event extraction action, such that a test corpus event set is obtained; all event pairs and feature information thereof are respectively extracted from the test corpus event set and a labelled corpus set; a maximum entropy event sequential relationship identifying model is trained according to the features of various event pairs in a labelled corpus feature set; the sequential relationship of each event pair in the test corpus feature set is identified by utilizing the maximum entropy event sequential relationship identifying model, such that a first event sequential relationship set is obtained; and, by taking the document as the unit, event sequential relationship reasoning of all the event pairs in the first event sequential relationship set is carried out by utilizing a word denoting time reasoning method, an event relationship reasoning method, a reflexive reasoning method and a transitive reasoning method, such that an event sequential relationship set is obtained.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to a method and a system for identifying temporal relationship of Chinese events. Background technique [0002] Event (Event) is a main form of information representation. It is an objective fact (also called "natural event") of specific people, things, and things interacting at a specific time and a specific place, such as human injury and death events. and food additive incidents, etc. Events are a unique pragmatic form in which objective facts are expressed in words, and an article generally consists of various events surrounding a certain topic or related topics, and these events express the core content of the article. Therefore, the events in the text are often not isolated individuals, and their occurrence and development often have a certain relationship with other external events surrounding the same topic, such as the "chronological" relati...

AI Technical Summary

Problems solved by technology

[0007] At present, there are two problems in the vast majority of event timing relationship identification methods: 1) Whether it is a rule method or a machine learning method, most of them isolate a pair of events and then identify their timing relationship

These methods generally do not consider the relationship between multiple events in a text and their mutual influence; 2) Existing methods focus on identifying the temporal relationship of events within a sentence or in adjacent sentences, and do not consider the timing of events in non-adjacent sentences relation

Images