Iterative construction method and device for military scenario text event extraction corpus

Abstract

The invention discloses an iterative construction method and device for a military scenario text event extraction corpus. The method comprises the following steps of 1, preprocessing, and obtaining anoriginal data set represented by a word sequence; 2, constructing a seed data set, defining an event template, constructing an event trigger word dictionary, forming the seed data set through manualannotation, and dividing the seed data set into a seed training set and a test set; 3, training a model, training a machine learning model by using the seed training set, testing the model by using the test set, and optimizing the model parameters according to a test result to obtain a first learning model; 4, selecting an unlabeled training corpus, and inputting the unlabeled training corpus intothe first learning model to obtain a prediction result set; 5, correcting the prediction result set to form a new annotation corpus; and 6, through the continuous iteration, generating the training sets in sequence to form the event extraction corpus. According to the iterative construction method for the military scenario text event extraction corpus, the corpus construction efficiency is improved, the manual annotation cost is reduced, and the relatively higher corpus annotation accuracy is obtained.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, in particular to a method and device for iteratively constructing a military scenario text event extraction corpus. Background technique [0002] With the rapid development of information technology, information is also growing explosively. How to extract and organize a large amount of unordered information in a timely manner, quickly and accurately obtain useful information needed by users, and transform it into a structured form that both humans and machines can understand and use has become the focus of research and development. Information extraction is generated and developed under this background. Event extraction is an advanced stage of information extraction, and it is the most challenging task. It mainly studies the extraction of event information of interest to users from various texts, presents it in a structured form, and provides it for other information extracti...

AI Technical Summary

Problems solved by technology

However, the construction of the corpus is mostly carried out by purely manual annotation, that is, the annotator directly performs the annotation operation on the original corpus, which has a low degree of automation, consumes a lot of time and manpower, is inefficient, and has poor security and confidentiality, which is not conducive to corpus management and protection of

Images