Priori knowledge-based financial field entity and intention recognition method

Abstract

The invention relates to the technical field of man-machine conversation and intelligent semantic search, in particular to a priori knowledge-based financial field entity and intention recognition method, which specifically comprises the following steps of defining a context model, defining related information of a whole relational database table as a context, including names, annotations and types of tables and column names, and defining a data dictionary, and performing context model analysis: performing analysis according to a database by using a data dictionary, attribute values of related fields and data types, and defining the attribute values of the fields as an open type or a limited enumerable type. Adaptive learning can be realized, and attribute values of intention entities, associated entities and most entities can be identified; the identified entities can be limited in a specific table, on one hand, similarity matching in a global space is avoided, and the step of synonym filtering and scoring is omitted; and secondly, recognized words can be limited in the context, and unidentified words can be filtered out as irrelevant vocabularies, so that the recognition accuracy is prevented from being influenced.

Description

technical field [0001] The invention relates to the technical fields of man-machine dialogue and intelligent semantic search, and specifically relates to a method for recognizing entities and intentions in the financial field based on prior knowledge. Background technique [0002] At present, there are mainly two types of entity recognition in this query sentence: one is to solve it in the form of a dictionary and a dictionary-based synonym configuration method, and the other is to identify common entities such as person names, organization names, and time based on deep learning methods. The first method can directly and quickly solve the recognition problem of some entities, but the generalization ability is insufficient. Although it can be alleviated by adding synonyms, the configuration of synonyms is too many, the operation is troublesome, and it cannot support semantic understanding. For example, "Shanghai Pudong Development Bank 2020 Transaction volume in May" and "Sha...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the entity recognition problem of the query statement in the complex context environment of the financial field, and provides a financial field entity and intent recognition method based on prior knowledge, specifically as follows:

Images