Data processing method, electronic device and readable storage medium for form question answering

Abstract

The present application provides a data processing method, electronic equipment, and readable storage medium for form questions and answers. In the method of the present application, the form data contained in the data form is stored in the document data of the ES, and a row of data in the data form corresponds to a document Data, the field name of the document data is determined according to the data type of the corresponding column and the order in which the corresponding column appears in the column of the same data type. The field type is determined according to the field name. It is not necessary to preset all the field names and their types in advance. Realize the storage of dynamic data tables, support the storage of large-volume table data, and support flexible changes in the structure of data tables. In the form of question and answer, the input question information is converted into a structured query statement, and the structured query statement is converted into a DSL statement. By executing the DSL statement in ES, the answer information of the question information can be obtained from the document data, and the realization Real-time retrieval ensures timely response to form questions and answers.

Description

technical field [0001] The present application relates to artificial intelligence technology, and in particular to a data processing method for form question and answer, electronic equipment and a readable storage medium. Background technique [0002] Form Q&A is aimed at natural language (Natural Language, NL) questions, based on natural language technology to give answers based on the content of the data table. In form question-and-answer scenarios such as product information consultation, the form data is usually at the level of hundreds of thousands or even millions, and the amount of form data is very large. [0003] At present, by storing tabular data in a relational database, the tabular data stored in the relational database is used for retrieval when managing tabular data and questions and answers. For complex and large data scenarios, for example, it is necessary to support fuzzy query (LIKE ) scenarios, it is difficult to achieve real-time response. In order to ...

AI Technical Summary

Problems solved by technology

[0004] However, when storing table data in ElasticSearch, it is necessary to pre-set the field name and field type of each column in the document (document) in ElasticSearch. When adding a new column to the data table, the new column must be specified in ES. The field name and field type are not conducive to the dynamic maintenance of the data table; as the number of columns in the data table increases, the number of fields in the ElasticSearch document continues to increase. When storing large amounts of table data, the retrieval performance based on ElasticSearch Decline, resulting in long response time and poor timeliness for form questions and answers

Images