Automatic reply method and device for user inquiry and electronic equipment

Abstract

The invention provides an automatic reply method and a device for a user inquiry and electronic equipment. The automatic reply method comprises the following steps: determining a question statement text of the inquiry input by a user; inputting the question statement text into a structured query language (SQL) conversion model, and outputting a corresponding question SQL statement; wherein the SQL conversion model is obtained by training based on a sample question statement text and a corresponding question SQL statement label, and the sample question statement text and the corresponding question SQL statement label are automatically constructed according to a preset rule; querying in a preset database by using the questioning SQL statement to output a query result; and splicing the query result into a natural reply statement and outputting the natural reply statement. According to the method, the device and the electronic equipment provided by the invention, automatic reply aiming at the inquiry of the user is realized, and the convenience and the high efficiency of using the intelligent database in reply are improved.

Description

technical field [0001] The invention relates to the technical field of intelligent databases, in particular to an automatic reply method, device and electronic equipment for user inquiries. Background technique [0002] Most of the existing natural language sentence text to Structured Query Language (SQL) Text-to-SQL tasks are based on the Seq2Seq model framework, which regards this task as similar to the machine translation task in neural networks. The Seq2Seq model adds a self-attention mechanism and a copy mechanism to make the Text-to-SQL task similar to the sequence-to-sequence generation task in machine translation. Although the model framework based on Seq2Seq can solve most problems in the case of single-domain datasets, it has not achieved good results in the case of multi-domain datasets. The main reason is that the Seq2Seq architecture cannot satisfy the programming language rules defined in the form of SQL. The SQL language has strict requirements on the restri...

AI Technical Summary

Problems solved by technology

[0006] The present invention provides an automatic reply method, device and electronic equipment for user inquiries, which are used to solve the problems caused by the manual labeling of training data sets that require a large amount of manpower during the training of the Text-to-SQL model in the existing intelligent database question answering system. The use of the intelligent database is not convenient and efficient. The SQL conversion model is trained through the sample query statement and the corresponding query SQL statement label, and the trained SQL conversion model is used in the automatic answer method for user inquiries provided by the present invention. , the key user-input natural language text is automatically converted to SQL statements through the above-mentioned trained SQL conversion model, and the query SQL statement output by the SQL model is used as a query statement to query in the preset database, and the obtained query results are spliced ​​into natural language output after the reply statement; wherein, the sample question statement and the corresponding query SQL statement label are automatically constructed according to preset rules, avoiding the problem of high labor costs caused by manual participation in the standard construction of the training data set in the traditional way, Improve the convenience and efficiency of the application of intelligent database

Images