A method to automatically generate natural language text from structured process model

Abstract

The invention discloses a method for automatically generating a natural language text from a structured flow model, belonging to the field of flow mining. The present invention first utilizes the label text analysis technology to obtain and analyze the label text information in the BPMN process model; secondly, uses the process model structure conversion technology to convert the structure of the BPMN process model into a process structure tree; finally, uses natural language to generate technology to generate natural language text. The invention can generate natural language texts with correct syntax and complete semantics, so that non-professionals can understand the BPMN process model by reading the natural language texts.

Description

technical field [0001] The invention belongs to the field of process mining, and in particular relates to a method for automatically generating natural language text from a structured process model. Background technique [0002] Currently, the generation of natural language text from the BPMN (Business Process Model and Notation, business process model and notation) process model includes two types of solutions: one is to generate corresponding natural language text through the reading and understanding of the model by BPMN process model experts; the other One is the natural language text automatically generated by the method of natural language information analysis. [0003] The first type of method is to rely on the help of process model experts. The experts master the knowledge related to the BPMN process model. The experts read and understand the BPMN process model and express the meaning of the model in natural language text. [0004] The second type of method is to us...

AI Technical Summary

Problems solved by technology

[0007] Artificially generated solutions cannot solve the problem of long time period

As the scale of the process model increases, for process model experts, the correctness of the generated text cannot be guaranteed, and the cost of manual generation is high and the efficiency is low

[0008] The natural language analysis and generation scheme is to use templates and label text information to generate natural language text. Since the grammar of the generated text is not considered, the grammar of the generated text cannot be guaranteed to be correct, so the readability of the generated text is poor

For the branches, selections, loops, concurrency and other structures existing in the process model, the natural language analysis method cannot handle them correctly, so the process structure information will be lost, which is fatal for understanding the process model

Images