Natural Language Workflow Generation via Sentence Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems that generate software products from natural language specifications struggle to construct and display workflows from sentences that do not follow specific rules, leading to incomplete or inconsistent process flows.

Innovation Solution

A non-transitory computer readable medium with a program that includes a receiving unit for natural sentences, a generator to create process workflows, and a display controller to visualize the workflow content, using a constituent element conversion dictionary and reference information to analyze and correct sentences for compliance with predetermined criteria.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If the system requires sentences to follow specific rules for workflow generation, then the manufacturing precision of the process flow is improved, but the ease of operation deteriorates because users must write in a restricted format

Engineering Contradiction:

Improveprocess flow consistencyVSAvoidinput flexibility

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent introduces a sentence analysis unit and constituent element conversion dictionary as intermediaries between the user's natural language input and the workflow generation system. These intermediaries analyze the input sentence, extract constituent elements, and convert them into structured data that the workflow generator can process, thereby enabling the system to handle flexible natural language input while maintaining consistent process flow output

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical approach of requiring users to follow strict syntax rules with an intelligent language analysis system. Instead of enforcing rigid input formats, the system uses natural language processing to automatically analyze and interpret user input, extracting meaningful elements and converting them into workflow structures without requiring users to adhere to specific grammatical rules

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If the system accepts any natural language input, then the ease of operation is improved, but the manufacturing precision of the generated workflow deteriorates due to incomplete or inconsistent process flows

Engineering Contradiction:

Improveinput flexibilityVSAvoidprocess flow completeness

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The patent implements a feedback mechanism where the sentence analysis unit continuously analyzes the input sentence, identifies constituent elements, and converts them into structured data. The workflow generation unit then generates the process flow based on this structured data, and the system provides feedback by displaying the generated workflow to the user for verification and correction, ensuring completeness and consistency

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary analysis of the input sentence by extracting constituent elements and converting them into structured data before workflow generation. This preliminary processing ensures that all necessary information is captured and organized properly, preventing incomplete or inconsistent process flows while maintaining input flexibility

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If the system uses complex sentence analysis and correction processes, then the manufacturing precision of workflow generation is improved, but the device complexity increases

Engineering Contradiction:

Improveworkflow accuracyVSAvoidsystem structure

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the workflow generation system into distinct functional units: a sentence analysis unit for analyzing input sentences, a constituent element conversion dictionary for converting extracted elements into structured data, and a workflow generation unit for generating the actual workflow. This segmentation allows each unit to specialize in its specific task, improving overall workflow accuracy while making the system structure more manageable and modular

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11461547B2Non-transitory computer readable medium for generating a target program source using software specification written in a natural language

Publication Date: 2022.10.04 FUJIFILM BUSINESS INNOVATION CORP

AI summary

A non-transitory computer readable medium storing a program that causes a processor to function as an information processing apparatus. The information processing apparatus includes a receiving unit configured to receive an input of a natural sentence that describes the content of a process for a process target, a generator configured to generate a series of processes including the process in accordance with the received natural sentence, and a display controller configured to perform control so as to display content information indicating contents of the generated series of processes.