Semantic Parsing with Knowledge-Based Editor for Data Operations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data science projects are time-consuming due to the repetitive and labor-intensive tasks of sourcing, cleaning, and preparing data, with software services being difficult to configure, leading to lost time savings.

Innovation Solution

A data processing apparatus that includes a software library, user interface, parser, knowledge base, software service execution scheduler, and result processor, allowing users to interact with software services via a domain-specific language, automating data processing tasks and recommending next actions based on previous user interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If users directly configure and interact with multiple individual software services, then they can access specific data processing functions, but they need to learn the syntax and semantics of each service, increasing the time and complexity of operation

Engineering Contradiction:
Improveease of accessing software servicesVSAvoidtime to learn and configure services
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The patent introduces a natural language interface as an intermediary between the user and multiple software services. Users interact with a single unified interface that translates natural language commands into appropriate service calls, eliminating the need to learn individual service syntax and semantics while maintaining access to diverse data processing functions

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If users manually configure software services to execute data processing tasks, then they can perform specific data operations, but the configuration process is time-consuming and offsets the time savings from automation

Engineering Contradiction:
Improveautomation of data processingVSAvoidtime to configure services
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system enables self-service by automatically translating user natural language intent into configured service execution. The interface autonomously handles service selection, parameter configuration, and task scheduling without requiring manual user configuration, thus preserving the productivity benefits of automation while eliminating configuration overhead

Inventive Principle:
Principle #25Self-service

3Ease of operation

If a unified user interface is provided to access multiple software services, then ease of operation improves, but the system complexity increases due to the need for parsing, scheduling, and coordination mechanisms

Engineering Contradiction:
Improvesingle interface accessVSAvoidsystem architecture complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent segments the complex system into distinct functional modules: a natural language parser that processes user input, a knowledge base that stores service metadata and relationships, a scheduler that coordinates service execution, and result processors that handle outputs. This modular segmentation manages system complexity by assigning specific responsibilities to each component while maintaining a simple unified user interface

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS10445427B2Semantic parsing with knowledge-based editor for execution of operations
Publication Date: 2019.10.15 FUJITSU LTD
  • US10445427B2 patent drawing
  • US10445427B2 patent drawing
  • US10445427B2 patent drawing

AI summary

A data processing apparatus including a library, storing software services, each executing a data processing function; a user interface receiving commands, expressed in a language and defining a data processing target and a data processing request; and a parser. The parser extracting from each command: the request from the language; and the target. A knowledge base maintaining a record of the request and the target; a scheduler obtaining the request from the parser, and compiling a schedule to fulfill the request; a controller controlling execution of the schedule, the target being the input data to the execution, and to output a result; and a result processor obtaining the result, and, based on records of requests and targets, to identify a request candidate, and to output to the user as a selectable command in the language, via the user interface, the candidate with the result defined as a data processing target.