Natural Language Dialogue Event Extraction for IoT Program Generation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technologies face difficulties in recording programs to obtain user intentions from natural language text data corresponding to dialogue, making it challenging to efficiently control IoT devices with natural spoken dialogue.

Innovation Solution

An information processing device equipped with a text data analysis unit to extract events and processes from dialogue and a record control unit to record these correspondences on a recording medium, allowing for automatic program generation and execution based on detected events.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If a pre-registered program is used to analyze text data and obtain user intentions, then the system can operate automatically, but it becomes difficult to easily record new programs for obtaining user intentions

Engineering Contradiction:
Improveautomatic operationVSAvoidease of recording program
Core Design Contradiction:
Extent of automationVSEase of manufacture

Solution Approach 1:

The system automatically extracts events and processes from text data and generates programs without requiring manual programming. The program generation is performed automatically by the system itself based on the extracted information, eliminating the need for providers to manually create programs.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary extraction of events and processes from text data before program generation. By pre-processing the text data to identify key elements, the system prepares the necessary information in advance for automatic program creation, making the overall process easier and more efficient.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If manual program recording is used to capture user intentions from natural language, then program accuracy can be maintained, but the burden on providers and users increases significantly

Engineering Contradiction:
Improveprogram accuracyVSAvoidoperational burden
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent replaces manual mechanical programming with an automatic information processing system. Instead of manually creating programs, the system automatically extracts events and processes from text data and generates programs, substituting human effort with automated computational processes.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system introduces an intermediary processing layer that automatically extracts events and processes from text data. This intermediary step bridges the gap between natural language input and program generation, maintaining accuracy while reducing the operational burden on users and providers.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If existing technology is used to analyze dialogue text data, then basic understanding can be achieved, but efficient control of IoT devices with natural spoken dialogue cannot be realized

Engineering Contradiction:
Improvebasic understandingVSAvoidefficiency of IoT device control
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system segments text data into distinct events and processes, allowing for more precise analysis and control. By dividing the text information into structured components, the system can efficiently map user intentions to specific IoT device control actions, improving overall productivity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically generates programs based on extracted events and processes rather than relying on static pre-registered programs. This dynamic approach allows the system to adapt to various IoT device control scenarios efficiently, improving productivity while maintaining reliability.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3418881B1Information processing device, information processing method, and program
Publication Date: 2020.04.01 SONY GROUP CORP
  • EP3418881B1 patent drawingFigure 1A
  • EP3418881B1 patent drawingFigure 1B
  • EP3418881B1 patent drawingFigure 2A

AI summary

It is desired to provide a technology capable of easily recording a program for obtaining an intention of a user from text data corresponding to dialogue using a natural language. Provided is an information processing device including: a text data analysis unit configured to analyze text data corresponding to dialogue using a natural language, and extract an event included in the text data and a process to be executed in accordance with the event; and a record control unit configured to record a correspondence relation between the event and the process on a recording medium as a program.