-

Mitsubishi Electric’s AI Creates Knowledge Graphs from Text and Graphics to Visualize Information Relevance

TOKYO--(BUSINESS WIRE)--Mitsubishi Electric Corporation (TOKYO: 6503) announced today that it has developed a technology based on its Maisart®1 AI technology that automatically constructs knowledge graphs by acquiring key phrases, authors, citation relationships and whole-part relationships of elements in various materials, including figures and tables, and then visualizes the relevance of the information so that users can identify and understand the most necessary information quickly and intuitively. The new technology is expected to greatly reduce the amount of time users spend gathering information.

Conventionally, it can take much time to collect necessary information from the overload of information people are exposed to these days. Moreover, to quickly find necessary or interesting information, the information must not only be digitalized but also managed based on information relationships within or among the materials. Mitsubishi Electric’s new AI technology structurally digitalizes materials and data by extracting important information and estimating the interrelationships in advance. Thereafter, when users explore the information, they can grasp the strengths of relationships based on the varying widths of bands in diagrams (see figure above) to quickly and intuitively find necessary, interesting and even previously unnoticed information.

For the full text, please visit: www.MitsubishiElectric.com/news/

Contacts

FOR IMMEDIATE RELEASE
Customer Inquiries
Information Technology R&D Center
Mitsubishi Electric Corporation
www.MitsubishiElectric.com/ssl/contact/company/rd/form.html
www.MitsubishiElectric.com/

Media Inquiries
Takeyoshi Komatsu
Public Relations Division
Mitsubishi Electric Corporation
Tel: +81-3-3218-2346
prd.gnews@nk.MitsubishiElectric.co.jp
www.MitsubishiElectric.com/news/

Mitsubishi Electric Corporation

TOKYO:6503

Release Versions

Contacts

FOR IMMEDIATE RELEASE
Customer Inquiries
Information Technology R&D Center
Mitsubishi Electric Corporation
www.MitsubishiElectric.com/ssl/contact/company/rd/form.html
www.MitsubishiElectric.com/

Media Inquiries
Takeyoshi Komatsu
Public Relations Division
Mitsubishi Electric Corporation
Tel: +81-3-3218-2346
prd.gnews@nk.MitsubishiElectric.co.jp
www.MitsubishiElectric.com/news/

More News From Mitsubishi Electric Corporation

Mitsubishi Electric Invests in Tulip Interfaces And Signs Strategic Alliance Agreement

TOKYO--(BUSINESS WIRE)--Mitsubishi Electric Corporation (TOKYO: 6503) announced today that it has invested in and signed a strategic alliance agreement with Tulip Interfaces, Inc., a Massachusetts, USA-based leader no-code platforms for system operations without programming to support manufacturing digitalization. Tulip Interfaces is also an expert in introducing manufacturing-targeted microservices, which divide large-scale systems into small, independent services to enable flexible developmen...

Mitsubishi Electric Publishes First TNFD Report

TOKYO--(BUSINESS WIRE)--Mitsubishi Electric Corporation (TOKYO: 6503) announced today that it has published its first Taskforce on Nature-related Financial Disclosures (TNFD) report, which summarizes the extent of Mitsubishi Electric group companies’ dependence on natural capital, the impact of their operations on the natural environment, the associated business risks and opportunities, and initiatives they are taking to reduce their environmental impact. The contents of the report are in accor...

Mitsubishi Electric Technology Detects Intoxication During Driving to Provide Driver Alerts and Even Vehicle-control Interventions

TOKYO--(BUSINESS WIRE)--Mitsubishi Electric Corporation (TOKYO: 6503) announced today that it has developed a technology that accurately detects intoxication levels indicated by driver distraction and drowsiness, and provides driver alerts and vehicle-control interventions as needed to help prevent alcohol-related accidents. The system detects intoxication using a combination of non-contact pulse-rate measurements, which are based on images captured with a driver monitoring system (DMS), and ve...
Back to Newsroom