NEC's text analysis technology for analyzing vast amounts of text data based on its "meaning" leads the world in accuracy!
NEC's Recognizing Textual Entailment technology is used in a range of scenarios
- MitaAre there any case studies that you could tell us about?
- MizuguchiLet me talk a bit about case studies, then.
- MitaPlease do.
- MizuguchiEnterprises that conduct business targeted at companies, such as banks, brokerage firms, and manufacturers, use Recognizing Textual Entailment technology in the form of solutions for bolstering information governance. At a certain company, it is used as a system for determining right away whether or not key information that should be managed strictly is included in documents created by sales representatives, due to the importance of governance to properly manage information handled as a prerequisite for utilizing information effectively.
- YamamotoOne of the virtues extolled by customers is the fact this technology facilitates a more accurate understanding of information that should be managed strictly, making it possible to recognize risks associated with information leaks, and leading to improved information governance.
- MitaThat's great!
- MizuguchiIn a similar monitoring example, email has been monitored to detect communications that could indicate collusion or cartel dealings. The concept involves searching email text for content that verges upon cartel dealings or collusion, and issuing an alert to the person responsible when something is found to prevent escalation.
- MitaDo you have any examples of applying this technology to consolidate customer feedback?
- MizuguchiAt many companies that conduct business targeted at consumers, such as those in the manufacturing or service industries, requests from customers are stored as feedback data. A certain company uses a system that can automatically analyze this customer feedback. This makes it possible to automatically group customer feedback collected in text form that was previously categorized and organized manually, and add titles regarding the product or service. As a result, more effort can be focused on improving the quality of customer-oriented services, raising customer satisfaction.
- MitaThat's wonderful!
- MizuguchiThere are also many cases of this technology being used at call centers. There are a large number of companies aiming to place more emphasis on improving their services and products by reducing the time it takes to consolidate complaints and requests from customers using Recognizing Textual Entailment technology. Incidentally, some companies looking at using this technology at call centers have indicated a desire to combine it with voice recognition, and NEC is current working on achieving this.
- MitaCan't they just analyze notes recorded by the operators?
- MizuguchiOf course, analyzing notes made by operators is an option, but this only captures the operator's perspective on things. Call center managers have expressed that they want to analyze a wider range of data, and they would like to be able to analyze requests and complaints from the conversations between customers and operators.
- MitaIt sounds like that would enable this technology to be used for even more applications. I hope to see it being utilized in a whole range of situations. Thank you for taking the time to speak with me today, Mr. Yamamoto and Mr. Mizuguchi!
- Yamamoto, MizuguchiNo, thank you!
In this installment, we heard about NEC's Recognizing Textual Entailment technology. I thought from its complex-sounding name that it wouldn't be very approachable, but after hearing more, the basic concept is very straightforward. It's exciting to think of all the different ways it could be put to use. It looks like it could be applied to consolidating all the requests and feedback our program is sent, so I might have to ask my boss about it next chance I get. See you in the next installment of "MiTA TV"!
(Published July 29, 2015)