Analyzes huge volumes of text data on issues under debate, and presents reasons and grounds for stances
Hitachi has developed a technology that analyzes huge volumes of text data on issues that are subject to debate, and presents reasons and grounds for either affirmative or negative opinions on those issues in English. This technology focuses on values such as health, economics and public safety, which are considered important to people and communities when expressing opinions, and uses correlations between those various values and relevant issues in the society to identify reasons and grounds with a high degree of reliability from among large volumes of news articles.
This will contribute to artificial intelligence enabling logical dialogue between humans and computers. The technology could be applied to future systems to analyze contents of company documents, published reports or electronic medical records, in order to form opinions and generate data to support decision making
In recent years, with the evolution of analysis technologies and information & telecommunication technologies such as the Internet, attention has been attracted to technologies that analyze “Big Data”, which is generated every day by various sensors and POS systems – and identify valuable information. At the same time, there has been an increasing demand for effective use of data such as company documents, published reports and electronic medical records to help give additional value and make management decisions.
Details of the technology developed are:
Hitachi focused on values such as health, economics and public safety, which are considered important to people and communities, and created a “Value Dictionary” that systematically organizes those values based on a database – a database that records affirmative and negative opinions regarding a large number of discussion topics.
Metadata is created by identifying correlations between issues and their values from huge volumes of text data. Using this method, the system created approximately 250 million metadata from around 9.7 million news articles.
Calculated reliability of the extracted sentences. By processing all of the sentences that could potentially serve as reasons or grounds for opinions, and evaluating scores, it is possible to select and present reliable grounds.
Constructed architecture to realize asynchronous distributed processing of multiple algorithms.
