TOKYO – Hakuhodo DY Holdings Inc. (Minato-ku, Tokyo; Hirokazu Toda, President & CEO) is pleased to announce that it has established a Data Exchange Platform Development Division, aiming for secure, broader utilization of the sei-katsu-sha and social data owned by companies and organizations.
Better integration of the wide range of data held by companies and organizations has been endorsed by Society 5.0, a basic plan on technology drawn up by the Japanese government, and Connected Industries, a concept advocated by the Ministry of Economy, Trade and Industry. Attention has focused on improving the convenience of services and infrastructure, and developing innovative services, by integrating and utilizing a wide variety of data from multiple data-holders. However, coordination of data among companies has made little headway as the actual use of data for marketing requires careful consideration and full compliance, including the handling of personal information and the protection of privacy.
Hakuhodo DY Group has promoted the use of various data, enhancing our capacity for sei-katsu-sha data-driven marketing*1 in line with the Group policy of using sei-katsu-sha insight. In so doing, we developed and obtained a patent on K-Statistics & Data Fusion*2, a technique for anonymizing and integrating personal information. This patented technique helps resolve the issue of how to handle personal information when using data, and facilitates the broader use of data securely. Our decision to establish the new division was motivated by our desire to build a data exchange platform that enables us to integrate and use data in response to the diverse needs of companies and organizations. By participating in the platform, data-holding companies and organizations can expect to capture new business opportunities by leveraging data. Meanwhile, data-using companies and organizations will gain increased possibilities to develop marketing and services that leverage hitherto-inaccessible external data.
In addition, Hakuhodo DY Holdings has been engaging in collaboration and cooperation with the National Institute of Advanced Industrial Science and Technology (AIST)*3, with the aim of becoming a society where we can distribute and utilize data securely. We are also a member of AIST’s Artificial Intelligence Technology Consortium (Chairman: Yoichi Motomura), a collection of companies that are highly interested in using data by leveraging AI technology and machine learning in the era of big data. To help the Consortium achieve its aim of generating co-created values, we are leading its Data Platform Working Group. The Working Group engages with AIST on technological research and undertakes such activities toward societal implementation as carrying out demonstration experiments on using data in business and solving social issues, working with a wide variety of companies and organizations across industries through the Consortium.*4
Going forward, the Data Exchange Platform Development Division will explore possibilities for data integration with various data-holding companies and organizations, and continue to build the foundation for secure data processing, as part of our Group-wide effort for driving data utilization.
*1 Sei-katsu-sha data-driven marketing
Sei-katsu-sha data-driven marketing is a marketing concept focusing on the collection, processing, integration and analysis of various sei-katsu-shadata to inform planning and action for resolving marketing issues.
Building on our proprietary Sei-katsu-sha Data Management Platform (DMP), developed by combining data on the information habits, purchasing behavior and attitudes of sei-katsu-sha with cutting-edge technology, Hakuhodo DY Group offers quality marketing solutions leveraging our strengths in planning, creative and executional capabilities.
*2 The K-Statistics & Data Fusion technique
K-Statistics & Data Fusion is a (patented) technique to integrate and utilize multiple datasets while protecting personal information. First, we process personal data from each data source and generate micro-clusters composed of k or more persons*5 to anonymize information while minimizing the loss of information from the raw data (k-statistics processing). Then, we integrate the data by statistically combining those micro-clusters from different data sources with similar characteristics (fusion processing). This sequenced processing generates rich data from multiple datasets that are not as detailed as personal data but nonetheless maintain the statistical granularity of the original data.
*3 National Institute of Advanced Industrial Science and Technology (AIST)
AIST is a public research institute under the Ministry of Economy, Trade and Industry, established to promote economic development by disseminating the outcomes of technological research and development to the industrial community.
*4 Working with AIST’s Artificial Intelligence Technology Consortium, we will leverage this innovative data exchange platform for data integration across companies and industries in various organs of the Consortium, including the Working Group on Data & Intelligence, Working Group on Manufacturing and Kyushu Branch.
*5 The number k will be selected to suit the type of data or the purpose of use, e.g., k=10, k=100, etc.
