Professor Yada started to specialize in data mining* when he was studying business science at an evening course for graduate studies in Kobe while working in retail during the day. This was in the wake of the Great Hanshin Awaji Earthquake. In his studies, he noticed that a huge amount of customer data was virtually untapped. This situation inspired Professor Yada to study computer literacy and programming independently to examine business applications for data mining while working in retail. Consequently, he began consulting with sales managers using his research data, and became involved in various company sales projects.
*Data Mining: Studying processes for analyzing knowledge consisting of rules and patterns found in large-scale databases.
When a customer shops at a store, shopping path refers to visualizing all in-store movements of a consumer such as which shelf he or she stops, what the customer is thinking, how he or she makes purchase decisions, and so on. Professor Yada has taken a leading role in this field of study and launched the Data Science Laboratory (formerly the Data Mining Applied Research Center) at Kansai University, in association with Columbia University, USA.
The most challenging problem in this research was how the burdens on retailers could be lessened. Shopping path data through the use of video monitoring systems requires installation work in the shop, and the expense becomes a burden to the retailer. Professor Yada examined various methods to acquire data, and found a solution in gathering location information via sensor network. By affixing an RFID tag to a shopping cart and tracking the positional information of carts at every access point (wireless LAN base station), the customer shopping path can be monitored with minimum wiring installation.
Shopping path shows data for which shelf a consumer stops, and how long that consumer takes in considering whether or not to purchase. It is also possible to incorporate with POS (point-of-sale) data, barcode information gathered at a checkout counter, to develop business insights and opportunities for retailers and manufactures. Information about which section in a store a customer spends time contributes toward developing effective promotion schemes. Knowing the length of time spent for considering purchase of every item enables to the retailer to make logical selling strategies for navigating customers toward purchasing. For example, customers tend to spend more time making purchase decisions on desserts than on brand-name products such as beer.
Moreover, obtaining information for how long a customer walks around in a store, and how he or she spends time has merits for designing each selling section corresponding to target age group. On the other hand, selling corners laid out based on customer shopping path are appealing to consumers, providing an easier environment to select items.
Professor Yada provided a theoretic model acquired from his analysis through open source on Internet, and is now teamed up with various product manufactures for items including amenities, food and beverages. Application of customer shopping path data is currently being expanded to various other business fields.
Professor Yada believes that on-site knowledge is most important for analyzing data, and sends his students to actual shops such as supermarkets when they start their research. They learn to analyze data in various industrial fields from Professor Yada, and many start working actively in areas of planning development to design data-based theoretical business plans after graduation. His belief is that research should be beneficial to people at workplaces, and this corresponds to Kansai University’s slogan “Knowledge for the Real World.” Professor Yada’s challenge for developing new business frontiers continues as he creates new business chances with the application of data mining.