Design of fusion technique-based mining engine for smart business

Keys to successful implementation of smart business require a wide spectrum of domain knowledge, experts, and their correlated experiences. Excluding those external factors—which can be collected by well-deployed sensors—being aware of user (or consumer) has the highest priority on the to-do-list. The more user is understood, the more user can be satisfied from an intuitive point of view, and thus, data plays a rather essential role in the scenario. However, it is never easy to achieve comprehensive understanding as the data requires further processing before its values can be extracted and used. So how the data can be properly transformed into something useful for smart business development is exactly what we pursue in this study. As a pioneer, three major tasks are focused. First, a data mining engine based on the concept of the KID model is designed and developed to be responsible for the universal collection of data and mining valuable information which is primarily from real world, cyber world, and social world. Second, we go further into the fusion process of the collected data and meaningful information extracted and interpreted by algorithms or fused algorithms in the data mining engine (e.g., the consumer purchase data shared by real-world company) and turn them into valuable knowledge about the situation of customers and business situations based on the concept of knowledge, information, and data. A three-layer analysis and mining procedure is designed to enhance the mining engine through conventional RFM (Recency, Frequency, and Monetary Value) model and a set of fusion techniques. And in the end, we make planning-based predictions for a real-world company for expansion of the business interests.