Data-mining in business performance database using explanation-based genetic algorithms

In recent environment of dynamic management, there is growing recognition that information and knowledge management systems are essential for efficient/effective decision making by CEO. To cope with this situation, we suggest the Data-Mining scheme as a key component of integrated information and knowledge management system. The proposed system measures business performance by considering both VA (Value-Added), which represents stakeholder's point of view, and EVA (Economic Value-Added), which represents shareholder's point of view. To mine the new information and knowledge discovery, we applied the improved genetic algorithms that consider predictability, understandability (lucidity) and reasonability factors simultaneously. We use a linear combination model for GAs learning structure. Although this model's predictability will be more decreased than non-linear model, this model can increase the knowledge's understandability, that is, meaning of induced values. Moreover, we introduce a random variable scheme based on normal distribution for initial chromosomes in GAs, so we can expect to increase the knowledge's reasonability, that is, degree of expert's acceptability. The random variable scheme based on normal distribution uses statistical correlation/determination coefficient that is calculated with training data. To demonstrate the performance of the system, we conducted a case study using financial data of Korean automobile industry over 16 years from 1981 to 1996, which are taken from a database of KISFAS (Korea Investors Services Financial Analysis System).