Conjoint Data Mining of Structured and Semi-structured Data

With the knowledge management requirement growing, enterprises are becoming increasingly aware of the significance of interlinking business information across structured and semi-structured data sources. This problem has become more important with the growing amount of semi-structured data often found in XML repositories, web logs, biological databases, etc. Effectively creating links between semi-structured and structured data is a challenging and unresolved problem. Once an optimized method has been formulated, the process of data mining can be implemented in a conjoint manner. This paper investigates a way in which this challenging problem can be tackled. The proposed method is experimentally evaluated using a real world database and the effectiveness and the potential in discovering collective information is demonstrated.