The graph clearly indicates that Intension Mining System is much more efficient than the existing approaches. And as the database grows, the efficiency increases. It neatly integrates all the steps of knowledge discovery process also. Our work is a modest attempt to develop a complete system for Intension Mining. Further work on this system will definitely yield better results. We propose the following future work in the same direction.
1. It is reasonably easy to design a knowledge discovery schema language for knowledge specification in relational databases. It is a great challenge to design languages for knowledge mining in other kinds of databases, such as transaction databases, objectoriented databases, spatial databases, multimedia databases, legacy databases, global information systems, etc. With the emerging activities for data mining in these databases, the design of schema languages for such mining tasks may become an important issue in future research.
2. A comprehensive Knowledge Discovery Query Language is needed to mine knowledge easily and efficiently from the Knowledge Concentrates.
3. More intensional algorithms need to be identified and integrated into the intension mining system.
Once the library of intension algorithms is made rich enough, our vision of a clean and integrated approach towards knowledge discovery in databases will be realized.