An Architecture for Integrated Online Analytical Mining
Journal of Emerging Technologies in Web Intelligence
Online Analytical Processing (OLAP) technology is an essential element of the decision support system and permits decision makers to visualize huge operational data for quick, consistent, interactive and meaningful analysis. More recently, data mining techniques are also used together with OLAP to analyze large data sets which makes OLAP more useful and easier to apply in decision support systems. Several works in the past proved the likelihood and interest of integrating OLAP with data mining
... nd as a result a new promising direction of Online Analytical Mining (OLAM) has emerged. In this paper, a variety of OLAM architectures in the literature were reviewed and the limitations in the previously reported work have been identified. Literature review reveals the fact that none of the previously reported OLAM architectures have integrated enhanced OLAP with data mining. We enhanced the performance of OLAP in terms of cube construction time and visualization by providing interactive visual exploration of data cube. Furthermore, automatic OLAP schema generation has never been introduced as a component of OLAM architecture. The aim of this paper is to propose an integrated OLAM architecture that not only overcomes the existing limitations but also extends the architecture by adding an automation layer for the schema generation. In the proposed work, hierarchical clustering has been used as the data mining technique and three types of schemas namely star, snowflake and galaxy were automatically generated. A prototype of automatic schema generation has been developed and schema generation algorithms have been provided. In addition to this, we implemented and deployed the proposed architecture. Validation has been done by performing experiments on real life data set. Experimental results prove that the proposed architecture improves the cube construction time, empowers interactive data visualization, automate schema generation, and enable targeted and focused analysis at the front-end. By integrating enhanced OLAP with data mining system a higher degree of development is achieved which makes significant advancement in the modern OLAM architectures.