Database Reference
In-Depth Information
A Case-Based Data Mining Platform
Xingwen Wang and Joshua Zhexue Huang
E-Business Technology Institute,
The University of Hong Kong, Pokfulam Road, Hong Kong
{xwwang, jhuang}@eti.hku.hk
Abstract. Data mining practice in industry heavily depends on experienced data
mining professionals to provide solutions. Normal business users cannot easily
use data mining tools to solve their business problems, because of the complexity
of data mining process and data mining tools. In this paper, we propose a case-
based data mining platform, which reuses the knowledge captured in past data
mining cases to semi-automatically solve new similar problems. We first extend
generic data mining model for knowledge reuse. Then we define data mining
case. And then we introduce this platform in detail from its storage bases,
functional modules, user interface, and application scenario. Theoretically, this
platform can simplify data mining process, reduce the dependency on data mining
professional, and shorten business decision time.
Keywords: Data Mining, Knowledge Reuse, Case-Based Reasoning, Case-
Based Data Mining Platform.
1 Introduction
Data mining is a technique of extracting useful but implicit knowledge from large
amounts of data. It has been widely used to solve business problems, such as,
customer segmentation, customer retention, credit scoring, product recommendation,
direct marketing campaigns, cross selling, fraud detection, and so on [2]. These
problems are ubiquitous in most companies regardless of their size. Data mining has
been an important technique applied in current business decision.
Data mining process is not trivial. It consists of many steps, such as, business
problem definition, data collection, data preprocessing, modelling, and model
deployment [4]. In each step, different techniques may be applied. For example, during
the modelling, techniques such as association analysis, decision trees, neural networks,
regression, clustering, and time sequence analysis can be used. On the other hand, many
commercial data mining tools, such as, Clementine, Enterprise Miner, and Intelligent
Miner, have been widely used to solve data mining problems. Even though they have
provided user-friendly graphical interfaces to drag-and-drop algorithms to form a
processing flow, the prerequisite to successfully conduct a data mining process is that
the user should know what those algorithms can do, how to make use of them
sequentially, and how to set the parameters.
Because of the complexity of data mining process and data mining tools, normal
business users cannot easily use data mining tools to solve their business problems.
 
Search WWH ::




Custom Search