GAM: a guidance enabled association mining environment
Aaron Ceglar, John F. Roddick · International Journal of Business Intelligence and Data Mining · 2007
The quality of data mining results is largely dependent on the ability to accommodate context and user requirements within the mining process. This is done effectively within the pre-processing and presentation stages, however the analysis (or mining) stage remains relatively autonomous and opaque with user input commonly limited to parameter setting. There is, at present, no direct manipulation of the analysis stage which results in the analysis of the domain space being statically constrained. This reduces the quality of results and increases the time needed for analysis. This paper presents a guided association mining environment, GAM, that enhances user-computer synergy by incorporating the user at fine level of granularity within the analysis stage. GAM extends the current state of the art and is based upon a generic guided knowledge discovery environment.