Data Mining with C4.5 and Interactive Cartographic Visualization
نویسندگان
چکیده
In the paper we consider application of techniques of knowledge discovery in databases (KDD) to spatially referenced data. We propose to combine such techniques with various methods of interactive classification of spatial objects supported by map displays. Maps help both to prepare source data for KDD procedures and to interpret results of their application. The latter function is realized through dynamic linking of spatial and non-
منابع مشابه
GIS Visualization Support to the C4.5 Classification Algorithm of KDD
Methods of Knowledge Discovery in Databases (KDD) are used for finding regularities and dependencies in collections of attribute data. These methods can be applied, in particular, to data referring to spatial objects. However, the general KDD methods do not make any respect of the spatial aspect of the data, and results of application of these methods are completely aspatial. Hence, to enable e...
متن کاملTowards Simple, Easy to Understand, an Interactive Decision Tree Algorithm
Data mining is intended to extract hidden useful knowledge from large datasets in a given application. This usefulness relates to the user goal, in other words only the user can determine whether the resulting knowledge answers his goal. Therefore, data mining tool should be highly interactive and participatory. This paper presents an interactive decision tree algorithm using visualization meth...
متن کاملSpace-Time-Attribute Analysis and Visualization of U.S. Company Data
This research integrates computational, visual, and cartographic methods to develop geo-visual analytic strategies through which analysts can detect and explore multivariate, spatio-temporal patterns. The paper introduces a new form of geographic small multiple display along with a novel integration of computational and visual methods, for clustering and sorting large, multivariate data sets an...
متن کاملIntelligent Icon Positioning for Interactive Map-based Information Systems
The combination of map displays with icon techniques is well-suited for visualizing geospatial dependent multivariate data. The requirements for icon placement on the map, although not trivial, are not properly attended by most interactive visualization systems. In cartography, a number of methods for automatic text label placement and map generalization has been developed. These are focused on...
متن کاملExploiting Visualization in Knowledge Discovery
Todate visualization has not been extensively harnessed in knowledge discovery in databases (KDD). In this paper, we show that a multidimensional visualization (MDV) technique can be used synergistically with a machine learning program like C4.5 to uncover new knowledge. Used together, the two approaches span the KDD spectrum between complete automation on one hand and fully manual on the other...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1999