Visual Data Mining Using Monotone Boolean Functions
نویسندگان
چکیده
This paper describes a new technique for extracting patterns and relations visually from multidimensional binary data using monotone Boolean functions. Visual Data Mining (VDM) has shown benefits in many areas when used with numerical data, but that technique is less beneficial for binary data. This problem is especially challenging in medical applications tracked with binary symptoms. The proposed method relies on monotone structural relations between Boolean vectors in the n-dimensional binary cube, E, and visualizes them in 2-D or 3-D as chains of Boolean vectors. Actual Boolean vectors are laid out on this chain structure. Currently the system supports two visual forms: the multiple disk form (MDF) and the “Yin/Yang” form (YYF). In the MDF, every vector has a fixed horizontal and vertical position. In the YYF, only the vertical position is fixed. The method is illustrated with an example from a breast cancer diagnosis based on mammographic X-ray images.
منابع مشابه
Visual Data Mining and Discovery in Multivariate Data Using Monotone n-D Structure
Visual data mining (VDM) is an emerging research area of Data Mining and Visual Analytics to gain a deep visual understanding of data. A border between patterns can be recognizable visually, but its analytical form can be quite complex and difficult to discover. VDM methods have shown benefits in many areas, but these methods often fail in visualizing highly overlapped multidimensional data and...
متن کاملUbiquitous monotone and almost-monotone Boolean functions in systems biology
Systems Biology is quickly becoming a major focus of applied and computational mathematics research. While traditionally numerics and continuous modeling techniques have been used, in recent years it has become apparent that qualitative modeling, data quantization and data mining questions play a central role for the field. Monotone Boolean functions are ubiquitous in these models, and, surpris...
متن کاملOn Monotone Data Mining Languages
We present a simple Data Mining Logic (DML) that can express common data mining tasks, like “Find Boolean association rules” or “Find inclusion dependencies.” At the center of the paper is the problem of characterizing DML queries that are amenable to the levelwise search strategy used in the a-priori algorithm. We relate the problem to that of characterizing monotone first-order properties for...
متن کاملAlmost all monotone Boolean functions are polynomially learnable using membership queries
We consider exact learning or identification of monotone Boolean functions by only using membership queries. It is shown that almost all monotone Boolean functions are polynomially identifiable in the input number of variables as well as the output being the sum of the sizes of the CNF and DNF representations. 2001 Elsevier Science B.V. All rights reserved.
متن کاملMonotone Boolean Functions Are Universal Approximators
The employment of proper codings, such as the base-2 coding, has allowed to establish the universal approximation property of Boolean functions: if a sufficient number b of inputs (bits) is taken, they are able to approximate arbitrarily well any real Borel measurable mapping. However, if the reduced set of monotone Boolean functions, whose expression involves only and and or operators, is cons...
متن کامل