Rough Support Vector Machine for Classification with Interval and Incomplete Data

Abstract

The paper presents the idea of connecting the concepts of the Vapnik's support vector machine with Pawlak's rough sets in one classification scheme. The hybrid system will be applied to classifying data in the form of intervals and with missing values [1]. Both situations will be treated as a cause of dividing input space into equivalence classes. Then, the SVM procedure will lead to a classification of input data into rough sets of the desired classes, i.e. to their positive, boundary or negative regions. Such a form of answer is also called a three-way decision. The proposed solution will be tested using several popular benchmarks.

Algorithmic Trading and Efficiency of the Stock Market in Poland

“…It is also worth noting that the algorithmic trade can be further enriched and extended with elements of new artificial intelligence methods e.g. fuzzy sets (Starczewski et al, 2020), neural networks (Bilski et al, 2020), particle swarm optimization (Dziwiński et al, 2020), rough support vector machines (Nowicki et al, 2019) or perhaps so-called ensemble techniques (Homenda et al, 2020).…”

Performance Analysis of Rough Set–Based Hybrid Classification Systems in the Case of Missing Values

“…The rough set-based classification systems considered in the paper have been described in detail in [21] and [22]. All systems have been defined using a common form of input and output data as well as notation.…”

“…the neighbour in the rough knearest neighbour classifier and the result of classification in each of the systems), or approximate set in the sense of Dubois and Prade [3,4]. The works published so far [21,22] presented the concept of rough set-based classification Systems, together with specific examples of such rough support vector machines, a rough k-nearest neighbour classifier, a rough neural network, and a rough fuzzy system. Individual The publications described their operation as independent classifiers [18,19,22], elements of an iterative system [21] or elements of assemblies [9,10,21,23].…”

“…The works published so far [21,22] presented the concept of rough set-based classification Systems, together with specific examples of such rough support vector machines, a rough k-nearest neighbour classifier, a rough neural network, and a rough fuzzy system. Individual The publications described their operation as independent classifiers [18,19,22], elements of an iterative system [21] or elements of assemblies [9,10,21,23]. The research results presented classification tasks in which the number of conditional attributes usually did not exceed 10.…”

“…rough support vector machines, a rough k-nearest neighbour classifier, a rough neural network, and a rough fuzzy system. More details have been presented in the works already published [18,19,20,21,22].…”