Wyniki wyszukiwania - BazTech

1

An experimental evaluation of two approaches to mining context based sequential patterns

Stefanowski J., Ziembiński R.

Control and Cybernetics

|

2009

|

Vol. 38, no 1

27-45

EN

The paper discusses the results of experiments with a new context extension of a sequential pattern mining problem. In this extension, two kinds of context attributes are introduced for describing the source of a sequence and for each element inside this sequence. Such context based sequential patterns may be discovered by a new algorithm, called Context Mapping Improved, specific for handling attributes with similarity functions. For numerical attributes an alternative approach could include their pre-discretization, transforming discrete values into artificial items and, then, using an adaptation of an algorithm for mining sequential patterns from nominal items. The aim of this paper is to experimentally compare these two approaches to mine artificially generated sequence databases with numerical context attributes where several reference patterns are hidden. The results of experiments show that the Context Mapping Improved algorithm has led to better re-discovery of reference patterns. Moreover, a new measure for comparing two sets of context based patterns is introduced.

2

Algorithms for Context Based Sequential Pattern Mining

Ziembiński R.

Fundamenta Informaticae

|

2007

|

Vol. 76, nr 4

495-510

EN

This paper describes practical aspects of a novel approach to the sequential pattern mining named Context Based Sequential Pattern Mining (CBSPM). It introduces a novel ContextMapping algorithm used for the context pattern mining and an illustrative example showing some advantages of the proposed method. The approach presented here takes into consideration some shortcomings of the classic problem of the sequential pattern mining. The significant advantage of the classic sequential patterns mining is simplicity. It introduces simple element construction, built upon set of atomic items. The comparison of sequence's elements utilizes simple inclusion of sets. But many practical problems like web event mining, monitoring, tracking and rules generation often require mining more complex data. The CBSPM takes into account non nominal attributes and similarity of sequence's elements. An approach described here extends traditional problem adding a vector of context attributes of any kind to sequences and sequence’s elements. Context vectors contain details about sequence's and element's origin. The mining process results in context patterns containing additional, valuable context information useful in interpretation of patterns origin.

3

An empirical study of context based sequential pattern mining algorithms efficiency

Ziembiński R.

Foundations of Computing and Decision Sciences

|

2007

|

Vol. 32, No. 1

63-84

EN

Methods of patterns detection in the sets of data are useful and demanded tools in a knowledge discovery process. The problem of searching patterns in set of sequences is named Sequential Patterns Mining. It can be defined as a way of finding frequent subsequences in the sequences database. The patterns selection procedure may be simply understood. Every subsequence must be enclosed in the required number of sequences from the database at least to become a pattern. The number of a pattern enclosing sequences is called a pattern support. The process of finding patterns may look trivial but its efficient solution is not. The efficiency plays a crucial role if the required support is lowered. The number of mined patterns may grow exponentially. Moreover, the situation may change if the problem of Sequential Patterns Mining will be extended further. In the classic definition the sequence is a list of ordered elements containing only non-empty sets of items. The Context Based Sequential Patterns Mining adds uniform and multi-attribute contexts (vectors) to the elements of the sequence and the sequence itself. Introducing contexts significantly enlarges the problem search space. However, it brings some additional occasions to constrain the mining process, too. This enhancement requires new algorithms. Traditional ones are not able to cope with non-nominal data directly. Algorithms derived straightly from traditional algorithms were verified to be inefficient. This study evaluates efficiency of novel ContextMapping and ContextMappingHeuristic algorithms. These innovative algoritnms are designed to solve the problem of Context Based Sequential Pattern Mining. This study answers in what scope the algorithms parameterization impacts on mining costs and accuracy. It also refers the modified problem to the traditional one pointing at the common and uncommon properties and drawing perspective for further research.