Building a new taxonomy for data discretization techniques

Bakar, Azuraliza Abu; Othman, Zulaiha Ali; Shuib, Liyana

doi:10.1109/dmo.2009.5341896

Cited by 25 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…At each step, they find the best candidate boundary to be used as a cut point and afterwards the rest of the decisions are made accordingly. Incremental discretizers are also known as hierarchical discretizers [9]. Both types of discretizers are widespread in the literature, although there is usually a more defined relationship between incremental and supervised ones.…”

Section: Main Characteristics Of a Discretizermentioning

confidence: 99%

“…The identification of the best discretizer for each situation is a very difficult task to carry out, but performing exhaustive experiments considering a representative set of learners and discretizers could help to make the best choice. Some reviews of discretization techniques can be found in the literature [9,36,75,123]. However, the characteristics of the methods are not studied completely, many discretizers, even classic ones, are not mentioned, and the notation used for categorization is not unified.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discretization

García

Luengo

Herrera

2014

Intelligent Systems Reference Library

View full text Add to dashboard Cite

Discretization is an essential preprocessing technique used in many knowledge discovery and data mining tasks. Its main goal is to transform a set of continuous attributes into discrete ones, by associating categorical values to intervals and thus transforming quantitative data into qualitative data. An overview of discretization together with a complete outlook and taxonomy are supplied in Sects. 9.1 and 9.2. We conduct an experimental study in supervised classification involving the most representative discretizers, different types of classifiers, and a large number of data sets (Sect. 9.4).

show abstract

Section: Main Characteristics Of a Discretizermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Discretization

García

Luengo

Herrera

2014

Intelligent Systems Reference Library

View full text Add to dashboard Cite

show abstract

“…The three of them will keep the numerical character of the output value (we are not discretising the problem). Bakar et al (Bakar et al, 2009) present an exhaustive taxonomy for data discretisation techniques. We will first explore two of the simplest and best-known unsupervised methods: Equal Frequency intervals (EF ) and Equal Width intervals (EW ) (Dougherty et al, 1995).…”

Section: Methods Based On Segmentationmentioning

confidence: 99%

Aggregative quantification for regression

Bella

Ferri

Hernández-Orallo

et al. 2013

Data Min Knowl Disc

View full text Add to dashboard Cite

The problem of estimating the class distribution (or prevalence) for a new unlabelled dataset (from a possibly different distribution) is a very common problem which has been addressed in one way or another in the past decades. This problem has been recently reconsidered as a new task in data mining, renamed quantification when the estimation is performed as an aggregation (and possible adjustment) of a single-instance supervised model (e.g., a classifier). However, the study of quantification has been limited to classification, while it is clear that this problem also appears, perhaps even more frequently, with other predictive problems, such asregression. In this case, the goal is to determine a distribution or an aggregated indicator of the output variable for a new unlabelled dataset. In this paper, we introduce a comprehensive new taxonomy of quantification tasks, distinguishing between the estimation of the whole distribution and the estimation of some indicators (summary statistics), for both classification and regression. This distinction is especially useful for regression, since predictions are numerical values that can be aggregated in many different ways, as in multi-dimensional hierarchical data warehouses. We focus on aggregative quantification for regression and see that the approaches borrowed from classification do not work. We present several techniques based on segmentation which are able to produce accurate estimations of the expected value and the distribution of the output variable. We show experimentally that these methods especially excel for the relevant scenarios where training and test distributions dramatically differ.

show abstract

“…The majority of the methods in time series data mining assume that the time series are discrete [1], [2] and [3]. Nevertheless, most applications generate and use floatingpoint data type.…”

Section: Introductionmentioning

confidence: 99%

“…The method processes a single time series at a time, so that the discretization criterion is not generalized to the complete dataset. represents time series values as a multi connected graph [2] [3]. Under this representation, similar time series are grouped into a graphical model.…”

Section: Introductionmentioning

confidence: 99%

Harmony Search algorithm for optimal word size in symbolic time series representation

Ahmed

Bakar

Hamdan

2011

2011 3rd Conference on Data Mining and Optimization (DMO)

Self Cite

View full text Add to dashboard Cite

Fast and high quality time series representation is a crucial task in data mining pre-pre-processing. Recent studies have shown that most representation methods based on improving classification accuracy and compress data sets rather than maximize data information. We attempt to improve the number of SAX (time series representation method) word size and alphabet size by searching for the optimal word size. In this paper we propose a new representation algorithm (HSAX) that deals with Harmony Search algorithm (HS) to explore optimal word size (Ws) and alphabet size (a) for SAX time series. Harmony search algorithm is an optimization algorithm that generates randomly solutions (Ws, a) and select two best solutions. H SAX algorithm is developed to maximize information, rather than improve classification accuracy. We have applied HSAX algorithm on some standard time series data sets. We also compare the HSAX with other meta-heuristic GENEBLA and original SAX algorithms The experimental results showed that the HSAX Algorithm compare to SAX manage to generate more word size and achieve less error rates, whereas HSAX compared to GENEBLA the quality of error rate is comparable with the advantage that HSAX generated high number of word and alphabet size .

show abstract

Building a new taxonomy for data discretization techniques

Cited by 25 publications

References 33 publications

Discretization

Discretization

Aggregative quantification for regression

Harmony Search algorithm for optimal word size in symbolic time series representation

Contact Info

Product

Resources

About