NACOD: A Naïve Associative Classifier for Online Data

The pedagogical orientation to the families of children with behavioral deficiencies needs a differentiated approach, accorded to the particular characteristics of each family. To accomplish this task, the personnel in charge of family orientation in the Schools for children with affective-behavioral maladies, diagnose the families, to obtain the peculiarities of the familiar dynamics. After that, they need to obtain groups of similar families, in order to carry out a better orientation. The data of the familiar dynamics are mixed an incomplete, and the desired group number is not given. In this paper we extend the  0-compact sets structuralization to cluster the families according to their characteristics. We found five groups, each of them with distinctive familiar dynamics. With the five groups, the family orientation specialists designed a personalized strategy with a more coherent and adequate orientation.

Section: B Second Stage Definition Of the Similarity Function To Comentioning

confidence: 99%

Using Extended Compact Sets to Cluster Educational Data

Rey-Benguría*

2019

“…In the case of dissimilarity, the HEOM function [75] was used for all algorithms . The reason for its use was its good results in the treatment of DMI [76][77][78]. The experiments were conducted as follows: For each database, the different algorithms were applied and Entropy was calculated.…”

Section:  mentioning

confidence: 99%

PANTSA Influence in grouping Mixed and Incomplete Data

Chávez-Castilla*

2019

Obtaining high quality groups and processing mixed and incomplete data (DMI) are still problems in the data clustering. Recently a method was proposed that improves the results obtained by clustering algorithms, the PAntSA; but this was only designed and tested for numerical data. For this reason, this paper analyzes the influence of applying the PAntSA in the performance of DMI restricted clustering algorithms. For this, the results of different algorithms are compared before and after applying the PAntSA. The comparisons made provide experimental evidence that the PAntSA algorithm improves the quality of the groups obtained by traditional DMI clustering methods.

“…For example, a customer description can include simultaneously attributes such as age (integer), sex (nominal), salary (real), educational degree (nominal), employed (Boolean), among others [10]. This type of object description is per say a challenge for any algorithm [11][12][13]. The lacking of a metric space makes impossible the definition of a sum operator and also the scalar multiplication [14][15][16][17][18][19][20][21][22][23].…”

Section: Introductionmentioning

confidence: 99%

Clustering mixed data using an Artificial Bee Colony

Cabrera-Venegas*,

Chávez-Castilla

2019

 Abstract: In this paper, we have proposed a clustering technique which optimizes the total compactness and separation (measured using the Silhouette index) of the clusters. The proposed algorithm uses an Artificial Bee Colony (ABC) based optimization method as the underlying optimization criterion. We used similarity based prototypes as cluster centers. The proposed clustering technique is able to suitably handle mixed and incomplete data types in such a way that the original characteristics of the data are preserved. Assignment of points to different clusters is done based on a dissimilarity function rather than the Euclidean distance. Results on real-life data sets show that the proposed technique is well-suited to detect true partitioning from data sets. Results are compared with those obtained by four existing clustering techniques, one genetic algorithm based clustering technique (AGKA), the k-Prototypes (KP) algorithm, well-known based K-means clustering technique for similarity functions (KMSF) and a newly developed algorithm with dissimilarity based clustering technique (AD2011).