Raudel Hernández-León

Classifying Using Specific Rules with High Confidence

In this paper, we introduce a new strategy for mining the set of Class Association Rules (CARs), that allows building specific rules with high confidence. Moreover, we introduce two propositions that support the use of a confidence threshold value equal to

Algorithms for mining frequent itemsets in static and dynamic datasets

0.5

Classification based on specific rules and inexact coverage

. We also propose a new way for ordering the set of CARs based on rule size and confidence values. Our results show a better average classification accuracy than those obtained by the best classifiers based on CARs reported in the literature.

On the design of hardware-software architectures for frequent itemsets mining on data streams

In this paper, two algorithms for mining frequent itemsets in large sparse datasets are proposed. The first one, named Compressed Arrays (CA), allows to process datasets that do not change along the time (static datasets) while the second one, based on the ideas of the former and named Dynamic Compressed Arrays (DCA), processes datasets that change along the time by adding/deleting transactions (dynamic datasets). Both algorithms introduce a novel way to use equivalence classes of itemsets by performing a breadth first search through them and by storing the class prefix support in compressed arrays, which allows fast itemset support computing. On the other hand, unlike previous algorithms for dynamic datasets that store the full dataset in main memory without reusing the current frequent itemsets, DCA algorithm stores the current frequent itemsets in binary files, grouped in equivalence classes, and reuses them to calculate the new frequent itemsets.

Studying Netconf in Hybrid Rule Ordering Strategies for Associative Classification

Association rule mining and classification are important tasks in data mining. Using association rules has proved to be a good approach for classification. In this paper, we propose an accurate classifier based on class association rules (CARs), called CAR-IC, which introduces a new pruning strategy for mining CARs, which allows building specific rules with high confidence. Moreover, we propose and prove three propositions that support the use of a confidence threshold for computing rules that avoids ambiguity at the classification stage. This paper also presents a new way for ordering the set of CARs based on rule size and confidence. Finally, we define a new coverage strategy, which reduces the number of non-covered unseen-transactions during the classification stage. Results over several datasets show that CAR-IC beats the best classifiers based on CARs reported in the literature.

A Novel Incremental Algorithm for Frequent Itemsets Mining in Dynamic Datasets

Frequent Itemsets Mining has been applied in many data processing applications with remarkable results. Recently, data streams processing is gaining a lot of attention due to its practical applications. Data in data streams are transmitted at high rates and cannot be stored for offline processing making impractical to use traditional data mining approaches (such as Frequent Itemsets Mining) straightforwardly on data streams. In this paper, two single-pass parallel algorithms based on a tree data structure for Frequent Itemsets Mining on data streams are proposed. The presented algorithms employ Landmark and Sliding Window Models for windows handling. In the presented paper, as in other revised papers, if the number of frequent items on data streams is low then the proposed algorithms perform an exact mining process. On the contrary, if the number of frequent patterns is large the mining process is approximate with no false positives produced. Experiments conducted demonstrate that the presented algorithms outperform the processing time of the hardware architectures reported in the state-of-the-art.

Approximate frequent itemsets mining on data streams using hashing and lexicographie order in hardware

In Associative Classification, building a classifier based on Class Association Rules (CARs) consists in finding an ordered CAR list by applying a rule ordering strategy. Since this CAR list will be used to build a classifier, it is important to develop a good rule ordering strategy. In this paper, we introduce four novel hybrid rule ordering strategies; the first three combine the Netconf measure with Support-Confidence based rule ordering strategies. The fourth strategy, called Hybrid Specific Rules/Netconf (SR/NF), combines the Netconf measure with a rule ordering strategy based on the CAR’s size. The experiments show that the proposed strategies obtain better classification accuracy than the best ordering strategies reported in the literature.

A Novel Hybrid Data Reduction Strategy and Its Application to Intrusion Detection

Frequent Itemsets (FI) Mining is one of the most researched areas of data mining. When some new transactions are appended, deleted or modified in a dataset, updating FI is a nontrivial task since such updates may invalidate existing FI or introduce new ones. In this paper a novel algorithm suitable for FI mining in dynamic datasets named Incremental Compressed Arrays is presented. In the experiments, our algorithm was compared against some algorithms as Eclat, PatriciaMine and FP-growth when new transactions are added or deleted.

Improving the Accuracy of the Sequential Patterns-Based Classifiers

Frequent Itemsets Mining is a Data Mining technique that has been employed to extract useful knowledge from datasets; and recently, from data streams. Data streams are an unbounded and infinite flow of data arriving at high rates; therefore, traditional Data Mining approaches for Frequent Itemsets Mining cannot be used straightforwardly. Finding alternatives to improve the discovery of frequent itemsets on data streams is an active research topic. This paper introduces the first hardware-based algorithm for such task. It uses the top-k frequent 1-itemsets detection, hashing and the lexicographic order of received items. Experimental results demonstrate the viability of the proposed method.

Improving the Accuracy of CAR-based Classifiers by Combining Netconf Measure and Dynamic\(-K\) Mechanism

The presence of useless information and the huge amount of data generated by telecommunication services can affect the efficiency of traditional Intrusion Detection Systems (IDSs). This fact encourage the development of data preprocessing strategies for improving the efficiency of IDSs. On the other hand, improving such efficiency relying on the data reduction strategies, without affecting the quality of the reduced dataset (i.e. keeping the accuracy during the classification process), represents a challenge. Also, the runtime of commonly used strategies is usually high. In this paper, a novel hybrid data reduction strategy is presented. The proposed strategy reduces the number of features and instances in the training collection without greatly affecting the quality of the reduced dataset. In addition, it improves the efficiency of the classification process. Finally, our proposal is favorably compared with other hybrid data reduction strategies.