Fuzzy-rough Set-based Feature Selection Research Articles

Using fuzzy-rough set evaluation for feature selection and Naive Bayes to classify the Parkinson disease

Feature selection is one of the issues in machine learning as well as statistical pattern recognition. This is important in many fields (such as classification) because there are many features in these areas, many of which are either unused or have little information load. Not eliminating these features does not make a problem in terms of information, but it does increase the computational burden for the intended application. Besides, it causes to store of so much useless information along with useful data. A problem for machine learning research occurs when there are many possible features with few attributes of training data. One way is to first specify the best attributes for prediction and then to classify features based on a measure of their dependence. In this study, the Fuzzy- Rough subset evaluation has been used to take features in core of similar features. Fuzzy-rough set-based feature selection (FS) has been demonstrated to be extremely advantageous at reducing dataset size but has various problems that yield it unproductive for big datasets. Fuzzy- Rough subset evaluation algorithm indicates that the techniques greatly decrease dimensionality while keeping classification accuracy. This paper considers classifying attributes by using fuzzy set similarity measures as well as the dependency degree as a relatedness measure. Here we use Artificial Neural Network, Naive Bayes as classifiers, and the performance of these techniques are compared by accuracy, precision, recall, and F-measure metrics.

An Efficient Classification Model using Fuzzy Rough Set Theory and Random Weight Neural Network

In the area of fuzzy rough set theory (FRST), researchers have gained much interest in handling the high-dimensional data. Rough set theory (RST) is one of the important tools used to pre-process the data and helps to obtain a better predictive model, but in RST, the process of discretization may loss useful information. Therefore, fuzzy rough set theory contributes well with the real-valued data. In this paper, an efficient technique is presented based on Fuzzy rough set theory (FRST) to pre-process the large-scale data sets to increase the efficacy of the predictive model. Therefore, a fuzzy rough set-based feature selection (FRSFS) technique is associated with a Random weight neural network (RWNN) classifier to obtain the better generalization ability. Results on different dataset show that the proposed technique performs well and provides better speed and accuracy when compared by associating FRSFS with other machine learning classifiers (i.e., KNN, Naive Bayes, SVM, decision tree and backpropagation neural network).

Machine learning approach for wart treatment selection: prominence on performance assessment

Warts are benign tumors, caused by human papillomavirus (HPV). The present study mainly emphasis on the selection of suitable methods for the removal of a common and plantar wart. There are numerous wart treatment methods for this disease, among them cryotherapy and immunotherapy are well-known approaches. Identifying the suitable wart treatment method manually is quite challenging. Moreover, existing machine learning (ML) techniques show a poor prediction accuracy towards the selection of wart treatment method, however, the prediction accuracy is not satisfactory and can be further improved. To achieve the same, the current study utilizes the advantage of fuzzy rough set based feature selection (FRFS) to generate the most optimal informative feature space, which in turn makes the ML algorithms more accurate and leads to a better prognosis. The proposed FRFS based Naive Bayes and FRFS based CART models outperform from the existing model in terms of prediction accuracy.

Adaptive intrusion detection via GA-GOGMM-based pattern learning with fuzzy rough set-based attribute selection

In this paper, an adaptive network intrusion detection method using fuzzy rough set-based feature selection and GA-GOGMM-based pattern learning is presented. Based on the fuzzy rough set theory, the optimal attribute subset of network connection records is achieved by the information gain ratio criterion in advance. A greedy algorithm-based global optimal Gaussian mixture model (GMM) clustering method, termed GA-GOGMM, is introduced, to extract the intrinsic structure of network instances to achieve highly-discernable and stable normal and intrusion pattern libraries for the subsequent network intrusion detection (NID). GA-GOGMM-based pattern learning can achieve the optimal GMM of network traffic instances for the pattern clustering while avoiding the negative effect of the empirical initialization of clustering numbers and random initialization of clustering centers with a low computational complexity. An adaptive model updating mechanism is further introduced for the online updating of normal and intrusion pattern libraries to ensure the adaptability of the NID model. Extensive validation and comparative experiments, conducted on a benchmark dataset NSL-KDD and a self-built Nidsbench-based network simulation platform, show that the proposed ANID approach leads to a significant improvement in detection accuracies with low false alarms and missing reports on both known and unknown attacks. It can effectively adapt to the dynamic changing network environments with high detection accuracy and low false alarm rate as well as low missing reporting rate, which has significant application prospects.

Intelligent temporal classification and fuzzy rough set-based feature selection algorithm for intrusion detection system in WSNs

At present, Internet-based information processing systems are challenged by different kinds of threats, which lead to various types of damages that in turn result in significant loss of information in Wireless Sensor Networks (WSNs). Moreover, the stream of data communication through WSNs is continuously increasing in volume. Such huge amounts of data are targeted by attackers frequently, and thus it is crucial to provide high-level security to ensure safe and effective communication of information through the Internet. In order to secure data communication over the Internet through WSNs, Intrusion Detection Systems (IDSs) must be provided as an essential component in networks, databases and cloud systems. Over the previous decade, various approaches were utilized for developing IDSs. Unfortunately, however,none of these systems are completely flawless due to uncertainty surrounding network traffic created by normal users and attackers. Hence, the need to develop efficient IDSs has increased continuously. The present study proposes an adaptive IDS based on Fuzzy Rough sets for attribute selection and Allen’s interval algebra, which is applied on network trace datasets in order to select a huge number of attack data for effective prediction of attacks in WSNs. In addition, a fuzzy and rough set based nearest neighborhood algorithm (FRNN) is proposed in this article for effective classification of network trace dataset. This model uses a biased dataset that has 50:50 normal and attack data as opposed to the conventional datasets that have 80:20 normal and attack data. The efficiency of the proposed IDS is increased due to the use of biased data. The combination of feature selection, temporal-based dataset selection, and classification using a biased dataset reduces the false alarm rate and increases detection accuracy.

A fuzzy rough set-based feature selection method using representative instances

The fuzzy rough set theory has been widely used to deal with uncertainty in real-valued or even complex data, in which one of the most concerned issues is feature selection. Since a real-world data set generally contains redundant data objects (or instances) and errors which lead to the fact that not all the instances are of equal importance, focusing on the representative instances for feature selection can not only acquire more convincing analysis results but also alleviate computational complexity in mining knowledge. At present, however, little attention has been paid on using representative instances to select features. In this paper, the issue of selecting features by using representative instances is investigated based on fuzzy rough sets and a representative instance-based feature selection approach is proposed. First, the fuzzy granular rule is employed to describe the discriminating information of an instance. Then, the representative instances are selected according to the coverage ability of the fuzzy granular rules induced by all of the instances. Furthermore, an implication relationship-preserved reduction is presented to maintain the discriminating information of the selected instances, and then a heuristic algorithm is presented to search for such a feature subset. Finally, a filter-wrapper approach is suggested to select the best subset of the features. Some numerical experiments are further conducted to show the performance of the proposed feature selection method and the results are satisfactory in terms of both efficiency and effectiveness.