Attribute Selection Approach Research Articles

Enhancing machine learning-based seismic facies classification through attribute selection: application to 3D seismic data from the Malay and Sabah Basins, offshore Malaysia

Over the past few years, the use of machine learning has gained considerable momentum in many industries, including exploration seismic. While supervised machine learning is increasingly being used in seismic data analysis, some obstacles hinder its widespread application. Seismic facies classification—a crucial aspect in this field—particularly faces challenges such as the selection of appropriate input attributes. Plethora of seismic attributes have been created over the years, and new ones are still coming out. Yet, several have been deemed redundant or geologically meaningless. In the context of machine learning, it is crucial to avoid these redundant and irrelevant attributes as they can result in overfitting, building unnecessary complex models, and prolonging computational time. The current study incorporates an attribute selection approach to seismic facies classification and evaluates the importance of several available seismic attributes. Two datasets from the AN Field and the Dangerous Grounds region offshore Malaysia were utilized. Several attribute selection techniques were evaluated, with most of them yielding perfect attribute subsets for the AN dataset. However, only the wrapper and embedded methods could produce optimal subsets for the more complex Dangerous Grounds dataset. In both datasets, distinguishing the targeted seismic facies was mainly dependent on amplitude, spectral, and gray-level co-occurrence matrix attributes. Furthermore, spectral magnitude components played a significant role in classifying the facies of the Dangerous Grounds broadband data. The study demonstrated the importance of attribute selection, established a workflow, and identified significant attributes that could enhance seismic facies classification in Malaysian basins and similar geologic settings.

Deep Learning-Inspired IoT-IDS Mechanism for Edge Computing Environments.

The Internet of Things (IoT) technology has seen substantial research in Deep Learning (DL) techniques to detect cyberattacks. Critical Infrastructures (CIs) must be able to quickly detect cyberattacks close to edge devices in order to prevent service interruptions. DL approaches outperform shallow machine learning techniques in attack detection, giving them a viable alternative for use in intrusion detection. However, because of the massive amount of IoT data and the computational requirements for DL models, transmission overheads prevent the successful implementation of DL models closer to the devices. As they were not trained on pertinent IoT, current Intrusion Detection Systems (IDS) either use conventional techniques or are not intended for scattered edge-cloud deployment. A new edge-cloud-based IoT IDS is suggested to address these issues. It uses distributed processing to separate the dataset into subsets appropriate to different attack classes and performs attribute selection on time-series IoT data. Next, DL is used to train an attack detection Recurrent Neural Network, which consists of a Recurrent Neural Network (RNN) and Bidirectional Long Short-Term Memory (LSTM). The high-dimensional BoT-IoT dataset, which replicates massive amounts of genuine IoT attack traffic, is used to test the proposed model. Despite an 85 percent reduction in dataset size made achievable by attribute selection approaches, the attack detection capability was kept intact. The models built utilizing the smaller dataset demonstrated a higher recall rate (98.25%), F1-measure (99.12%), accuracy (99.56%), and precision (99.45%) with no loss in class discrimination performance compared to models trained on the entire attribute set. With the smaller attribute space, neither the RNN nor the Bi-LSTM models experienced underfitting or overfitting. The proposed DL-based IoT intrusion detection solution has the capability to scale efficiently in the face of large volumes of IoT data, thus making it an ideal candidate for edge-cloud deployment.

Intrusion detection systems for wireless sensor networks using computational intelligence techniques

Network Intrusion Detection Systems (NIDS) are utilized to find hostile network connections. This can be accomplished by looking at traffic network activity, but it takes a lot of work. The NIDS heavily utilizes approaches for data extraction and machine learning to find anomalies. In terms of feature selection, NIDS is far more effective. This is accurate since anomaly identification uses a number of time-consuming features. Because of this, the feature selection method influences how long it takes to analyze movement patterns and how clear it is. The goal of the study is to provide NIDS with an attribute selection approach. PSO has been used for that purpose. The Network Intrusion Detection System that is being developed will be able to identify any malicious activity in the network or any unusual behavior in the network, allowing the identification of the illegal activities and safeguarding the enormous amounts of confidential data belonging to the customers from being compromised. In the research, datasets were produced utilising both a network infrastructure and a simulation network. Wireshark is used to gather data packets whereas Cisco Packet Tracer is used to build a network in a simulated environment. Additionally, a physical network consisting of six node MCUs connected to a laptop and a mobile hotspot, has been built and communication packets are being recorded using the Wireshark tool. To train several machine learning models, all the datasets that were gathered—created datasets from our own studies as well as some common datasets like NSDL and UNSW acquired from Kaggle—were employed. Additionally, PSO, which is an optimization method, has been used with these ML algorithms for feature selection. In the research, KNN, decision trees, and ANN have all been combined with PSO for a specific case study. And it was found demonstrated the classification methods PSO + ANN outperformed PSO + KNN and PSO + DT in this case study.

Smart screening system for breast cancer: the use of thermographic images, evolutionary algorithms, and oversampling.

Purpose.This study aimed to develop a computer system for automatic detection of thermographic changes indicating breast malignancy risk.Materials and methods. The database contained 233 thermograms of women, including 43 with malignant lesions and 190 with no malignant lesions. Five classifiers were evaluated (k-Nearest Neighbor, Support Vector Machine, Decision Tree, Discriminant Analysis, and Naive Bayes) in combination with oversampling techniques. An attribute selection approach using genetic algorithms was considered. Performance was assessed using accuracy, sensitivity, specificity, AUC, and Kappa statistics.Results. Support vector machines combined with attribute selection by genetic algorithm and ASUWO oversampling obtained the best performance. Attributes were reduced by 41.38%, and accuracy was 95.23%, sensitivity was 93.65%, and specificity was 96.81%. The Kappa index was 0.90, and AUC was 0.99.Conclusion. The feature selection process lowered computational costs and improved diagnostic accuracy. A high-performance system using a new breast imaging modality could positively aid breast cancer screening.

Modified Multi-Criteria Decision Making Methods to Assess Classification Methods

During the last few decades, many academic and professional groups gave attention to adopting the multi-criteria decision-making methods in a variety of contexts for decision-making that are given to the diversity and sophistication of their selections. Five different classification methods are tested and assessed in this paper. Each has its own set of five attribute selection approaches. By using the multi-criteria decision-making procedures, these data can be used to rate options. Technique for order of preference by similarity to ideal solution (TOPSIS) is designed utilizing a modified fuzzy analytic hierarchy process (MFAHP) to compute the weight alternatives for TOPSIS in order to obtain the confidence value of each classifier for each feature selection approach individually. Defuzzification of the fuzzy values to obtain the final criteria weights, the rank function is used. The modification of TOPSIS is assessed in tests using five prediction models (alternatives) and six performance measurements (criteria) to analyze the German credit data sets. Overall the results of the experiment show that the proposed strategies are successful in credit approval data.

Intrusion Detection Using an Improved Cuckoo Search Optimization Algorithm

These days, intelligent cybersecurity models based on machine learning and data mining techniques are prevalent. Several factors might affect the quality of these models, including the accuracy, the ability to train new models quickly, the quick decision-making process, the simplicity of the created models, and the model’s interpretability. Feature selection algorithms can help achieve all these characteristics by isolating the crucial features from the unimportant ones during the model creation phase. The current article proposes an intelligent intrusion detection model based on an improved cuckoo search algorithm which is a nature-inspired optimization algorithm. The improved cuckoo search algorithm proposed in this article may tolerate several bad steps toward determining the set of effective features that would preserve or maximize the capabilities of the produced classification models. The generalization ability of such an algorithm is examined by applying it to 10 benchmark datasets, and it showed superior outcomes compared with several nature-inspired attribute selection approaches. Later, the improved cuckoo search algorithm is used to develop intelligent intrusion detection systems using the well-known “NSL-KDD” dataset. The obtained outcomes are appealing regarding the general performance, the time required to develop the intelligent intrusion detection models, and the number of rules generated.

A Decisive Metaheuristic Attribute Selector Enabled Combined Unsupervised-Supervised Model for Chronic Disease Risk Assessment.

Advanced predictive analytics coupled with an effective attribute selection method plays a pivotal role in the precise assessment of chronic disorder risks in patients. Traditional attribute selection approaches suffer from premature convergence, high complexity, and computational cost. On the contrary, heuristic-based optimization to supervised methods minimizes the computational cost by eliminating outlier attributes. In this study, a novel buffer-enabled heuristic, a memory-based metaheuristic attribute selection (MMAS) model, is proposed, which performs a local neighborhood search for optimizing chronic disorders data. It is further filtered with unsupervised K-means clustering to remove outliers. The resultant data are input to the Naive Bayes classifier to determine chronic disease risks' presence. Heart disease, breast cancer, diabetes, and hepatitis are the datasets used in the research. Upon implementation of the model, a mean accuracy of 94.5% using MMAS was recorded and it dropped to 93.5% if clustering was not used. The average precision, recall, and F-score metric computed were 96.05%, 94.07%, and 95.06%, respectively. The model also has a least latency of 0.8 sec. Thus, it is demonstrated that chronic disease diagnosis can be significantly improved by heuristic-based attribute selection coupled with clustering followed by classification. It can be used to develop a decision support system to assist medical experts in the effective analysis of chronic diseases in a cost-effective manner.

A data-level fusion model for unsupervised attribute selection in multi-source homogeneous data

Information fusion refers to derive an overall precise description of data by using certain fusion technique for utilizing the complementary information from multiple sources of data, which can facilitate effective decision-making, prediction and classification, etc. Multi-source homogeneous data, characterizing the data type of variables in the sample in form of one type (i.e., numerical or categorical) in different information sources, which widely exists in many practical applications. This paper concentrates on efficient fusion of multi-source homogeneous data with a data-level fusion model which involves the consolidation of multiple information sources and unsupervised attribute selection of the fused data. A unified description and modeling method of a multi-source homogeneous information system is introduced. The neighborhood rough sets model is used to construct the neighborhood granular structure, which uses the idea of granular computing to build methods of uncertainty measures. Given the uncertainty of fusing multiple information sources, Sup–Inf fusion functions are developed based on the proposed uncertainty measures, which can fuse the multi-source homogeneous information system into a single-source information system. Finally, an unsupervised attribute selection approach is employed to eliminate redundant attribute of the single-source information system. Theoretical analysis and comprehensive experiments on several datasets demonstrate the feasibility and superiority of our method.

Attribute selection approaches for incomplete interval-value data

Attribute selection in an information system (IS) is an important issue when dealing with a large amount of data. An IS with incomplete interval-value data is called an incomplete interval-valued information system (IIVIS). This paper proposes attribute selection approaches for an IIVIS. Firstly, the similarity degree between two information values of a given attribute in an IIVIS is proposed. Then, the tolerance relation on the object set with respect to a given attribute subset is obtained. Next, θ-reduction in an IIVIS is studied. What is more, connections between the proposed reduction and information entropy are revealed. Lastly, three reduction algorithms base on θ-discernibility matrix, θ-information entropy and θ-significance in an IIVIS are given.

An automated ontology learning for benchmarking classifier models through gain-based relative-non-redundant feature selection: a case-study with erythemato-squamous disease

Erythemato-squamous disease (ESD) is one of the complex diseases in the dermatology field, the diagnosis of which is challenging, due to common morphological features and often leads to inconsistent results. Besides, diagnosis has been done on the basis of inculcated visible symptoms pertinent with the expertise of the physician. Hence, ontology construction for prediction of erythemato-squamous disease through data mining techniques was believed to yield a clear representation of the relationships between the disease, symptoms and course of treatment. However, the classification accuracy required to be high in order to obtain a precise ontology. This required identifying the correct set of optimal features required to predict ESD. This paper proposes the Gain based Relative-Non-Redundant Attribute selection approach for diagnosis of ESD. This methodology yielded 98.1% classification accuracy with Adaboost algorithm that executed J48 as the base classifier. The feature selection approach revealed an optimal feature set comprising of 19 selected features.

A New Cygnus Optimization Algorithm for Prediction Of Cardio Vascular Disease

Machine learning is an emerging field in the present day due to a massive improvement in the size of data. However, with the current studies, the prologue of artificial intelligence and medical sciences, help in averting any such category of disease. With the intention to take good decision in health care, data mining methodology and technology plays a foremost role to transmit the enormous data into valuable information. Emerging as a primary source of fatality in the early 20th century and peaking in prevalence from 1980s, Cardio-Vascular Disease (CVD) remains a major global threat. Early prediction of CVD is most vital for sensible preclusion and treatment. This paper is to investigate the work on the attribute selection approach and propose an improved Cygnus Optimization Algorithm by carrying a random search through the whole attributes, which was obtained from UCI machine learning repository. The proposed method shows 97% of accuracy, which is better than early.

An Adaptive Threshold-Based Attribute Selection to Classify Requests Under DDoS Attack in Cloud-Based Systems

Cloud-based services are increasing day by day for various purposes due to its perpetuity and diverse dexterity. However, offensive network traffic such as distributed denial of service (DDoS) plays a significant role in threatening the cloud-based services. Therefore, defense for such attacks is required to save the cloud resources. Most of the attribute selection approaches for request classification are based on static threshold statistics. These statistics are the default values used to reduce the dimensionality of data. However, these default statistics do not work well with varying network conditions and for different intensities of DDoS attack. Therefore, an adaptive statistic is required to deal with different network and incoming traffic conditions. This paper also gives an comprehensive analysis of TCP, UDP and ICMP protocol-based DDoS attack and its effects on the cloud network. Based on the analysis and above issues, this paper presents an adaptive hybrid approach for attribute selection and classification of incoming traffic. The proposed approach consists of three subsystems such as (1) preprocessing subsystem, (2) adaptive attribute selection subsystem and (3) detection and prevention subsystem. The work utilizes NSL-KDD dataset which helps in the evaluation of the proposed approach. It is concluded that the combination of Mean Absolute Deviation technique with Random Forest classifier (MAD-RF) outperforms the other combinations. Therefore, MAD-RF is selected for further analysis. MAD-RF is also capable of dealing with TCP, UDP and ICMP protocol-based DDoS attack. The result shows that MAD-RF outperforms dimensionality reduction, traditional attribute selection methods and state-of-the-art approaches.

A New Cygnus Optimization Algorithm for Prediction Of Cardio Vascular Disease

Machine learning is an emerging field in the present day due to a massive improvement in the size of data. However, with the current studies, the prologue of artificial intelligence and medical sciences, help in averting any such category of disease. With the intention to take good decision in health care, data mining methodology and technology plays a foremost role to transmit the enormous data into valuable information. Emerging as a primary source of fatality in the early 20th century and peaking in prevalence from 1980s, Cardio-Vascular Disease (CVD) remains a major global threat. Early prediction of CVD is most vital for sensible preclusion and treatment. This paper is to investigate the work on the attribute selection approach and propose an improved Cygnus Optimization Algorithm by carrying a random search through the whole attributes, which was obtained from UCI machine learning repository. The proposed method shows 97% of accuracy, which is better than early

Quad-PolSAR data classification using modified random forest algorithms to map halophytic plants in arid areas

In this work, fully polarized SAR (PolSAR) data are exploited to characterize halophyte plants in lakeside saline wetland environments thanks to their different scattering properties. To this aim, several polarization signatures and morphological profiles (MPs) are used as inputs to the proposed “random M5 model forest” (RM5MF) and “classification via random forest regression” (CVRFR) classifiers. The experimental results show that parameters such as pedestal height (PH), as well as 3D co-polarization and cross-polarization signature plots, are more suited than biomass index (BMI), volume scattering index (VSI), and canopy scattering index (CSI) to map halophytic plants in arid environments. When we compare the suitability of PolSAR features using RM5MF, random forest (RaF), and other five popular attribute selection approaches, all the results uniformly show that span, MPs and entropy are the most valuable features, while PH and BMI are more valuable than CSI, VSI and the radar forest degradation index (RFDI). Additionally, the diagonal elements of the coherency matrix are more valuable than are the off-diagonal elements, and double-bounce, odd-bounce and wire elements are more valuable than helix bounce and volume bounce. The study results are obtained from PolSAR L-band quad-polarimetric high-sensitivity stripmap data over two study regions by comparing RM5MF and CVRFR with more traditional algorithms (support vector machine (SVM), RaF, rotation forest (RoF), and MultiBoostAB). The RM5MF model achieves the highest accuracy value in the study regions. However, due to the binary splitting criteria in the M5 model tree, it is more computationally intensive than all the others. In contrast, the CVRFR model consumes much less time—approximately 10 times less than RM5MF, and 5 times less than RoF—but still achieves better (3%–8%) classification accuracy than SVM or RoF, and its results are comparable (less than 1% difference) to those by RaF.

A Data Mining Approach for Attribute Selection in Intrusion Detection System

A Data Mining Approach for Attribute Selection in Intrusion Detection System

Attribute Selection for Partially Labeled Categorical Data By Rough Set Approach.

Attribute selection is considered as the most characteristic result in rough set theory to distinguish itself to other theories. However, existing attribute selection approaches can not handle partially labeled data. So far, few studies on attribute selection in partially labeled data have been conducted. In this paper, the concept of discernibility pair based on rough set theory is raised to construct a uniform measure for the attributes in both supervised framework and unsupervised framework. Based on discernibility pair, two kinds of semisupervised attribute selection algorithm based on rough set theory are developed to handle partially labeled categorical data. Experiments demonstrate the effectiveness of the proposed attribute selection algorithms.

Joint rough sets and Karhunen-Loève transform approach to seismic attribute selection for porosity prediction in a Chinese sandstone reservoir

The Paleogene Lower Dongying Formation of the Bohai Bay Basin, China, consists of a series of vertically stacked sand-filled delta distributary channels. These laterally complex sandstones create the need for a precise interwell estimation of reservoir porosity. In this study, we integrated wireline logs from 101 wells and [Formula: see text] of seismic data to directly predict porosity in the area of an existing heavy-oil field in the middle of the Liaoxi Uplift, Bohai Bay Basin. The top and the bottom horizons of the target oil unit interpreted on high-quality 3D seismic data are used to constrain the time window for 41 horizon-based attribute extractions. Next, we used the joint rough sets and Karhunen-Loève transform (K-L transform) selection method to choose the optimal number and the type of seismic attributes that exhibit a high correlation with porosity. Finally, a method combinings multiple linear regression and radial basis function neural network was used to predict porosity based on the selected attribute subsets. After error analysis of the 101 wells, the results from the joint attribute selection approach showed the least prediction error at well locations because the joint attribute selection captured the linear and nonlinear relationship between attributes and porosities. We obtained a geologically realistic predicted porosity distribution using the joint approach, which suggested a network of main distributary channel sands characterized by high-porosity zones (approximately 32%).

A Novel Approach Based on Rough Conditional Entropy for Attribute Reduction

Z. Pawlak’s rough set theory has been widely applied in analyzing ordinary information systems and decision tables. While few studies have been conducted on attribute selection problem in incomplete decision systems because of its complexity. Therefore, it is necessary to investigate effective algorithms to tackle this issue. In this paper, In this paper, a new rough conditional entropy based uncertainty measure is introduced to evaluate the significance of subsets of attributes in incomplete decision systems. Moreover, some important properties of rough conditional entropy are derived and three attribute selection approaches are constructed, including an exhaustive approach, a heuristic approach, and a probabilistic approach. In the end, a series of experiments on practical incomplete data sets are carried out to assess the proposed approaches. The final experimental results indicate that two of these approaches perform satisfyingly in the process of attribute selection in incomplete decision systems.

Attribute selection for product configurator design based on Gini index

Customer needs play a critical role in product design. Matching diversified customer needs with a company’s product offerings has been a challenge for academia and industry. To this end, product configurator systems have been accepted as important tools to elicit customer needs to meet the challenge. However for complex products with many product possibilities and intricate inter-product relationships, current product configuration systems may become tedious and time consuming for customers’ choice navigation process. They cannot adapt to each individual customer’s preferences by leveraging on the attribute specification information captured in previous configuration steps. This paper presents a Gini index-based attribute selection approach for configurator design. Product configuring is modelled as a sequential query–answer process. In each configuring step, the Gini index is deployed to quantify the clarity of the designer’s beliefs about the customer’s needs. The attribute which contributes most to the clarity will be selected for the customer to configure. A product recommendation module is also integrated with the configurator to further improve the efficiency. As a result, designers obtain clarification on the customer’s needs and preferences in an accelerated manner. An example is presented to test the viability of the method.

A Novel Approach of Rough Conditional Entropy-Based Attribute Selection for Incomplete Decision System

Pawlak's classical rough set theory has been applied in analyzing ordinary information systems and decision systems. However, few studies have been carried out on the attribute selection problem in incomplete decision systems because of its complexity. It is therefore necessary to investigate effective algorithms to deal with this issue. In this paper, a new rough conditional entropy-based uncertainty measure is introduced to evaluate the significance of subsets of attributes in incomplete decision systems. Furthermore, some important properties of rough conditional entropy are derived and three attribute selection approaches are constructed, including an exhaustive search strategy approach, a heuristic search strategy approach, and a probabilistic search strategy approach for incomplete decision systems. Moreover, several experiments on real-life incomplete data sets are conducted to assess the efficiency of the proposed approaches. The final experimental results indicate that two of these approaches can give satisfying performances in the process of attribute selection in incomplete decision systems.