Hybrid Data Mining Techniques Research Articles

Monthly streamflow forecasting based on meteorological data from a nearby station

ABSTRACT Monthly streamflow forecasting is critical for improving water resource management. In this study, several base-classifier data-mining algorithms – conjunctive rule (CR), isotonic regression (ISOR), sequential minimal optimization regression (SMOR) – as well as several hybrid data-mining techniques – disjoint aggregating or dagging (DA)-CR, DA-ISOR, and DA-SMOR – that combine dagging with these algorithms were developed and applied to forecasting streamflow 3 and 12 months into the future (i.e., Qt+3 and Qt+12) based on meteorological data from a nearby station. Thirty years of data (from 1988 to 2018) that included precipitation, minimum relative humidity, maximum relative humidity, evaporation, hours of sunshine, maximum temperature, minimum temperature, wind speed, and streamflow were collected at the Kermanshah synoptic station were input to develop and evaluate several models. Varying combinations of the input data were tested to find the optimal set to employ. The models were validated and compared using several quantitative statistical indices. Though all satisfactorily predicted monthly streamflow, the hybrid models (DA-CR, DA-ISOR, and DA-SMOR) outperformed the base-classifier models (CR, ISOR, and SMOR), proving that dagging improved data-mining models significantly. Of the hybrid models, D-SMOR was the best. The models developed in this study are cost-effective tools for quick and accurate monthly streamflow forecasting.

Overview of Logistics Demand Forecasting Methods

Accurate forecasting of logistics demand is of great theoretical significance and practical application value for the formulation of national policies and the satisfaction of actual demand in the logistics industry. From the modeling form, the existing logistics demand forecasting methods are divided into four categories: single traditional forecasting method, single intelligent forecasting method, combined forecasting method and mixed forecasting method. Among them, single traditional forecasting methods mainly include simple time series method, regression analysis, mathematical and statistical methods, etc.; single intelligent forecasting methods mainly involve gray forecasting method, neural network, support vector machine and their improved forms; combined forecasting methods are mainly summarized into three combined forms: linear combination of single forecasting results, nonlinear combination of single forecasting results, modified single forecasting results; mixed forecasting methods are mainly summarized into three hybrid forms: hybrid intelligent optimization algorithms with single prediction methods, hybrid data dimensionality reduction techniques with intelligent prediction methods, and hybrid data mining techniques with intelligent prediction methods. The four major types of forecasting methods are reviewed, and each forecasting model in the four major types of methods is evaluated in terms of modeling principles, advantages and disadvantages, and applicability, in order to find forecasting methods suitable for different logistics demand forecasting tasks for logistics demand researchers.

Applying Data Mining in Surveillance

In the current times where human safety is threatened by man-made and natural calamities, surveillance systems have gained immense importance. But, even in presence of high definition (HD) security cameras and manpower to monitor the live feed 24/7, room for missing important information due to human error exists. In addition to that, employing an adequate number of people for the job is not always feasible either. The solution lies in a system that allows automated surveillance through classification and other data mining techniques that can be used for extraction of useful information out of these inputs. In this research, a data mining-based framework has been proposed for surveillance. The research includes interpretation of data from different networks using hybrid data mining technique. In order to show the validity of the proposed hybrid data mining technique, an online data set containing network of a suspicious group has been utilized and main leaders of network has been identified.

Improvement of HCF life of automotive safety components considering a novel design of wheel alignment based on a Hybrid multibody dynamic, finite element, and data mining techniques

The main aim of the present study is to improve the High-Cycle Fatigue (HCF) life of automotive safety components subjected to multi-input random non-proportional loading. A case study was conducted on the automotive knuckle as one of the high-critical mechanical components of the suspension and steering systems. The steering knuckle is subjected to the highest road- and steering-induced random loads that are exerted on several points (i.e., joints of knuckle with lower control arm, steering linkage, and Macpherson strut). Moreover, incorrect adjustment of wheel angles increases the intensity of road loads. In this regard, previous studies have shown that wheel angles affect the service time of this component. In the present study, the authors attempted to present a novel design of wheel alignment to reduce the intensity of 3D stress components, and subsequently, to increase the fatigue life of the steering knuckle. To this end, a Hybrid multibody dynamic, finite element, and data mining techniques was used. Accordingly, different load histories extracted through Multi-Body Dynamics (MBD) analysis of a full vehicle model were applied on various joints of this component. Next, a transient dynamic analysis was performed, and the time histories of 3D stress fields were obtained in the critical zone. The fatigue life of the component was predicted using the Critical Plane Method (CPM) via employing the Rain-flow cycle counting technique and Miner-Palmgren damage accumulation rule. Finally, the results of Taguchi Sensitivity Analysis (TSA) were used to investigate the effect of wheel primary angles, including Camber and Toe, on the fatigue life of the cast iron knuckle. Moreover, Response Surface Method (RSM) was utilized to optimize the wheel alignment, and the cyclic lifetime of the component was compared under the pre- and post-optimization process. The results reveal that knuckle life improves by 33.66% using the optimum settings of wheel alignment.

Optimization of association rules using hybrid data mining technique

Software quality has been the important area of interest for decades now in the IT sector and software firms. Defect prediction gives the tester the pointers as to where the bugs will most likely be hidden in the software product. Identifying and reporting the defect probe areas is the main job of software defect prediction techniques. Early detection of software defects during Software Development Life Cycle could lead to a reduction in cost of development, time involved in further testing activities and rework effort post-production and maintenance phase, thus resulting in more reliable software. Software metrics can be used for developing the defect prediction models. Several data mining techniques can be applied on the available open-source software datasets. These datasets are extracted from software programs. Such datasets made publicly available by National Aeronautics and Space Administration for their various softwares have been extensively used in software engineering-related research activities. These datasets contain information on associated Software Metrics at module level. The proposed idea is a novel hybrid data mining technique consisting of Clustering and Modified Apriori Algorithm that results in improved efficiency and reliability of Software Defect Prediction. This technique works by reducing the number of association rules generated. The results are achieved by using interestingness measure called spread. The paper also does a comparative analysis of the results obtained from the novel technique with the existing hybrid technique of Clustering and Apriori.

Bedload transport rate prediction: Application of novel hybrid data mining techniques

The accurate prediction of bedload transport in gravel-bed rivers remains a significant challenge in river science. However the potential for data mining algorithms to provide models of bedload transport have yet to be explored. This study provides the first quantification of the predictive power of a range of standalone and hybrid data mining models. Using bedload transport data collected in laboratory flume experiments, the performance of four types of recently developed standalone data mining techniques - the M5P, random tree (RT), random forest (RF) and the reduced error pruning tree (REPT) - are assessed, along with four types of hybrid algorithms trained with a Bagging (BA) data mining algorithm (BA-M5P, BA-RF, BA-RT and BA-REPT). The main findings are four-fold. First, the BA-M5P model had the highest prediction power (R2 = 0.943; RMSE = 0.061 kg m−1 s−1; MAE = 0.040 kg m−1 s−1; NSE = 0.945; PBIAS = −1.60) followed by M5P, BA-RT, RT, BA-RF, RF, BA-REPT, and REPT. All models displayed ‘very good’ performance except the BA-REPT and REPT model, which were ‘satisfactory’. Second, the M5P, BA-RT, and RT models underestimated, and the BA-M5P, BA-RF, RF, BA-REPT and REPT models overestimated, bedload transport rates. Third, flow velocity had the most significant impact on bedload transport rate (PCC = 0.760) followed by shear stress (PCC = 0.709), discharge (PCC = 0.668), bed shear velocity (PCC = 0.663), bed slope (PCC = 0.490), flow depth (PCC = 0.303), median sediment diameter (PCC = 0.247), and relative roughness (PCC = 0.003). Fourth, the maximum depth of tree was the most sensitive operator in decision tree-based algorithms, and batch size, number of execution slots and number of decimal places did not have any impact on model’ prediction power. Overall the results revealed that hybrid data mining techniques provide more accurate predictions of bedload transport rate than standalone data mining models. In particular, M5P models, trained with a Bagging data mining algorithm, have great potential to produce robust predictions of bedload transport in gravel-bed rivers.

An e-healthcare system for disease prediction using hybrid data mining technique

Purpose The purpose of this study is to alleviate the specified issues to a great extent. To promote patients’ health via early prediction of diseases, knowledge extraction using data mining approaches shows an integral part of e-health system. However, medical databases are highly imbalanced, voluminous, conflicting and complex in nature, and these can lead to erroneous diagnosis of diseases (i.e. detecting class-values of diseases). In literature, numerous standard disease decision support system (DDSS) have been proposed, but most of them are disease specific. Also, they usually suffer from several drawbacks like lack of understandability, incapability of operating rare cases, inefficiency in making quick and correct decision, etc. Design/methodology/approach Addressing the limitations of the existing systems, the present research introduces a two-step framework for designing a DDSS, in which the first step (data-level optimization) deals in identifying an optimal data-partition (Popt) for each disease data set and then the best training set for Popt in parallel manner. On the other hand, the second step explores a generic predictive model (integrating C4.5 and PRISM learners) over the discovered information for effective diagnosis of disease. The designed model is a generic one (i.e. not disease specific). Findings The empirical results (in terms of top three measures, namely, accuracy, true positive rate and false positive rate) obtained over 14 benchmark medical data sets (collected from https://archive.ics.uci.edu/ml) demonstrate that the hybrid model outperforms the base learners in almost all cases for initial diagnosis of the diseases. After all, the proposed DDSS may work as an e-doctor to detect diseases. Originality/value The model designed in this study is original, and the necessary parallelized methods are implemented in C on Cluster HPC machine (FUJITSU) with total 256 cores (under one Master node).

Spatial prediction of groundwater potentiality using ANFIS ensembled with teaching-learning-based and biogeography-based optimization

Groundwater is one of the most beneficial natural resources worldwide. The main aim of the present study is the mapping of groundwater potential of the Zhangjiamao area in China using two novel hybrid data mining techniques that involve the adaptive neuro-fuzzy inference system (ANFIS) ensembled with teaching-learning-based optimization (TLBO) and biogeography-based optimization (BBO). First, 93 spring locations were identified in the study area, which were randomly divided in the ratio of 70:30 to be used for building models and validation, respectively. A total of 16 spring affecting factors viz., slope aspect, altitude, slope angle, plan curvature, profile curvature, curvature, sediment transport index, stream power index, topographic wetness index, distance to roads, distance to rivers, rainfall, lithology, soil, NDVI, and land use/land cover, were used as input variables. The probability certainty factor (PCF) method was applied for the correlation analysis between spring occurrences and the affecting factors. Subsequently, the ANFIS-TLBO and ANFIS-BBO models were generated to produce groundwater spring potential maps (GSPMs). Finally, the GSPMs were evaluated by using the area under the receiver operating characteristic (AUROC). The results showed that the AUROC values for the ANFIS-TLBO model based on the training and validation datasets were 0.866 and 0.905, respectively. In contrast, the AUROC values for the ANFIS-BBO model based on the training and validation datasets were 0.861 and 0.887, respectively. The results of ANFIS-TLBO showed that 25.66% of the study area could be classified into the high and very high groundwater spring potential classes as compared to the ANFIS-BBO model (30.52%). The results of the present study can be useful for groundwater management.

Detection and classification of islanding and power quality disturbances in microgrid using hybrid signal processing and data mining techniques

This study presents multi-scale morphological gradient filter (MSMGF) and short-time modified Hilbert transform (STMHT) techniques, respectively, to detect and classify multiclass power system disturbances in a distributed generation (DG)-based microgrid environment. The non-stationary power signal samples measured near the target DG's are processed through the proposed MSMGF and STMHT techniques, respectively, and some computations over them generates the target parameter sets. Depending on the complexity of the overlapping in the target attribute values for different disturbance patterns, fuzzy judgment tree structure is incorporated for multiclass event classification, which proves to be robust for most of the classes. In this regard, an extensive simulation on the proposed microgrid models, subjected to a number of multiclass disturbances has been performed in MATLAB/Simulink environment. The faster execution, lower computational burden, superior efficiency as well as better accuracy in multiclass power system disturbance classification by the proposed judgment tree-based MSMGF and STMHT techniques, respectively, as compared to some of the conventional techniques, is significantly illustrated in the performance evaluation section. Further, as illustrated in this section, the real-time capability of the proposed techniques has been verified in the hardware environment, where the results shown are satisfactory.

Developing constitutive models from EPR‐based self‐learning finite element analysis

SummaryA constitutive model that captures the material behavior under a wide range of loading conditions is essential for simulating complex boundary value problems. In recent years, some attempts have been made to develop constitutive models for finite element analysis using self‐learning simulation (SelfSim). Self‐learning simulation is an inverse analysis technique that extracts material behavior from some boundary measurements (eg, load and displacement). In the heart of the self‐learning framework is a neural network which is used to train and develop a constitutive model that represents the material behavior. It is generally known that neural networks suffer from a number of drawbacks. This paper utilizes evolutionary polynomial regression (EPR) in the framework of SelfSim within an automation process which is coded in Matlab environment. EPR is a hybrid data mining technique that uses a combination of a genetic algorithm and the least square method to search for mathematical equations to represent the behavior of a system. Two strategies of material modeling have been considered in the SelfSim‐based finite element analysis. These include a total stress‐strain strategy applied to analysis of a truss structure using synthetic measurement data and an incremental stress‐strain strategy applied to simulation of triaxial tests using experimental data. The results show that effective and accurate constitutive models can be developed from the proposed EPR‐based self‐learning finite element method. The EPR‐based self‐learning FEM can provide accurate predictions to engineering problems. The main advantages of using EPR over neural network are highlighted.

Enhancement of Effective Spatial Data Analysis using R

Background: The availability of Spatial Data which is a part of GIS is growing day by day in an exponential manner. This high availability of data is throwing challenges to the research community to analyze and draw effective conclusions. The Present Study aims at requirement for effective analysis and to draw Conclusions. Methods/Statistical Analysis: Spatial Analysis requires logical relationships between attribute data and map features.Spatial data Analysis is not a simple single task it requires complex procedures in which combinational techniques namely Hybrid techniques are required for effective analysis. Mathematics and statistics are the fundamentals to spatial data analytics. In this paper, a realistic Spatial crime data set was considered for analysis. It involves different types of data mining Techniques like Clustering, Classification and Association rule mining techniques apart from Hybrid techniques. These hybrid Data mining Techniques were applied using R. Findings: The Hybrid Data Mining techniques with K-means Clustering and J48 Decision Tree Algorithm was developed and Applied for the enhancement of accuracy. Association Rule generation Apriori algorithm was applied on the resultant K-means clustered data set. The application of 3D visualization techniques also made for further analysis. Applications/Improvements: It is essentially required to analyze these complex spatial data sets effectively. So there is a need of hybrid Spatial Data Mining Techniques requirement for effective analysis and to draw Conclusions.

A Comparative Study of Utilization of Single and Hybrid Data Mining Techniques in Heart Disease Diagnosis and Treatment Plan

In clinical medicine ,data mining deals with learning models to predict patient’s health. The models can be dedicated to support clinicians in diagnostic and monitoring tasks. Data mining methods are commonly applied in clinical contexts to analyze retrospective data, thus giving healthcare professionals the opportunity to exploit enormous amounts of data routinely collected during their day-by-day activity. Nowadays, clinicians can take advantage of data mining techniques to deal with the huge amount of research results obtained by molecular medicine such as genomic signatures or genetic which may allow transition from population-based to personalized medicine.The different classification and prediction models can be devoted to support medical practitioners in diagnosis and formation of treatment plans. There is need of powerful data analysis tool to extract useful knowledge from huge amount of data available in health care field. Last few years, heart disease is the major cause of death all over the world. In heart disease diagnosis and treatment, single data mining techniques are showing satisfactory level of accuracy. Nowadays, researchers are experimenting the deployment of hybrid data mining techniques showing great level of accuracy. In this paper,single data mining techniques like Naive base,Decision tree, Association rule, Neural network and Regression are studied and compared with hybrid data mining algorithm to achieve an efficient results in heart disease diagnosis and to formulate treatment plan.

Knowledge-based system, equipped with cluster analysis for eco-material selection: an automobile structure case study

The aim of this study is to develop a material selection framework structured around a knowledge-based system (KBS). Specifically, a hybrid data mining technique is employed to extract knowledge from large datasets using cluster analysis techniques; the mined knowledge then serves as the inference logic within the KBS designed for material selection purposes. Cluster analysis results are used as a basis for the tree-based structure of the KBS where if–then rules are developed based on the general cluster properties; that is, inference logic is structured in a way such that it can predict general sustainability characteristics of the material as well as its exact mechanical, cost and physical properties. To develop the structure of the KBS, the selection structure employs sustainable material indices. Additionally, the proposed material selection model of the KBS is purposefully composed of material sustainability, functionality and cost indices. The constructed knowledge is then demonstrated for selecting automobile structural panels.

HYBRID OF FUZZY CLUSTERING NEURAL NETWORK OVER NSL DATASET FOR INTRUSION DETECTION SYSTEM

Intrusion Detection System (IDS) is one of the component that take part in the system defence, to identify abnormal activities happening in the computer system. Nowadays, IDS facing composite demands to defeat modern attack activities from damaging the computer systems. Anomaly-Based IDS examines ongoing traffic, activity, transactions and behavior in order to identify intrusions by detecting anomalies. These technique identifies activities which degenerates from the normal behaviours. In recent years, data mining approach for intrusion detection have been advised and used. The approach such as Genetic Algorithms , Support Vector Machines, Neural Networks as well as clustering has resulted in high accuracy and good detection rates but with moderate false alarm on novel attacks. Many researchers also have proposed hybrid data mining techniques. The previous resechers has intoduced the combination of Fuzzy Clustering and Artificial Neural Network. However, it was tested only on randomn selection of KDDCup 1999 dataset. In this study the framework experiment introduced, has been used over the NSL dataset to test the stability and reliability of the technique. The result of precision, recall and f-value rate is compared with previous experiment. Both dataset covers four types of main attacks, which are Derial of Services (DoS), User to Root (U2R), Remote to Local (R2L) and Probe. Results had guarenteed that the hybrid approach performed better detection especially for low frequent over NSL datataset compared to original KDD dataset, due to the removal of redundancy and uncomplete elements in the original dataset. This electronic document is a &#34live&#34 template. The various components of your paper [title, text, tables, figures and references] are already defined on the style sheet, as illustrated by the portions given in this document.

EPR-based material modelling of soils considering volume changes

In this paper an approach is presented for developing material models for soils based on evolutionary polynomial regression (EPR), taking into account its volumetric behaviour. EPR is a recently developed hybrid data mining technique that searches for structured mathematical equations (representing the behaviour of a system) using genetic algorithm and the least squares method. Stress–strain data from triaxial test are used to train and develop EPR-based material models for soil. The developed models are compared with some of the well known conventional material models. In particular, the capability of the developed EPR models in predicting volume change behaviour of soils is illustrated. It is also shown that the developed EPR-based material models can be incorporated in finite element (FE) analysis. Two geotechnical examples are presented to verify the developed EPR-based FE model (EPR-FEM). The results of the EPR-FEM are compared with those of a standard FEM where conventional constitutive models are used to describe the material behaviour. The results show that EPR-FEM can be successfully employed to analyse geotechnical engineering problems. The advantages of the proposed EPR models are highlighted.

Credit risk Evaluation by hybrid data mining technique

Most studies have concentrated on building an accurate credit scoring model to decide whether or not to grant credit to new applicants and the efforts to build more accurate credit scoring model seems to be not significant. In this paper, we proposed a hybrid data mining technique which contains two processing stages. In the clustering stage, the samples of the accepted and new applicants are grouped into homogeneous clusters, the isolated samples are deleted and inconsistent samples are relabeled. In the classification stage, support vector machines used samples with new labels to build the scoring model. The difference from the other credit scoring model is that the samples were classified into three or four classes, rather than two the good and the bad credit classes. Experimental results based on the credit data set provided by a local bank in China showed that by choosing a proper cut-off point, super classification accuracy of the good and the bad credit is obtained. Risk management strategies are developed according to the characteristic of each class.

Facilitating cross-selling in a mobile telecom market to develop customer classification model based on hybrid data mining techniques

As the competition between mobile telecom operators becomes severe, it becomes critical for operators to diversify their business areas. Especially, the mobile operators are turning from traditional voice communication to mobile value-added services (VAS), which are new services to generate more average revenue per user (ARPU). That is, cross-selling is critical for mobile telecom operators to expand their revenues and profits. In this study, we propose a customer classification model, which may be used for facilitating cross-selling in a mobile telecom market. Our model uses the cumulated data on the existing customers including their demographic data and the patterns for using old products or services to find new products and services with high sales potential. The various data mining techniques are applied to our proposed model in two steps. In the first step, several classification techniques such as logistic regression, artificial neural networks, and decision trees are applied independently to predict the purchase of new products, and each model produces the results of their prediction as a form of probabilities. In the second step, our model compromises all these probabilities by using genetic algorithm (GA), and makes the final decision for a target customer whether he or she would purchase a new product. To validate the usefulness of our model, we applied it to a real-world mobile telecom company’s case in Korea. As a result, we found that our model produced high-quality information for cross-selling, and that GA in the second step contributed to significantly improve the performance.

Customer churn prediction by hybrid neural networks

As churn management is a major task for companies to retain valuable customers, the ability to predict customer churn is necessary. In literature, neural networks have shown their applicability to churn prediction. On the other hand, hybrid data mining techniques by combining two or more techniques have been proved to provide better performances than many single techniques over a number of different domain problems. This paper considers two hybrid models by combining two different neural network techniques for churn prediction, which are back-propagation artificial neural networks (ANN) and self-organizing maps (SOM). The hybrid models are ANN combined with ANN (ANN + ANN) and SOM combined with ANN (SOM + ANN). In particular, the first technique of the two hybrid models performs the data reduction task by filtering out unrepresentative training data. Then, the outputs as representative data are used to create the prediction model based on the second technique. To evaluate the performance of these models, three different kinds of testing sets are considered. They are the general testing set and two fuzzy testing sets based on the filtered out data by the first technique of the two hybrid models, i.e. ANN and SOM, respectively. The experimental results show that the two hybrid models outperform the single neural network baseline model in terms of prediction accuracy and Types I and II errors over the three kinds of testing sets. In addition, the ANN + ANN hybrid model significantly performs better than the SOM + ANN hybrid model and the ANN baseline model.

Applying hybrid data mining techniques to web-based self-assessment system of Study and Learning Strategies Inventory

Traditional assessment tools, such as “Learning and Study Strategy Scale Inventory (LASSI)”, are typically pen-and-paper tests that require responses to a multitude of questions. This may easily lead to student’s resistance, fatigue and unwillingness to complete the assessment. To improve the situation, a hybrid data mining technique was applied to analyze the LASSI surveys of freshmen students at Tamkang University. The most significant contribution of this research is in dynamically reducing the number of questions while the LASSI assessment is proceeding. To verify the appliance of the proposed method, a web-based LASSI self-assessment system (Web-LSA) was developed. This system can be used as a guide to determine study disturbances for high-risk groups, and can provide counselors with fundamental information on which to base follow-up counseling services to its users.

Two-step filtering datamining method integrating case-based reasoning and rule induction

Case-based reasoning (CBR) methods are applied to various target problems on the supposition that previous cases are sufficiently similar to current target problems, and the results of previous similar cases support the same result consistently. However, these assumptions are not applicable for some target cases. There are some target cases that have no sufficiently similar cases, or if they have, the results of these previous cases are inconsistent. That is, the appropriateness of CBR is different for each target case, even though they are problems in the same domain. Thus, applying CBR to whole datasets in a domain is not reasonable. This paper presents a new hybrid datamining technique called two-step filtering CBR and rule induction (TSFCR), which dynamically selects either CBR or RI for each target case, taking into consideration similarities and consistencies of previous cases. We apply this method to three medical diagnosis datasets and one credit analysis dataset in order to demonstrate that TSFCR outperforms the genuine CBR and RI.