Fuzzy Decision Tree Research Articles

Broad area mapping of monthly soil erosion risk using fuzzy decision tree approach: integration of multi-source data within GIS

Soil erosion poses a serious problem for sustainable agriculture and the environment. There is a need to develop a simple and practical approach for broad area mapping of soil erosion risk that uses the uncertain but available information as input data within the constraints of reasonable cost and time. In this work, a predictive approach for conducting analytical erosion risk assessment across broad areas is developed, which combines a fuzzy decision tree (FDT), remote sensing and Geographic Information System (GIS). This approach is applicable to situations with a limited amount of input data and can easily adjust assessment factors according to actual need. In this study, four dominating factors affecting soil erosion were considered: soil, topography, land cover and climate. GIS thematic layers of these factors were constructed followed by fuzzified analysis through trapezoidal shaped membership functions. Based on subdivided erosion response units (ERUs), an optimal FDT was determined to classify monthly soil erosion risk into five levels. High-risk and very high-risk soil erosion in the study area is mainly concentrated from June to August, with July and August showing the highest risk covering more than 80% of the study area. November to March is dominated by low risk over more than 90% of the study area, while medium risk is dominant in April, May, September and October. Compared with field survey data, the fuzzy decision erosion risk assessment approach was shown to be applicable and economical for rapidly identifying and locating soil erosion risk with limited input data by means of remote sensing and GIS.

Feature subset selection Filter–Wrapper based on low quality data

Today, feature selection is an active research in machine learning. The main idea of feature selection is to choose a subset of available features, by eliminating features with little or no predictive information, as well as redundant features that are strongly correlated. There are a lot of approaches for feature selection, but most of them can only work with crisp data. Until now there have not been many different approaches which can directly work with both crisp and low quality (imprecise and uncertain) data. That is why, we propose a new method of feature selection which can handle both crisp and low quality data. The proposed approach is based on a Fuzzy Random Forest and it integrates filter and wrapper methods into a sequential search procedure with improved classification accuracy of the features selected. This approach consists of the following main steps: (1) scaling and discretization process of the feature set; and feature pre-selection using the discretization process (filter); (2) ranking process of the feature pre-selection using the Fuzzy Decision Trees of a Fuzzy Random Forest ensemble; and (3) wrapper feature selection using a Fuzzy Random Forest ensemble based on cross-validation. The efficiency and effectiveness of this approach is proved through several experiments using both high dimensional and low quality datasets. The approach shows a good performance (not only classification accuracy, but also with respect to the number of features selected) and good behavior both with high dimensional datasets (microarray datasets) and with low quality datasets.

An interpretable classifier for detection of cardiac arrhythmias by using the fuzzy decision tree

An extraction of medical knowledge from cardiological data is proposed in this work, it is based on relevant intelligent method called fuzzy decision tree. It could lead to increase understanding the cause of various abnormal beats in cardiac activity, leading to a better medical diagnosis. The performance of this technique is evaluated on the MIT-BIH Arrhythmia Database following the AAMI recommendations. The first part of this paper discusses the characterization of heart beats. It is considered as an important step in arrhythmias classification. In a second part we apply the fuzzy decision tree to recognize some cardiac abnormalities. In the last part we discuss the activity of fuzzy decision rules extracted from cardiological data analyzing.

Fuzzy Decision Tree Approach for Optimal Supplier Base

Minimizing supplier base allows better supplier relationship management, but any supplier failure will interrupt the supplies chain. Earth quake in Fukushima, Japan and flood in Bangkok, Thailand, caused supply chain breakdown and affect the global production. Decision tree method is recommended to determine optimal number of supplier in supplier base with the risk of supplier failure. This paper presented an approach to integrate fuzzy logic in decision tree approach by rating the supplier risk in linguistic terms for imperfect environment. Numerical example and sensitivity are carried out and presented, and optimal supplier base was identified.

Analysis of Textual Data Based on Inductive Learning Techniques

This paper introduces knowledge discovery methods based on inductive learning techniques from textual data. The author argues three methods extracting features of the textual data. First one activates a key concept dictionary, second one does a key phrase pattern dictionary, and third one does a named entity extractor. These features are used in order to generate rules representing relationships between the features and text classes. The rules are described in the format of a fuzzy decision tree. Also, these features are used in order to acquire a classification model based on SVM (Support Vector Machine). The model can classify new textual data into the text classes with high classification accuracy. Lastly, this paper introduces two application tasks based on these methods and verifies the effect of the methods.

Estimating phosphorus delivery from land to water in headwater catchments using a fuzzy decision tree approach

AbstractThe pathway that delivers mobilized phosphorus (P) from source to surface water is conceptually complex. Firstly, P mobilization itself, either through mechanical disturbance of the soil or through chemical dissolution or weathering, is difficult to measure in situ; secondly, the annual flow‐weighted mean concentration of P is a difficult metric to gather. Summarizing these pathway processes as the ratio of the delivered P to the mobilized fraction of P assists in the quantification and assessment of the P transfer continuum. The proportion of this mobilized P that is subsequently delivered to the watercourse is also difficult to quantify. A range of P models already exist that aim to predict P mobilization and delivery. Reasonable estimates can be made for the outputs of P from a plot or small catchment over a series of events. Current models mainly rely on empirical data or expert judgment to derive appropriate coefficients of P transfer and delivery. This paper describes the research designed to evaluate the range of delivery coefficients from headwater catchments that might be anticipated for different land‐use regions across the UK. These coefficients were evaluated within a fuzzy decision tree framework. The delivery coefficients used in this study covered a broad range of possible values from all available data describing P fluxes from headwater catchments in the UK. The results of this work indicate that delivery coefficients are a useful way of summarizing the mobilization behaviour of a headwater catchment. Mean delivery coefficients above unity were present in four of 18 catchments, indicating higher than average delivery, probably due to intensive agricultural management practices. This first use of a fuzzy or uncertain approach to model P delivery in the UK has generated potentially promising results, and a methodology has been developed for producing delivery coefficients using sparse data, which may be applicable to other headwater catchments.

Interpretability-based fuzzy decision tree classifier a hybrid of the subtractive clustering and the multi-objective evolutionary algorithm

The fuzzy decision tree (FDT) is a powerful, top---down, hierarchical search scheme to extract human interpretable classification rules. Furthermore, the FDT is considered an approach to model a system by making use of a descriptive language with fuzzy logic and fuzzy predicates. There exist two contradictory requirements with fuzzy modeling, namely, accuracy and interpretability. In this study, the subtractive clustering and a multi-objective evolutionary algorithm are used to develop a novel fuzzy modeling scheme based on the FDT classifier to construct an accurate and interpretable system that is defined as the interpretability-based fuzzy decision tree classifier. Two interpretability measures, namely, complexity interpretability and semantics interpretability are considered to reach acceptable accuracy. These measures are optimized as different objectives within the multi-objective framework. Results obtained in several benchmark classification problems are encouraging because they show the ability of the developed scheme while yielding accuracy comparable to that achieved by other methods like neural networks and the crisp decision tree (C4.5). The experimental results demonstrate the superiority of the developed scheme.

Analysis of Complex Decisional Situations in Companies with the Support of AHP Extension of Vroom-Yetton Contingency Model

Decision-making processes are crucial for individuals, groups, and organisations to gain and sustain the competitive advantage. The choice of the most appropriate decision style remains the very complex issue. Weaknesses of the original Vroom-Yetton contingency model, referring to the validity and reliability of the model were addressed by fuzzy decision trees or expert systems utilisation. In Analytic Hierarchy Process approach to decision making, priorities of alternatives are derived objectively after subjective judgments are made. In practical part, three individual decision making personalities on four different standardised situational cases are exemplified to compare recommendations provided by the contingency tree with subjective preferences of decision makers in the proposed model.

Research on Method of Ecomaterials Life Cycle Assessment

It is generally considered, as the material basis of human survival and production, materials industry is one of the main reasons causing energy shortages, excessive consumption of resources and environmental pollution, the issues on ecomaterials life cycle assessment are also increasingly subject to attention. In this paper, Fuzzy decision trees is applied due to they are powerful, top-down, hierarchical search methodology to extract human interpretable classification rules. And Fuzzy ID3 algorithm is used, a popular and particularly efficient method, to construct fuzzy decision tree. We use weighted fuzzy production rules generated from FDT to establish assessment model. In addition, a new method is proposed, based on fuzzy rules and degree of confidence, to set the band of assessment. Classification accuracy was used to examine results of the system. The results indicated that the system is very valid for eco-materials life cycle assessment.

An application model of fuzzy clustering analysis and decision tree algorithms in building web mining

An application model of fuzzy clustering analysis and decision tree algorithms in building web mining

Extraction of fuzzy rules from fuzzy decision trees: An axiomatic fuzzy sets (AFS) approach

In this study, we introduce a new type of coherence membership function to describe fuzzy concepts, which builds upon the theoretical findings of the Axiomatic Fuzzy Set (AFS) theory. This type of membership function embraces both the factor of fuzziness (by capturing subjective imprecision) and randomness (by referring to the objective uncertainty) and treats both of them in a consistent manner. Furthermore we propose a method to construct a fuzzy rule-based classifier using coherence membership functions. Given the theoretical developments presented there, the resulting classification systems are referred to as AFS classifiers. The proposed algorithm consists of three major steps: (a) generating fuzzy decision trees by assuming some level of specificity (detailed view) quantified in terms of threshold; (b) pruning the obtained rule-base; and (c) determining the optimal threshold resulting in a final tree. Compared with other fuzzy classifiers, the AFS classifier exhibits several essential advantages being of practical relevance. In particular, the relevance of classification results is quantified by associated confidence levels. Furthermore the proposed algorithm can be applied to data sets with mixed data type attributes. We have experimented with various data commonly present in the literature and compared the results with that of SVM, KNN, C4.5, Fuzzy Decision Trees (FDTs), Fuzzy SLIQ Decision Tree (FS-DT), FARC-HD and FURIA. It has been shown that the accuracy is higher than that being obtained by other methods. The results of statistical tests supporting comparative analysis show that the proposed algorithm performs significantly better than FDTs, FS-DT, KNN and C4.5.

Justification for the selection of manufacturing technologies: a fuzzy-decision-tree-based approach

In this paper, a developed model for the justification of alternative manufacturing technologies is presented. The approach, based on fuzzy decision trees, provides a methodology capable of identifying patterns within a technology case repository to support the evaluation of manufacturing systems. Experts are highly influential individuals in the decision process; they provide support and guidance when selecting investments. The experience-oriented task is founded on previous cases or an experts’ experience, and therefore difficult to express in a rational form. The concept is based on a number of characteristics of the case-based reasoning, rule induction and expert system theory. Structured around the fuzzy-decision-tree data-mining technique, the framework provides the ability of using regulated case information to act as structured experience for assisting in the decision process. Fuzzy induction extracts formal rules from a set of experience data, and the expert system philosophy computes the experience base of human expertise for problem-solving. A test case indicates the stability of the classification algorithm and verifies the applicability within the domain.

Enhanced Encoding with Improved Fuzzy Decision Tree Testing Using CASP Templates

A novel protein model assessment technique using an improved fuzzy decision tree was tested using CASP8 and CASP9 templates. The testing was conducted in three phases. In the first two phases, templates were classified without explicit scoring. In phase three, a new scoring method was created for performance evaluation. The new protein model assessment technique was compared with several common prominent model assessment techniques for CASP competitions. The performance was analyzed based on correlation between our scores and GDT_TS scores of the templates. Finally, it is concluded that although the novel protein model assessment technique resulted in reduced correlation when compared to best competitors in CASP, its uniqueness in using the improved fuzzy decision tree and a single model stands as a new paradigm in the protein model assessment field.

Feature Selection and Fuzzy Decision Tree for Network Intrusion Detection

Extra features can increase computation time, and can impact the accuracy of the Intrusion Detection System. Feature selection improves classification by searching for the subset of features, which best classify the training data. This paper proposed approach uses Mutual Correlation for feature selection which reduces from 34 continuous attributes to 10 continuous attributes and Fuzzy Decision Tree for detection and diagnosis of attacks. Experimental results on the 10% KDD Cup 99 benchmark network intrusion detection dataset demonstrate that the proposed learning algorithm achieved high true positive rate (TPR) and significant reduce false positive rate (FP ). DOI: http://dx.doi.org/10.11591/ij-ict.v1i2.591

INDUCING FUZZY REGRESSION TREE FORESTS USING ARTIFICIAL IMMUNE SYSTEMS

Fuzzy decision forests aim to improve the predictive power of single fuzzy decision trees by allowing multiple views of the same domain to be modelled. Such forests have been successfully created for classification problems where the outcome field is discrete; however predicting a continuous output value is more challenging in combining the output from multiple fuzzy decision trees. This paper presents a new approach to creating fuzzy regression tree forests based upon the induction of multiple fuzzy regression decision trees from one training sample, where each tree will represent a different view of the data domain. The singular fuzzy regression trees are induced using a proven algorithm known as Elgasir which fuzzifies crisp CHAID decision trees using trapezoidal membership functions for fuzzification and applies Takagi-Sugeno inference to obtain the final predicted values. A modified version of Artificial Immune System Network model (opt-aiNet) is then used for the simultaneous optimization of the membership functions across all trees within the forest. A strength of the proposed method is that data does not require fuzzification before forest induction this reducing pre-processing time and the need for subjective human experts. Five problem sets from the UCI repository and KEEL repository are used to evaluate the approach. The experimental results have shown that fuzzy regression tree forests reduce the error rate compared with single fuzzy regression trees.

IIVFDT: IGNORANCE FUNCTIONS BASED INTERVAL-VALUED FUZZY DECISION TREE WITH GENETIC TUNING

The choice of membership functions plays an essential role in the success of fuzzy systems. This is a complex problem due to the possible lack of knowledge when assigning punctual values as membership degrees. To face this handicap, we propose a methodology called Ignorance functions based Interval-Valued Fuzzy Decision Tree with genetic tuning, IIVFDT for short, which allows to improve the performance of fuzzy decision trees by taking into account the ignorance degree. This ignorance degree is the result of a weak ignorance function applied to the punctual value set as membership degree. Our IIVFDT proposal is composed of four steps: (1) the base fuzzy decision tree is generated using the fuzzy ID3 algorithm; (2) the linguistic labels are modeled with Interval-Valued Fuzzy Sets. To do so, a new parametrized construction method of Interval-Valued Fuzzy Sets is defined, whose length represents such ignorance degree; (3) the fuzzy reasoning method is extended to work with this representation of the linguistic terms; (4) an evolutionary tuning step is applied for computing the optimal ignorance degree for each Interval-Valued Fuzzy Set. The experimental study shows that the IIVFDT method allows the results provided by the initial fuzzy ID3 with and without Interval-Valued Fuzzy Sets to be outperformed. The suitability of the proposed methodology is shown with respect to both several state-of-the-art fuzzy decision trees and C4.5. Furthermore, we analyze the quality of our approach versus two methods that learn the fuzzy decision tree using genetic algorithms. Finally, we show that a superior performance can be achieved by means of the positive synergy obtained when applying the well known genetic tuning of the lateral position after the application of the IIVFDT method.

High-speed corner detection based on fuzzy ID3 decision tree

A high-speed corner detection algorithm based on fuzzy ID3 decision tree was proposed. In the algorithm, the Bresenham circle with 3-pixel radius was used as the test mask, overlapping the candidate corners with the nucleus. Connected pixels on the circle were applied to compare the intensity value with the nucleus, with the membership function used to give the fuzzy result. The pixel with maximum information gain was chosen as the parent node to build a binary decision tree. Thus, the corner detector was derived. The pictures taken in Fengtai Railway Station in Beijing were used to test the method. The experimental results show that when the number of pixels on the test mask is chosen to be 9, best result can be obtained. The corner detector significantly outperforms existing detector in computational efficiency without sacrificing the quality and the method also provides high performance against Poisson noise and Gaussian blur.

Improving Classifications of Medical Data Based on Fuzzy ART2 Decision Trees

Analyzing given medical databases provide valuable references for classifying other patients symptoms. This study presents a strategy for discovering fuzzy decision trees from medical databases, in particular Harbeman's Survival database and the Blood Transfusion Service Center database. Harbeman's Survival database helps doctors treat and diagnose a group of patients who show similar past medical symptoms and the Blood Transfusion Service Center database advises individuals about when to donate blood. The proposed data mining procedure involves neural network based clustering using Adaptive Resonance Theory 2 (ART2), and the extraction of fuzzy decision trees for each homogeneous cluster of data records using fuzzy set theory. Besides, another objective of this paper is to examine the effect of the number of membership functions on building decision trees. Experiments confirm that the number of erroneously clustered patterns is significantly reduced compared to other methods without preprocessing data using ART2.

Ant colony optimization based fuzzy binary decision tree for bimodal hand knuckle verification system

In the recent trends of touch-less biometric authentication systems, hand knuckles from dorsal part of the hand is gaining popularity as a potential candidate for verification/recognition in variety of security applications. However, most of the available knuckle verification systems offer fixed security achieved for desired level of accuracy which cannot meet the varying levels of security requirements. This paper presents a bimodal knuckle verification system which is designed to meet a wide range of applications varying from civilian to high security regions. We use ant colony optimization (ACO) to choose the optimal fusion parameters corresponding to each level of security. The developed verification system utilizes fuzzy binary decision tree (FBDT) which is aimed at decision making in two classes: genuine (accept) and imposter (reject) using matching scores computed from the knuckle database. The FBDT is implemented using fuzzy Gini index for the selection of the tree nodes. The experiments are carried out on four publicly available HongKong PolyU knuckle databases named as: left index, right index, left middle and right middle with four bimodal systems: left–right index, left–right middle, left index–middle and right index–middle. The experimental results from these four bimodal knuckle databases validate the contributions of the proposed work.

Fuzzy binary decision tree for biometric based personal authentication

The use of fuzzy decision trees is yet to be ascertained for the biometric based personal authentications. This paper therefore presents a fuzzy binary decision tree (FBDT) algorithm for decision making on two classes: genuine and imposter using matching scores computed from the biometric databases. The proposed FBDT makes use of two criteria: fuzzy Gini index and fuzzy entropy for the selection of the tree nodes. The fuzzy membership functions can be automatically computed from the training scores and these are employed in two modes: Same function mode, where only one membership function is used for both the classes and Different function mode, where separate functions are used for both the classes. The parameters computed at the learning stages are used for the classification of the claimed identity in any of the two classes. Over-fitting of feature data often results in false branches in the decision trees. So the pruning of the tree is required with the consequent increase in computational complexity. Most of the FBDTs in this work are found to have lesser size than DTs as ascertained from the experimental results. The proposed FBDT is tested on two publically available databases and it fares well over its crisp counterpart.