Binary Decision Tree Classifier Research Articles

Evaluation of machine learning models in the classification of pool boiling regimes up to critical heat flux based on boiling acoustics

The present study focuses on the performance of the machine learning methods in classifying the boiling regimes of water up to critical heat flux conditions based on the acoustic characteristics of boiling. The data set is generated by conducting a pool boiling experiment on a wire heater at various heat fluxes varying from 54.95 kW/m2 to 2898.67 kW/m2. A Kanthal D wire of standard wire gauge 36 is used. The data set is divided into three classes: no boiling, nucleate boiling, and critical heat flux to identify the boiling incipience and critical heat flux. Much focus is insisted on identifying critical heat flux as it carries more practical importance in the safety of the cooling systems. Data set size optimization is performed to find the lowest number of records required for each method. Three machine-learning methods are employed to predict the boiling regime, namely, binary decision tree method, decision tree ensemble method and naive Bayes method. Out of these, the decision tree ensemble outperformed the binary decision tree and naive Bayes classifiers. The decision tree ensemble classified the regimes in the given data with the lowest classification error and inference time. The accurate classification of boiling regimes based on boiling acoustics strengthens the safety measures in real-time monitoring of cooling systems.

A multivariate binary decision tree classifier based on shallow neural network

In this paper, a novel decision tree classifier based on shallow neural networks with piecewise and nonlinear transformation activation functions are presented. A shallow neural network is recursively employed into linear and non-linear multivariate binary decision tree methods which generates splitting nodes and classifier nodes. Firstly, a linear multivariate binary decision tree with a shallow neural network is proposed which employs a rectified linear unit function. Secondly, there is presented a new activation function with non-linear property which has good generalization ability in learning process of neural networks. The presented method shows high generalization ability for linear and non-linear multivariate binary decision tree models which are called a Neural Network Decision Tree (NNDT). The proposed models with high generalization ability ensure the classification accuracy and performance. A novel split criterion of generating the nodes which focuses more on majority objects of classes on the current node is presented and employed in the new NNDT models. Furthermore, a shallow neural network based NNDT models are converted into a hyperplane based linear and non-linear multivariate decision trees which has high speed in the processing classification decisions. Numerical experiments on publicly available datasets have showed that the presented NNDT methods outperform the existing decision tree algorithms and other classifier methods.

A Study on Halftoning Improvement for Low-Resolution Digital Print Engines With Machine Learning Methods

As today’s printing volume worldwide decreases, and most traditional printing engines are expensive non-digital devices (offset), the demand for a low-cost digital replacement is rapidly increasing. A main disadvantage of digital presses is the low-resolution capabilities, introducing a compromise in the print quality (PQ). A key factor of print quality is the halftoning algorithm. A very common halftoning method is amplitude modulation (AM) halftone screening, in which dots are placed on a repetitive lattice, varying in size as a function of the grey level. The main AM screen design PQ challenge for low-resolution devices is the quantization frequencies, a disturbing pattern that usually emerges when a screen is approximated to a rational angle due to low resolution. Fourier-based analysis is a classical rule-based method to filter out screens that suffer from visually disturbing quantization patterns. This work presents a new approach that tackles this challenge by incorporating machine learning with the classic Fourier-based approach. Particularly, we show that a binary decision tree classifier with a Fourier-based feature vector has an accuracy of 95% in identifying quantization-free screens compared to the classic rule-based method, which has an accuracy of 66%. We conclude by demonstrating the use of the screen classifier to design a quantization-free screen set. This is done by first applying the screen classifier to the entire screen pool, that is, the set of all possible screens for a given print engine, followed by a rosette zero-moiré offset-like screen design.

Multiparametric Ultrasound Approach Using a Tree-Based Decision Classifier for Inconclusive Focal Liver Lesions Evaluated by Contrast Enhanced Ultrasound.

Background: Multiparametric ultrasound (MPUS) is a concept whereby the examiner is encouraged to use the latest features of an ultrasound machine. The aim of this study was to reanalyze inconclusive focal liver lesions (FLLs) that had been analyzed via contrast enhanced ultrasound (CEUS) using the MPUS approach with the help of a tree-based decision classifier. Materials and methods: We retrospectively analyzed FLLs that were inconclusive upon CEUS examination in our department, focusing our attention on samples taken over a period of two years (2017−2018). MPUS reanalysis followed a three-step algorithm, taking into account the liver stiffness measurement (LSM), time–intensity curve analysis (TIC), and parametric imaging (PI). After processing all steps of the algorithm, a binary decision tree classifier (BDTC) was used to achieve a software-assisted decision. Results: Area was the only TIC-CEUS parameter that showed a significant difference between malign and benign lesions with a cutoff of >−19.3 dB for washout phenomena (AUROC = 0.58, Se = 74.0%, Sp = 45.7%). Using the binary decision tree classifier (BDTC) algorithm, we correctly classified 71 out of 91 lesions according to their malignant or benignant status, with an accuracy of 78.0% (sensitivity = 62%, specificity = 45%, and precision = 80%). Conclusions: By reevaluating inconclusive FLLs that had been analyzed via CEUS using MPUS, we managed to determine that 78% of the lesions were malignant and, in 28% of them, we established the lesion type.

Machine learning enables noninvasive prediction of atrial fibrillation driver location and acute pulmonary vein ablation success using the 12-lead ECG

BackgroundAtrial fibrillation (AF) is the most common supraventricular arrhythmia, characterized by disorganized atrial electrical activity, maintained by localized arrhythmogenic atrial drivers. Pulmonary vein isolation (PVI) allows to exclude PV-related drivers. However, PVI is less effective in patients with additional extra-PV arrhythmogenic drivers.ObjectivesTo discriminate whether AF drivers are located near the PVs vs extra-PV regions using the noninvasive 12-lead electrocardiogram (ECG) in a computational and clinical framework, and to computationally predict the acute success of PVI in these cohorts of data.MethodsAF drivers were induced in 2 computerized atrial models and combined with 8 torso models, resulting in 1128 12-lead ECGs (80 ECGs with AF drivers located in the PVs and 1048 in extra-PV areas). A total of 103 features were extracted from the signals. Binary decision tree classifier was trained on the simulated data and evaluated using hold-out cross-validation. The PVs were subsequently isolated in the models to assess PVI success. Finally, the classifier was tested on a clinical dataset (46 patients: 23 PV-dependent AF and 23 with additional extra-PV sources).ResultsThe classifier yielded 82.6% specificity and 73.9% sensitivity for detecting PV drivers on the clinical data. Consistency analysis on the 46 patients resulted in 93.5% results match. Applying PVI on the simulated AF cases terminated AF in 100% of the cases in the PV class.ConclusionMachine learning–based classification of 12-lead-ECG allows discrimination between patients with PV drivers vs those with extra-PV drivers of AF. The novel algorithm may aid to identify patients with high acute success rates to PVI.

A linear multivariate binary decision tree classifier based on K-means splitting

A novel linear multivariate decision tree classifier, Binary Decision Tree based on K-means Splitting (BDTKS), is presented in this paper. The unsupervised K-means clustering is recursively integrated into the binary tree, building a hierarchical classifier. The introduction of the unsupervised K-means clustering provides the powerful generalization ability for the resulting BDTKS model. Then, the good generalization ability of BDTKS ensures the classification performance. A novel non-split condition with an easy-setting hyperparameter which focuses more on minority classes of the current node is proposed and applied in the BDTKS model, avoiding ignoring the minority classes in the class imbalance cases. Furthermore, the K-means centroid based BDTKS model is converted into the hyperplane based decision tree, speeding up the process of classification. Extensive experiments on the publicly available data sets have demonstrated that the proposed BDTKS matches or outperforms the previous decision trees.

Drowsiness Detection using Band Power and log Energy Entropy Features Based on EEG Signals

Sleeping on the wheels due to drowsiness is one of the significant causes of death tolls all over the world. The primary reason for the sleepiness is due to lack of sleep and irregular sleep patterns. Several methods such as unobtrusive sensors, vehicle dynamics and obtrusive physiology sensors are used to diagnose drowsiness in drivers. However, the unobtrusive sensors detect drowsiness in the later stage. Whereas the physiological methods use obtrusive sensors such as electro-ocular, electro-myo and electro-encephalograms produce high accuracy in the early detection of drowsiness, which makes them a preferable solution. The objective of this research article is to classify drowsiness with alertness based on the electroencephalographic (EEG) signals using band power and log energy entropy features. A publicly available ULg DROZY database used in this research. The raw multimodal signal is processed to extract the five EEG channels. A passband filter with the cut off frequencies of 0.1 Hz and 50 Hz attenuates the high-frequency components. Another bandpass filter bank is designed to slice the raw signals into eight sub-bands, namely delta, theta, low alpha, high alpha, low beta, mid-beta, high beta and gamma. The preprocessed signals are segmented into an equal number of frames with a frame duration of 2 seconds using a rectangular time windowing approach with an overlap of 50%. Frequency domain features such as log energy entropy and band power were extracted. The extracted feature sets were further normalised between 0 and 1 and labelled as drowsy and alert and then combined to form the final dataset. The K-fold cross-validation method is used to divide the dataset into training and testing sets. The processed dataset is then trained using Discriminant analysis, k-nearest neighbour network, Binary decision tree, ensemble, Naive Bayes and support vector machine classifiers and the results are compared with the literature. The kNN classifier produces 95% classification accuracy. The developed model can provide a tool for drowsiness detection in drivers.

Machine learning algorithms for activity recognition in ambulant children and adolescents with cerebral palsy

BackgroundCerebral palsy (CP) is the most common physical disability among children (2.5 to 3.6 cases per 1000 live births). Inadequate physical activity (PA) is a major problem effecting the health and well-being of children with CP. Practical, yet accurate measures of PA are needed to evaluate the effectiveness of surgical and therapy-based interventions to increase PA. Accelerometer-based motion sensors have become the standard for objectively measuring PA in children and adolescents; however, current methods for estimating physical activity intensity in children with CP are associated with significant error and may dramatically underestimate HPA in children with more severe mobility limitations. Machine learning (ML) models that first classify the PA type and then predict PA intensity or energy expenditure using activity specific regression equations may be more accurate than standalone regression models. However, the feasibility and validity of ML methods has not been explored in youth with CP. Therefore, the purpose of this study was to develop and test ML models for the automatic identification of PA type in ambulant children with CP.MethodsTwenty two children and adolescents (mean age: 12.8 ± 2.9 y) with CP classified at GMFCS Levels I to III completed 7 activity trials while wearing an ActiGraph GT3X+ accelerometer on the hip and wrist. Trials were categorised as sedentary (SED), standing utilitarian movements (SUM), comfortable walking (CW), and brisk walking (BW). Random forest (RF), support vector machine (SVM), and binary decision tree (BDT) classifiers were trained with features extracted from the vector magnitude (VM) of the raw acceleration signal using 10 s non-overlapping windows. Performance was evaluated using leave-one-subject out cross validation.ResultsSVM (82.0–89.0%) and RF (82.6–88.8%) provided significantly better classification accuracy than BDT (76.1–86.2%). Hip (82.7–85.5%) and wrist (76.1–82.6%) classifiers exhibited comparable prediction accuracy, while the combined hip and wrist (86.2–89.0%) classifiers achieved the best overall performance. For all classifiers, recognition accuracy was excellent for SED (94.1–97.9%), good to excellent for SUM (74.0–96.6%) and brisk walking (71.5–86.0%), and modest for comfortable walking (47.6–70.4%). When comfortable and brisk walking were combined into a single walking class, recognition accuracy ranged from 90.3 to 96.5%.ConclusionsML methods provided acceptable classification accuracy for detection of a range of activities commonly performed by ambulatory children with CP. The resultant models can help clinicians more effectively monitor bouts of brisk walking in the community. The results indicate that 2-step models that first classify PA type and then predict energy expenditure using activity specific regression equations are worthy of exploration in this patient group.

Binary Decision Trees for Preoperative Periapical Cyst Screening Using Cone-beam Computed Tomography

IntroductionCone-beam computed tomographic (CBCT) analysis allows for 3-dimensional assessment of periradicular lesions and may facilitate preoperative periapical cyst screening. The purpose of this study was to develop and assess the predictive validity of a cyst screening method based on CBCT volumetric analysis alone or combined with designated radiologic criteria. MethodsThree independent examiners evaluated 118 presurgical CBCT scans from cases that underwent apicoectomies and had an accompanying gold standard histopathological diagnosis of either a cyst or granuloma. Lesion volume, density, and specific radiologic characteristics were assessed using specialized software. Logistic regression models with histopathological diagnosis as the dependent variable were constructed for cyst prediction, and receiver operating characteristic curves were used to assess the predictive validity of the models. A conditional inference binary decision tree based on a recursive partitioning algorithm was constructed to facilitate preoperative screening. ResultsInterobserver agreement was excellent for volume and density, but it varied from poor to good for the radiologic criteria. Volume and root displacement were strong predictors for cyst screening in all analyses. The binary decision tree classifier determined that if the volume of the lesion was >247 mm3, there was 80% probability of a cyst. If volume was <247 mm3 and root displacement was present, cyst probability was 60% (78% accuracy). ConclusionsThe good accuracy and high specificity of the decision tree classifier renders it a useful preoperative cyst screening tool that can aid in clinical decision making but not a substitute for definitive histopathological diagnosis after biopsy. Confirmatory studies are required to validate the present findings.

A machine vision system for automatic detection of parasites Edotea magellanica in shell-off cooked clam Mulinia edulis

In this work, a machine vision system for the automatic online detection of the parasite Edotea magellanica inside of the clam Mulinia edulis is presented. The machine vision system uses a transillumination technique to acquire images of the clam-parasite tandem. To improve the light transmission properties of the clam and parasite tissues, a novel online flattening system to flatten the clam thickness was developed. The automatic detection of the parasite in a clam is accomplished by a binary decision tree classifier that analyzes clam images. The classifier was developed using a supervised pattern-recognition approach, and proper features were created using spectral, spatial, geometrical, and biological information, such as the relative transmittance, the location of the parasite in the clam, the shape of the parasite, and the anatomical regions of the clam. The prototype system was built and tested, achieving an average classification accuracy of 98%, for a laboratory training sample set of 200 clams, and an accuracy of 73% for another set of 200 clams in an industrial setting.

A Dictionary Approach to Electron Backscatter Diffraction Indexing

We propose a framework for indexing of grain and subgrain structures in electron backscatter diffraction patterns of polycrystalline materials. We discretize the domain of a dynamical forward model onto a dense grid of orientations, producing a dictionary of patterns. For each measured pattern, we identify the most similar patterns in the dictionary, and identify boundaries, detect anomalies, and index crystal orientations. The statistical distribution of these closest matches is used in an unsupervised binary decision tree (DT) classifier to identify grain boundaries and anomalous regions. The DT classifies a pattern as an anomaly if it has an abnormally low similarity to any pattern in the dictionary. It classifies a pixel as being near a grain boundary if the highly ranked patterns in the dictionary differ significantly over the pixel's neighborhood. Indexing is accomplished by computing the mean orientation of the closest matches to each pattern. The mean orientation is estimated using a maximum likelihood approach that models the orientation distribution as a mixture of Von Mises-Fisher distributions over the quaternionic three sphere. The proposed dictionary matching approach permits segmentation, anomaly detection, and indexing to be performed in a unified manner with the additional benefit of uncertainty quantification.

Rank discrimination measures for enforcing monotonicity in decision tree induction

Monotone classification is a relatively recent topic in machine learning in which the classification function to learn is asked to guarantee a sort of monotonicity of the class with respect to attribute values. Nevertheless, real datasets are quite far from being monotone and this can sharply limit the performance of purely monotone classifiers while standard classifiers are simply insensitive to monotonicity. Here we focus on rank discrimination measures to be used in decision tree induction, i.e., functions able to measure the discrimination power of an attribute with respect to the class taking into account the monotonicity of the class with respect to the attribute. Three new measures are studied in detail and a hierarchical construction model is derived allowing the formal definition of a general rank discrimination measure. Our measures have been compared with other well-known proposals, quantifying both the accuracy and the monotonicity of the resulting binary decision tree classifiers.

A Low-Power Hardware-Friendly Binary Decision Tree Classifier for Gas Identification

In this paper, we present a hardware friendly binary decision tree (DT) classifier for gas identification. The DT classifier is based on an axis-parallel decision tree implemented as threshold networks—one layer of threshold logic units (TLUs) followed by a programmable binary tree implemented using combinational logic circuits. The proposed DT classifier circuit removes the need for multiplication operation enabling up to 80% savings in terms of silicon area and power compared to oblique based-DT while achieving 91.36% classification accuracy without throughput degradation. The circuit was designed in 0.18 μm Charter CMOS process and tested using a data set acquired with in-house fabricated tin-oxide gas sensors.

New variants of pairwise classification

Abstract Pairwise classification is the technique that deals with multi-class problems by converting them into a series of binary problems, one for each pair of classes. Typically, K-class classification rules tend to be easier to learn for K = 2 than for K > 2 – only one decision boundary requires attention. This paper presents new methods for obtaining class membership probability estimates for multi-class classification problems by coupling the probability estimates created by binary classifiers. Classifiers used include linear Bayes normal classifier, Parzen density based classifier, naive Bayes classifier, binary decision tree classifier and random neural net classifier. The accuracy of new pairwise classifiers is examined on some real data sets. The classification errors were estimated by stratified version of 10-fold cross-validation technique, i.e. the training examples were partitioned into 10 equal-sized blocks with similar class distributions as in the original set. The validation technique was repeated 10 times for each data set.

Self-localization algorithm for sensor networks using SVM classification region

Focused on the requirements of low cost and low power in Wireless Sensor Network(WSN),this paper proposed a range-free localization algorithm based on Support Vector Machine(SVM) classification regions.First,SVM constructed a binary decision tree classifier via learning of the training data.Then the classifier determined the certain classification region where the unknown nodes were located.Finally,the study used the region's center point as the estimated position of the unknown node.The proposed algorithm required mere connectivity information(i.e.,hop counts only),so as to reduce the network cost and communication load.The simulation results show that this algorithm alleviates the coverage holes and border problem significantly while certain location accuracy is assured.

Understory Bamboo Discrimination Using a Winter Image

In this study, a new approach is presented that combines forest phenology and Landsat vegetation indices to estimate evergreen understory bamboo coverage in a mixed temperate forest. It was found that vegetation indices, especially the normalized difference vegetation index (NDVI) derived from leaf-off (winter) images were significantly correlated with percent understory bamboo cover for both deciduous and mixed coniferous/deciduous forests. Winter NDVI was used to map bamboo coverage using a binary decision tree classifier. A high mapping accuracy for understory bamboo presence/absence was achieved with an overall accuracy of 89 percent ( k5 0.59). In addition, for the first time, we successfully classified three density classes of bamboo with an overall accuracy of 68 percent ( k5 0.48). These results were compared to three traditional multispectral bandsbased methods (Mahalanobis distance, maximum likelihood, and artificial neural networks). The highest mapping accuracy was again obtained from winter images. However, the kappa z-test showed that there was no statistical difference in accuracy between the methods. The results suggest that winter is the optimal season for quantifying the coverage of evergreen understory bamboos in a mixed forest area, regardless of the classification methods use.

Geometric probing of dense range data

A new method is presented for the efficient and reliable pose determination of 3D objects in dense range image data. The method is based upon a minimalistic Geometric Probing strategy that hypothesizes the intersection of the object with some selected image point, and searches for additional surface data at locations relative to that point. The strategy is implemented in the discrete domain as a binary decision tree classifier. The tree leaf nodes represent individual voxel templates of the model, with one template per distinct model pose. The internal nodes represent the union of the templates of their descendant leaf nodes. The union of all leaf node templates is the complete template set of the model over its discrete pose space. Each internal node also encodes a single voxel which is the most common element of its child node templates. Traversing the free is equivalent to efficiently matching the large set of templates at a selected image seed location. The method was implemented and extensive experiments were conducted for a variety of combinations of tree designs and traversals under isolated, cluttered, and occluded scene conditions. The results demonstrated a tradeoff between efficiency and reliability. It was concluded that there exist combinations of tree design and traversal which are both highly efficient and reliable.

Construction and application of hierarchical decision tree for classification of ultrasonographic prostate images.

A non-parametric algorithm is described for the construction of a binary decision tree classifier. This tree is used to correlate textural features, computed from ultrasonographic prostate images, with the histopathology of the imaged tissue. The algorithm consists of two parts; growing and pruning. In the growing phase an optimal tree is grown, based on the concept of mutual information. After growing, the tree is pruned by an alternating interaction of two data sets. Moreover, the structure and performance of the constructed tree are compared to the results using a slightly modified corresponding growing and pruning algorithm. The modified algorithm provides better retrospective and prospective classification results than the original algorithm. The use of the tree for automated cancer detection in ultrasonographic prostate images results in retrospective and prospective accuracy of 77.9% and 72.3%, respectively. Using this tissue characterisation, a supporting tool is provided for the interpretation of transrectal ultrasonographic images.

Data fusion with artificial neural networks for classification of earth surface from microwave satellite measurements

A data fusion system with artificial neural networks (ANN) is used for fast and accurate classification of five earth surface conditions and surface changes, based on seven Special Sensor Microwave Imager (SSMI) multichannel microwave satellite measurements. The measurements include brightness temperatures at 19, 22, 37, and 85 GHz at both horizontal and vertical polarizations (only vertical at 22 GHz). The seven channel measurements are processed through a convolution computation such that all measurements are located at same grid. Five surface classes including nonscattering surface, precipitation over land, over ocean, snow, and desert are identified from ground-truth observations. The system processes sensory data in three consecutive phases: (a) preprocessing to extract feature vectors and enhance separability among detected classes; (b) preliminary classification of Earth surface patterns using two separate and parallel-acting classifiers: back-propagation neural network and binary decision tree classifiers; and (c) data fusion of results from preliminary classifiers to obtain the optimal performance in over-all classification. Both the binary decision tree classifier and the fusion processing centers are implemented by neural network architectures. The fusion system configuration is a hierarchical neural network architecture, in which each functional neural net will handle different processing phases in a pipelined fashion. There is a total of around 13,500 samples for this analysis, of which 4% are used as the training set and 96% as the testing set. After training, this classification system is able to bring up the detection accuracy to 94% compared with 88% for back-propagation artificial neural networks and 80% for binary decision tree classifiers. The neural network data fusion classification is currently under progress to be integrated in an image processing system at NOAA and to be implemented in a prototype of a massively parallel and dynamically reconfigurable Modular Neural Ring (MNR).

Nonparametric classification using matched binary decision trees

Abstract The paper describes a new technique for designing binary decision tree classifiers. Decision features are chosen at each node in such a way that the classifier formed by combining the decision trees behaves in an optimal fashion.