Tree-based Ensemble Classifier Research Articles

Detecting anomalies in blockchain transactions using machine learning classifiers and explainability analysis

As the use of blockchain for digital payments continues to rise, it becomes susceptible to various malicious attacks. Successfully detecting anomalies within blockchain transactions is essential for bolstering trust in digital payments. However, the task of anomaly detection in blockchain transaction data is challenging due to the infrequent occurrence of illicit transactions. Although several studies have been conducted in the field, a limitation persists: the lack of explanations for the model's predictions. This study seeks to overcome this limitation by integrating explainable artificial intelligence (XAI) techniques and anomaly rules into tree-based ensemble classifiers for detecting anomalous Bitcoin transactions. The shapley additive explanation (SHAP) method is employed to measure the contribution of each feature, and it is compatible with ensemble models. Moreover, we present rules for interpreting whether a Bitcoin transaction is anomalous or not. Additionally, we introduce an under-sampling algorithm named XGBCLUS, designed to balance anomalous and non-anomalous transaction data. This algorithm is compared against other commonly used under-sampling and over-sampling techniques. Finally, the outcomes of various tree-based single classifiers are compared with those of stacking and voting ensemble classifiers. Our experimental results demonstrate that: (i) XGBCLUS enhances true positive rate (TPR) and receiver operating characteristic-area under curve (ROC-AUC) scores compared to state-of-the-art under-sampling and over-sampling techniques, and (ii) our proposed ensemble classifiers outperform traditional single tree-based machine learning classifiers in terms of accuracy, TPR, and false positive rate (FPR) scores.

Classification of operational states in porous journal bearings using a semi-supervised multi-sensor Machine Learning approach

Tribosystems are designed for long-lasting operation necessitating condition monitoring to identify critical regimes in time. On the example of porous journal bearings we propose a new approach for the classification of the operational states.Tribometer data from accelerated lifetime tests were used for developing a semi-supervised multi-sensor Machine Learning (ML) classifier. To advance the state-of-the art and gain higher classification accuracy, data from multiple sensors were combined by calculating time and frequency feature sets. The frequencies found in the sensor data were traced using a 3D scanning vibrometer and originated from eigenmodes and vibration states of the system. Relevant features were selected using Sequential Forward Floating Selection (SFFS) to enhance the performance of the model. Labeling was assisted by explorative data analysis techniques, i.e., principal component analysis (PCA) and subsequent hierarchical (Ward) clustering of the feature sets.A tree-based ensemble classifier (Extra-Trees) differentiating four tribological regimes from ‘Run-in’ to ‘Critical’ yielded a classification accuracy of 0.98. The algorithm was able to reliably detect states even when applied to data from previously unseen experiments and to predict terminal failure of a bearing up to 50 h before it occurred.

Predicting residual stress of aluminum nitride thin-film by incorporating manifold learning and tree-based ensemble classifier

The optical emission spectroscopy (OES) data provide multi-featured and high-dimension data, which exhibit rich physical phenomena. The novel unsupervised learning uniform manifold approximation and projection (UMAP) and t-distributed stochastic neighborhood embedding (t-SNE) are implemented to deal with OES data and keeping data structure during the aluminum nitride (AlN) film deposition process on Si (100) substrate. After dimensionality reduction, we hand over data to a tree-based ensemble model for training, and the performance of the model through indicators such as recall, precision, and F1-score was also determined. The optimal testing results of this study of compressive and tensile stress classification prediction were presented by UMAP including true positive (TP) 34.32%, false positive (FP) 0.21%, false negative (FN) 0.71%, and true negative (TN) 64.76%. Critical indicators were also evaluated with excellent results such as recall 0.9797, precision 0.9939, and F1-score 0.9867. Therefore, we proved a residual stress prediction with machine learning studies would become workable in the thin film deposition process.

Computational Intelligence for Observation and Monitoring: A Case Study of Imbalanced Hyperspectral Image Data Classification.

Imbalance in hyperspectral images creates a crisis in its analysis and classification operation. Resampling techniques are utilized to minimize the data imbalance. Although only a limited number of resampling methods were explored in the previous research, a small quantity of work has been done. In this study, we propose a novel illustrative study of the performance of the existing resampling techniques, viz. oversampling, undersampling, and hybrid sampling, for removing the imbalance from the minor samples of the hyperspectral dataset. The balanced dataset is classified in the next step, using the tree-based ensemble classifiers by including the spectral and spatial features. Finally, the comparative study is performed based on the statistical analysis of the outcome obtained from those classifiers that are discussed in the results section. In addition, we applied a new ensemble hybrid classifier named random rotation forest to our dataset. Three benchmark hyperspectral datasets: Indian Pines, Salinas Valley, and Pavia University, are applied for performing the experiments. We have taken precision, recall, F score, Cohen kappa, and overall accuracy as assessment metrics to evaluate our model. The obtained result shows that SMOTE, Tomek Links, and their combinations stand out to be the more optimized resampling strategies. Moreover, the ensemble classifiers such as rotation forest and random rotation ensemble provide more accuracy than others of their kind.

Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [18F]FDG PET/CT features.

One of the challenges in the management of patients with follicular lymphoma (FL) is the identification of individuals with histological transformation, most commonly into diffuse large B-cell lymphoma (DLBCL). [18F]FDG-PET/CT is used for staging of patients with lymphoma, but visual interpretation cannot reliably discern FL from DLBCL. This study evaluated whether radiomic features extracted from clinical baseline [18F]FDG PET/CT and analyzed by machine learning algorithms may help discriminate FL from DLBCL. Patients were selected based on confirmed histopathological diagnosis of primary FL (n=44) or DLBCL (n=76) and available [18F]FDG PET/CT with EARL reconstruction parameters within 6 months of diagnosis. Radiomic features were extracted from the volume of interest on co-registered [18F]FDG PET and CT images. Analysis of selected radiomic features was performed with machine learning classifiers based on logistic regression and tree-based ensemble classifiers (AdaBoosting, Gradient Boosting, and XG Boosting). The performance of radiomic features was compared with a SUVmax-based logistic regression model. From the segmented lesions, 121 FL and 227 DLBCL lesions were included for radiomic feature extraction. In total, 79 radiomic features were extracted from the SUVmap, 51 from CT, and 6 shape features. Machine learning classifier Gradient Boosting achieved the best discrimination performance using 136 radiomic features (AUC of 0.86 and accuracy of 80%). SUVmax-based logistic regression model achieved an AUC of 0.79 and an accuracy of 70%. Gradient Boosting classifier had a significantly greater AUC and accuracy compared to the SUVmax-based logistic regression (p≤0.01). Machine learning analysis of radiomic features may be of diagnostic value for discriminating FL from DLBCL tumor lesions, beyond that of the SUVmax alone.

Modeling freight mode choice using machine learning classifiers: a comparative study using Commodity Flow Survey (CFS) data

ABSTRACT This study explores the usefulness of machine learning classifiers for modeling freight mode choice. We investigate eight commonly used machine learning classifiers, namely Naïve Bayes, Support Vector Machine, Artificial Neural Network, K-Nearest Neighbors, Classification and Regression Tree, Random Forest, Boosting and Bagging, along with the classical Multinomial Logit model. US 2012 Commodity Flow Survey data are used as the primary data source; we augment it with spatial attributes from secondary data sources. The performance of the classifiers is compared based on prediction accuracy results. The current research also examines the role of sample size and training-testing data split ratios on the predictive ability of the various approaches. In addition, the importance of variables is estimated to determine how the variables influence freight mode choice. The results show that the tree-based ensemble classifiers perform the best. Specifically, Random Forest produces the most accurate predictions, closely followed by Boosting and Bagging. With regard to variable importance, shipment characteristics, such as shipment distance, industry classification of the shipper and shipment size, are the most significant factors for freight mode choice decisions.

In-sensor classification with boosted race trees

When extremely low-energy processing is required, the choice of data representation makes a tremendous difference. Each representation (e.g., frequency domain, residue coded, and log-scale) embodies a different set of tradeoffs based on the algebraic operations that are either easy or hard to perform in that domain. We demonstrate the potential of a novel form of encoding, race logic, in which information is represented as the delay in the arrival of a signal. Under this encoding, the ways in which signal delays interact and interfere with one another define the operation of the system. Observations of the relative delays (for example, the outcome of races between signals) define the output of the computation. Interestingly, completely standard hardware logic elements can be repurposed to this end and the resulting embedded systems have the potential to be extremely energy efficient. To realize this potential in a practical design, we demonstrate two different approaches to the creation of programmable tree-based ensemble classifiers in an extended set of race logic primitives; we explore the trade-offs inherent to their operation across sensor, hardware architecture, and algorithm; and we compare the resulting designs against traditional state-of-the-art hardware techniques.

Spline-rule ensemble classifiers with structured sparsity regularization for interpretable customer churn modeling

An important business domain that relies heavily on advanced statistical- and machine learning algorithms to support operational decision-making is customer retention management. Customer churn prediction is a crucial tool to support customer retention. It allows an early identification of customers who are at risk to abandon the company and provides the ability to gain insights into why customers are at risk. Hence, customer churn prediction models should complement predictive performance with model insights. Inspired by their ability to reconcile strong predictive performance and interpretability, this study introduces rule ensembles and their extension, spline-rule ensembles, as a promising family of classification algorithms to the customer churn prediction domain. Spline-rule ensembles combine the flexibility of a tree-based ensemble classifier with the simplicity of regression analysis. They do, however, neglect the relatedness between potentially conflicting model components which can introduce unnecessary complexity in the models and compromises model interpretability. To tackle this issue, a novel algorithmic extension, spline-rule ensembles with sparse group lasso regularization (SRE-SGL) is proposed to enhance interpretability through structured regularization. Experiments on fourteen real-world customer churn data sets in different industries (i) demonstrate the superior predictive performance of spline-rule ensembles with sparse group lasso over a set well yet powerful benchmark methods in terms of AUC and top decile lift; (ii) show that spline-rule ensembles with sparse group lasso regularization significantly outperform conventional rule ensembles whilst performing at least as well as conventional spline-rule ensembles; and (iii) illustrate the interpretable nature of a spline-rule ensemble model and the advantage of structured regularization in SRE-SGL by means of a case study on customer churn prediction for a telecommunications company.

A Multi-Stage predictive model for missed appointments at outpatient primary care settings serving rural areas

ABSTRAT Missed appointments are a significant cause of inefficiency in the healthcare industry. Many researchers have studied this problem in various healthcare settings. However, a few studies are concerned with predicting missed appointments at outpatient primary care settings serving rural areas. This study holistically investigates the factors behind two types of missed appointments - no shows and cancelations - at an outpatient primary care medical center serving rural areas and develops a predictive model to reduce their incidence. The study was carried out in three main phases. First, exploratory data analysis was conducted to discover the patterns related to missed appointments. Second, the association between some of the attributes and appointment status was analyzed. Third, three prediction models – binary, multi-class, multi-stage chain - were considered for missed appointments. The third model is a new proposed multi-stage chain model to predict missed appointments. Machine learning classifiers including logistic regression, decision tree, and tree-based ensemble classifiers were used in the three models. It was found that appointment lead time is a key driver for missed appointments. The multi-stage chain model produced the best results with 73.0% precision, 73.3% recall, 73.0% F1-score, and 73.3% accuracy. Based on this analysis, several interventions were proposed to reduce missed appointments.

ForgeNet: a graph deep neural network model using tree-based ensemble classifiers for feature graph construction.

A unique challenge in predictive model building for omics data has been the small number of samples (n) versus the large amount of features (p). This 'n≪p' property brings difficulties for disease outcome classification using deep learning techniques. Sparse learning by incorporating known functional relationships between the biological units, such as the graph-embedded deep feedforward network (GEDFN) model, has been a solution to this issue. However, such methods require an existing feature graph, and potential mis-specification of the feature graph can be harmful on classification and feature selection. To address this limitation and develop a robust classification model without relying on external knowledge, we propose a forest graph-embedded deep feedforward network (forgeNet) model, to integrate the GEDFN architecture with a forest feature graph extractor, so that the feature graph can be learned in a supervised manner and specifically constructed for a given prediction task. To validate the method's capability, we experimented the forgeNet model with both synthetic and real datasets. The resulting high classification accuracy suggests that the method is a valuable addition to sparse deep learning models for omics data. The method is available at https://github.com/yunchuankong/forgeNet. tianwei.yu@emory.edu. Supplementary data are available at Bioinformatics online.

An Improved Scheme for Vibration-Based Rolling Bearing Fault Diagnosis Using Feature Integration and AdaBoost Tree-Based Ensemble Classifier

Bearings are key components in modern power machines. Effective diagnosis of bearing faults is crucial for normal operation. Recently, the deep convolutional neural network (DCNN) with 2D visualization technology has shown great potential in bearing fault diagnosis. Traditional DCNN-based fault diagnosis mostly adopts a single learner with one input and is time-consuming in sample and network construction to obtain a satisfied performance. In this paper, a scheme combining diverse DCNN learners and an AdaBoost tree-based ensemble classifier is proposed to improve the diagnosis performance and reduce the requirement of sample and network construction simultaneously. In this scheme, multiple types of samples can be constructed independently and employed for diagnosis simultaneously; next, the same number of DCNN learners are built for underlying features extraction and the obtained results are integrated and finally fed into the ensemble classifier for fault diagnosis. An illustration based on the Case Western Reserve University datasets is given, which proves the superiority of the proposed scheme in both accuracy and robustness. Herein, we present a universal scheme to improve the diagnosis performance, and give an example for practical application, where the signal preprocessing and image sample construction methods can also be applied in other vibration-based analysis.

Potential of Ensemble Learning to Improve Tree-Based Classifiers for Landslide Susceptibility Mapping

Ensemble learning methods have been widely used due to their remarkable generalized performance, but their potential in landslide spatial prediction application is not fully studied. To take full advantage of ensemble learning techniques, the classification and regression tree classifier and four tree-based ensemble classifiers of random forest, extremely randomized tree, gradient boosting decision trees, and extreme gradient boosting decision trees are used in this study for landslide susceptibility assessment. Specifically, a stacking ensemble learning framework coupled with embedded feature selection is presented, consisting of multiple tree-based classifiers mentioned previously as base learners and logistic regression as a metalearner in a two-layer structure. In the study area of Yongxin, China, 364 historical landslide locations were first randomly partitioned into a ratio of 7/3 for training and testing the model. Then, a spatial database of 16 landslide causative factors was constructed for landslide prediction. Meanwhile, the relative importance of these factors were quantified by using the total number of feature splits and the average Gini index during the training process, and a novel embedded feature selection method was used in the base learner of the proposed framework to further improve the computational efficiency and predictive performance by allowing each base learner to obtain its own optimal subfeature space. Finally, different methods were assessed by using several evaluation criteria. Experimental results demonstrated that the proposed ensemble learning framework had the highest area under the curve value of 0.864, and it is more effective than the conventional tree-based classifiers and other ensemble learning methods.

A Comprehensive Study of Smartphone-Based Indoor Activity Recognition via Xgboost

Recently, multi-floor indoor positioning has become increasingly interesting for researchers, in which accurate recognition of indoor activities is critical for the detection of floor changes and the improvement of positioning accuracy according to indoor landmarks. However, we have not found a comprehensive study for recognizing indoor activities related to multi-floor indoor positioning based on a robust machine learning algorithm. In this work, we propose a framework for recognizing five indoor activities, i.e., walking, stillness, stair climbing, escalator, or elevator taking. In this framework, we investigate the relevant sensors and features to improve the recognition accuracy of these activities, especially some specific features in the frequency domain and wavelet domain. We propose to utilize a promising tree-based ensemble learning classifier, XGBoost, to recognize these activities. Based on our dataset created by 40 volunteers, we provide a comprehensive analysis of the proposed framework for indoor activity recognition. Considering both accuracy and computational cost, the XGBoost-based indoor activity recognition algorithm outperforms the other ensemble learning classifiers and single classifiers, and the average recognition F-score of XGBoost reaches 84.41%. In addition, our introduced specific features in the frequency domain and wavelet domain can significantly improve the recognition accuracy. Moreover, we use a publicly available dataset to verify our proposed framework and XGBoost classifier reaches 84.19% that outperforms the other classifiers.

AF detection from ECG recordings using feature selection, sparse coding, and ensemble learning

Objective: The objective of this paper is to provide an algorithm for accurate, automated detection of atrial fibrillation (AF) from ECG signals. Four types of ECG signals are considered: normal signals, signals representing symptoms of AF, other signals, and noisy signals. This paper represents follow-up work to the authors’ entry in the 2017 PhysioNet Challenge as reported in the 2017 Computing in Cardiology Conference. Approach: Our approach involves extracting features from the ECG waveform and training a machine learning classifier. In feature extraction, we calculate several statistical features related to the ECG signal and fiduciary points. We also used a disciplined method of feature selection to reduce the dimensionality of the feature space. We also employ sparse coding as an unsupervised feature extraction tool. The classifier we use is a decision tree-based ensemble learning classifier. Main results: When applied to the hidden test data reserved by the PhysioNet Challenge organizers, our classifier reports F1 scores of 0.91, 0.78, and 0.71 for the Normal, AF, and Other classes, respectively. The overall test score is 0.80, and is obtained by averaging the F1 scores for these three classes. Significance: This work demonstrates that feature selection and ensemble learning can be used to improve the performance of ECG-based classification of AF.

Tree-based Ensemble Classifier Learning for Automatic Brain Glioma Segmentation

Abstract We introduce a dynamic multiscale tree (DMT) architecture that learns how to leverage the strengths of different state-of-the-art classifiers for supervised multi-label image segmentation. Unlike previous works that simply aggregate or cascade classifiers for addressing image segmentation and labeling tasks, we propose to embed strong classifiers into a tree structure that allows bi-directional flow of information between its classifier nodes to gradually improve their performances. Our DMT is a generic classification model that inherently embeds different cascades of classifiers while enhancing learning transfer between them to boost up their classification accuracies. Specifically, each node in our DMT can nest a Structured Random Forest (SRF) classifier or a Bayesian Network (BN) classifier. The proposed SRF-BN DMT architecture has several appealing properties. First, while SRF operates at a patch-level (regular image region), BN operates at the super-pixel level (irregular image region), thereby enabling the DMT to integrate multi-level image knowledge in the learning process. Second, although BN is powerful in modeling dependencies between image elements (superpixels, edges) and their features, the learning of its structure and parameters is challenging. On the other hand, SRF may fail to accurately detect very irregular object boundaries. The proposed DMT robustly overcomes these limitations for both classifiers through the ascending and descending flow of contextual information between each parent node and its children nodes. Third, we train DMT using different scales for input patches and superpixels. Basically, as we go deeper along the tree edges nearing its leaf nodes, we progressively decrease the patch and superpixel sizes, producing segmentation maps that capture a coarse-to-fine image details. Last, DMT demonstrates its outperformance in comparison to several state-of-the-art segmentation methods for multi-labeling of brain images with gliomas.

Bearing Capacity of Shallow Foundations on Cohesionless Soils: A Random Forest Based Approach

Determining the ultimate bearing capacity (UBC) is vital for design of shallow foundations. Recently, soft computing methods (i.e. artificial neural networks and support vector machines) have been used for this purpose. In this paper, Random Forest (RF) is utilized as a tree-based ensemble classifier for predicting the UBC of shallow foundations on cohesionless soils. The inputs of model are width of footing (B), depth of footing (D), footing geometry (L/B), unit weight of sand (γ) and internal friction angle (ϕ). A set of 112 load tests data were used to calibrate and test the developed RF-based model. The used data set consists of 47 full-scale observations and 65 small-scale laboratory footing load tests. To demonstrate the efficiency of proposed RF-based model, the results are compared with some popular classical formulas that are most commonly used for determining the UBC. The results show the efficiency and capabilities of the proposed RF-based model as a practical tool in evaluating the UBC of shallow foundations in a fast and accurate way.

Identifying Different Transportation Modes from Trajectory Data Using Tree-Based Ensemble Classifiers

Recognition of transportation modes can be used in different applications including human behavior research, transport management and traffic control. Previous work on transportation mode recognition has often relied on using multiple sensors or matching Geographic Information System (GIS) information, which is not possible in many cases. In this paper, an approach based on ensemble learning is proposed to infer hybrid transportation modes using only Global Position System (GPS) data. First, in order to distinguish between different transportation modes, we used a statistical method to generate global features and extract several local features from sub-trajectories after trajectory segmentation, before these features were combined in the classification stage. Second, to obtain a better performance, we used tree-based ensemble models (Random Forest, Gradient Boosting Decision Tree, and XGBoost) instead of traditional methods (K-Nearest Neighbor, Decision Tree, and Support Vector Machines) to classify the different transportation modes. The experiment results on the later have shown the efficacy of our proposed approach. Among them, the XGBoost model produced the best performance with a classification accuracy of 90.77% obtained on the GEOLIFE dataset, and we used a tree-based ensemble method to ensure accurate feature selection to reduce the model complexity.

Automated classification of oral premalignant lesions using image cytometry and Random Forests-based algorithms.

A major challenge for the early diagnosis of oral cancer is the ability to differentiate oral premalignant lesions (OPL) at high risk of progressing into invasive squamous cell carcinoma (SCC) from those at low risk. Our group has previously used high-resolution image analysis algorithms to quantify the nuclear phenotypic changes occurring in OPLs. This approach, however, requires a manual selection of nuclei images. Here, we investigated a new, semi-automated algorithm to identify OPLs at high risk of progressing into invasive SCC from those at low risk using Random Forests, a tree-based ensemble classifier. We trained a sequence of classifiers using morphometric data calculated on nuclei from 29 normal, 5 carcinoma in situ (CIS) and 28 SCC specimens. After automated discrimination of nuclei from other objects (i.e., debris, clusters, etc.), a nuclei classifier was trained to discriminate abnormal nuclei (8,841) from normal nuclei (5,762). We extracted voting scores from this trained classifier and created an automated nuclear phenotypic score (aNPS) to identify OPLs at high risk of progression. The new algorithm showed a correct classification rate of 80% (80.6% sensitivity, 79.3% specificity) at the cellular level for the test set, and a correct classification rate of 75% (77.8% sensitivity, 71.4% specificity) at the tissue level with a negative predictive value of 76% and a positive predictive value of 74% for predicting progression among 71 OPLs, performed on par with the manual method in our previous study. We conclude that the newly developed aNPS algorithm serves as a crucial asset in the implementation of high-resolution image analysis in routine clinical pathology practice to identify lesions that require molecular evaluation or more frequent follow-up.

A comparative study of different classification techniques for marine oil spill identification using RADARSAT-1 imagery

The discrimination of oil spills and look-alike phenomena (e.g., low wind area, wind front area and natural slicks) on Synthetic Aperture Radar (SAR) images is a crucial task in marine oil spill detection. Many classification techniques can be employed for this purpose. In order to make the best use of the large variety of statistical and machine learning classification methods, it is necessary to assess their performance differences and make recommendations for classifier selection and improvement. The objective of this paper is to compare different classification techniques for oil-spill detection in RADARSAT-1 imagery. The data of this study consists of 15 features of 192 oil spills and look-alikes identified by Canadian Ice Service between 2004 and 2008 off Canada's east and west coastal areas. The studied classifiers include the Support Vector Machine (SVM), Artificial Neural Network (ANN), tree-based ensemble classifiers (bagging, bundling and boosting), Generalized Additive Model (GAM) and Penalized Linear Discriminant Analysis (PLDA). Two performance measures, the specificity at fixed sensitivity (80%) and the area under the Receiver Operating Characteristic (ROC) curve (AUC), were estimated using cross-validation to evaluate the performance of classifiers at a high sensitivity. Overall, the bundling technique which achieved a median specificity of 90.7%, at sensitivity of 80%, significantly outperformed the second best (i.e. bagging) by 1.5 percentage points, and the worst (i.e. ANN) by 15 percentage points. The median values of AUC measure indicated consistent results. Bundling and bagging achieved comparable median AUC values of about 92%, followed by GAM and PLDA, with ANN yielding the smallest. Most classifiers (SVM, bundling and especially PLDA and ANN) performed significantly better on datasets pre-processed by log-transformation and standardization than on the original dataset. These results demonstrate the importance and benefit of selecting the optimal classifiers for oil spill classification, and configuring the classifiers by proper feature construction techniques.

Towards a comprehensive evaluation of V-I-S sub-pixel fractions and land surface temperature for urban land-use classification in the USA

Remote-sensing image classification based on the vegetation–impervious surface–soil (V-I-S) model and land-surface temperature (LST) has proved to be more efficient in characterizing the urban landscape than conventional spectral-based classification. However, current literature emphasizes discussion of the classifier's accuracy improvement achieved by the input of V-I-S fractions and LST over conventional spectral-based classification while ignoring the stability evaluation. Hence, this study proposes an evaluation framework for exploring the superiority of the input features and the stability of classifiers by integrating statistical randomization techniques and a kappa-error diagram. The evaluation framework was applied to case studies for demonstrating the superiority of V-I-S fractions and LST in the context of urban land-use classification with five different types of classifiers, including the maximum likelihood classifier (MLC), the tree classifier, the Bagging classifier, the random forest (RF) and the support vector machine (SVM). It followed that the use of V-I-S fractions and LST (1) could alleviate the ‘salt and pepper’ effect; (2) is preferred by tree and tree-based ensembles for branch splitting; (3) could produce classification trees with less complexity; (4) could benefit the stability of classifiers in addition to the accuracy improvement; and (5) could allow histograms following nearly normal distribution in its feature space, which boosts the performance of MLC. It is shown that MLC becomes comparable with modern classifiers when trained with V-I-S fractions and LST combination. Because of its adequacy and simplicity, MLC is recommended for urban land-use classification when V-I-S fractions and LST are used as the only input features. However, replacing them with, or including, the band reflectance might degrade MLC. A direct use of spectral band reflectance is not recommended for any of the classification approaches being considered in this study, except for SVM, which is the most robust classifier as it has a consistently high performance for all the input feature combinations. We recommend using tree-based ensemble classifiers or SVM when V-I-S fractions and LST as well as the band reflectance are all used in the classification. The proposed evaluation framework can also be applied to the assessment of input features and classifiers in other remote-sensing classification endeavours.