Narrow Results

Facial expression detection technology is a flexible tool utilized in many different industries that allows for a better understanding of human behavior and emotional responses. Its applications span a broad spectrum of societal requirements, from enhancing marketing strategies to bolstering security protocols and assisting in the diagnosis of mental health issues. By using the power of artificial intelligence and computer vision, facial expression recognition (FER) helps businesses to adapt, grow and build deeper interactions between people and the technology and environments they live in. Its primary goal remains to facilitate better understanding, communication and decision-making in order to bring in a new wave of technical innovation that is driven by people’s needs. Facial emotion recognition, which makes use of computer vision and artificial intelligence, not only improves the capacity to understand and react to human emotions but also opens the door to more personalized and empathic interactions in an increasingly digital environment. In this paper, unique method of facial emotion recognition based on sophisticated feature extraction techniques combined with a hybrid ensemble learning technique. In particular, hierarchical cascade regression neural networks (HCRNN) for face landmark identification, VGG-19 for feature extraction and Random Forests are considered as classification techniques. In order to accomplish reliable and accurate emotion recognition from facial images, this method takes advantage of the complementary capabilities of both approaches. Initially, we utilized a deep convolutional neural network (CNN) called VGG-19 to extract high-level features from face images. Pre-trained on extensive image datasets, VGG-19 has proven to be highly effective at capturing complex visual representations. These characteristics function as an all-inclusive description of facial expressions, encompassing both general trends and minute details. HCRNN for facial landmark detection, which precisely locates important face features including the mouth, nose and eyes. Facial landmark predictions are iteratively improved by this hierarchical cascade architecture, which also handles occlusions and changes in illumination and position. Lastly, we use an ensemble learning technique called Random Forests to classify emotions. Multiple decision trees’ predictions are combined by random forests, which offer robustness against overfitting and efficiently handle high-dimensional feature fields. High accuracy and generalization capabilities are achieved in the facial emotion classification process by Random Forests, which combine the feature representations retrieved by VGG-19 with the localized facial landmarks recognized by HCRNN. We tested the suggested strategy on benchmark face emotion detection datasets and compared its results with the most advanced techniques available. Performance measures like accuracy of 98.89%, precision 96.52%, recall 96%, F1-score 99% and receiver operating characteristic (ROC) curve of 97.24% analysis were used to evaluate how well hybrid strategy worked and how much better it was at identifying emotions in facial images. The outcomes of experiments show how method has practical applications in human–computer interaction, healthcare and other fields.

Software defect prediction models that use software metrics such as code-level measurements and defect data to build classification models are useful tools for identifying potentially-problematic program modules. Effectiveness of detecting such modules is affected by the software measurements used, making data preprocessing an important step during software quality prediction. Generally, there are two problems affecting software measurement data: high dimensionality (where a training dataset has an extremely large number of independent attributes, or features) and class imbalance (where a training dataset has one class with relatively many more members than the other class). In this paper, we present a novel form of ensemble learning based on boosting that incorporates data sampling to alleviate class imbalance and feature (software metric) selection to address high dimensionality. As we adopt two different sampling methods (Random Undersampling (RUS) and Synthetic Minority Oversampling (SMOTE)) in the technique, we have two forms of our new ensemble-based approach: selectRUSBoost and selectSMOTEBoost. To evaluate the effectiveness of these new techniques, we apply them to two groups of datasets from two real-world software systems. In the experiments, four learners and nine feature selection techniques are employed to build our models. We also consider versions of the technique which do not incorporate feature selection, and compare all four techniques (the two different ensemble-based approaches which utilize feature selection and the two versions which use sampling only). The experimental results demonstrate that selectRUSBoost is generally more effective in improving defect prediction performance than selectSMOTEBoost, and that the techniques with feature selection do help for getting better prediction than the techniques without feature selection.

The challenge of assessing semantic similarity between pieces of text through computers has attracted considerable attention from industry and academia. New advances in neural computation have developed very sophisticated concepts, establishing a new state of the art in this respect. In this paper, we go one step further by proposing new techniques built on the existing methods. To do so, we bring to the table the stacking concept that has given such good results and propose a new architecture for ensemble learning based on genetic programming. As there are several possible variants, we compare them all and try to establish which one is the most appropriate to achieve successful results in this context. Analysis of the experiments indicates that Cartesian Genetic Programming seems to give better average results.

For a new project, it is impossible to get a reliable prediction model because of the lack of sufficient training data. To solve the problem, researchers proposed cross-project defect prediction (CPDP). For CPDP, most researchers focus on how to reduce the distribution difference between training data and test data, and ignore the impact of class imbalance on prediction performance. This paper proposes a hybrid multiple models transfer approach (HMMTA) for cross-project software defect prediction. First, several instances that are most similar to each target project instance are selected from all source projects to form the training data. Second, the same number of instances as that of the defected class are randomly selected from all the non-defect class in each iteration. Next, instances selected from the non-defect classes and all defected class instances are combined to form the training data. Third, the transfer learning method called ETrAdaBoost is used to iteratively construct multiple prediction models. Finally, the prediction models obtained from multiple iterations are integrated by the ensemble learning method to obtain the final prediction model. We evaluate our approach on 53 projects from AEEEM, PROMISE, SOFTLAB and ReLink four defect repositories, and compare it with 10 baseline CPDP approaches. The experimental results show that the prediction performance of our approach significantly outperforms the state-of-the-art CPDP methods. Besides, we also find that our approach has the comparable prediction performance as within-project defect prediction (WPDP) approaches. These experimental results demonstrate the effectiveness of HMMTA approach for CPDP.

Software defect prediction can detect modules that may have defects in advance and optimize resource allocation to improve test efficiency and reduce development costs. Traditional features cannot capture deep semantic and grammatical information, which limits the further development of software defect prediction. Therefore, it has gradually become a trend to use deep learning technology to automatically learn valuable deep features from source code or relevant data. However, most software defect prediction methods based on deep learning extraction features from a single information source or only use a single deep learning model, which leads to the fact that the extracted features are not comprehensive enough to affect the final prediction performance. In view of this, this paper proposes a Hierarchical Feature Ensemble Deep Learning (HFEDL) Approach for software defect prediction. Firstly, the HFEDL approach needs to obtain three types of information sources: abstract syntax tree (AST), class dependency network (CDN) and traditional features. Then, the Convolutional Neural Network (CNN) and the Bidirectional Long Short-Term Memory based on Attention mechanism (BiLSTM+Attention) are used to extract different valuable features from the three information sources and multiple prediction sub-models are constructed. Next, all the extracted features are fused by a filter mechanism to obtain more comprehensive features and construct a fusion prediction sub-model. Finally, all the sub-models are integrated by an ensemble learning method to obtain the final prediction model. We use 11 projects in the PROMISE defect repository and evaluate our approach in both non-effort-aware and effort-aware scenarios. The experimental results show that the prediction performance of our approach is superior to state-of-the-art methods in both scenarios.

There has been a recent push for a new framework of learning, due in part to the availability of storage, networking and the abundance of very large datasets. This framework argues for parallelized learning algorithms that can operate on a distributed platform and have the ability to terminate early in the likely event that data size is too inundating. Methods described herein propose a subsampled model aggregation technique based on the bagging algorithm. It is capable of significant run-time reduction with no loss in modeling performance. These claims were validated with a variety of base-learning algorithms on large web and newswire datasets.

In this paper we present a novel method that forms a weighted combination of a range of Stacking based methods for regression problems, without adding any major computational overhead in comparison to stacking itself. The intention of the technique is to benefit from the variation in performance of individual Stacking methods as demonstrated with different data sets, in order to provide a more robust technique overall. We detail an empirical analysis of the technique referred to as weighted Meta–Combiner (wMetaComb) and compare its performance to its underlying techniques.

We present a new ensemble learning method that employs a set of regional classifiers, each of which learns to handle a subset of the training data. We split the training data and generate classifiers for different regions in the feature space. When classifying an instance, we apply a weighted voting scheme among the classifiers that include the instance in their region. We used 11 datasets to compare the performance of our new ensemble method with that of single classifiers as well as other ensemble methods such as RBE, bagging and Adaboost. As a result, we found that the performance of our method is comparable to that of Adaboost and bagging when the base learner is C4.5. In the remaining cases, our method outperformed other approaches.

We present a new co-training style framework and combine it with ensemble learning to further improve the generalization ability. By employing different strategies to combine co-training with ensemble learning, two learning algorithms, Sequential Ensemble Co-Learning (SECL) and Parallel Ensemble Co-Learning (PECL) are developed. Furthermore, we propose a weighted bagging method in PECL to generate an ensemble of diverse classifiers at the end of co-training. Finally, based on the voting margin, an upper bound on the generalization error of multi-classifier voting systems is given in the presence of both classification noise and distribution noise. Experimental results on six datasets show that our method performs better than other compared algorithms.

The vast amount of data available on social media and microblogs can be a valuable resource for mining opinions or for analyzing the overall mood of the public. This helps in identifying potential customers, exploring market trends and predicting events. Analyzing twitter data is comparatively difficult due to the large amount of irregularities present in tweets. Many approaches that use sentiment dictionaries and machine learning have been proposed until now. In this paper, we present a new feature that is extracted using dependency parsing and an emotion lexicon. This feature, along with n-grams, syntactic n-grams and lexicon-based features, is used to classify the tweets. We also use custom dictionaries to identify slang words, SMS short forms, emoticons and word contractions. The performance of various classification algorithms and ensemble techniques is compared. Our results show that the new feature along with the ensemble framework improves sentiment classification.

In the machine learning field, the technique known as ensemble learning aims at combining different base learners in order to increase the quality and the robustness of the predictions. Indeed, this approach has widely been applied to tackle, with success, real world problems from different domains, including computational biology. Nevertheless, despite their potential, ensembles combining results from different base learners have been understudied in the context of gene regulatory network inference. In this paper we applied genetic algorithms and frequent itemset mining, to design small but effective ensembles of gene regulatory network inference methods. These ensembles were evaluated and compared to well-established single and ensemble methods, on both real and synthetic datasets. Results showed that small ensembles, consisting of few but diverse base learners, enhance the exploration of the solution space, and compensate base learners biases, outperforming state-of-the-art methods. Results advocate for the use of such methods as gene regulatory network inference tools.

In modern times, system energy load forecasting is an extremely important process in a variety of contexts. Moreover, energy load time series fluctuations are influenced by a wide range of factors, ranging, inter alia, from environmental conditions, natural events, and demographics to both regional and global geopolitical contexts, economic conditions, energy sources, policies, and regulations. Given these, this paper examines the integration of news information from the global scene into Greek energy load forecasting schemes through the use of sentiment analysis. Investigating the ways the general emotional footprint of news worldwide affects and can be used in an energy modeling context, we benchmark possible ensemble configurations incorporating a multitude of 31 emotion polarities. Building on our previous work, an ensemble method that exploits specific outputs from a multi-label sentiment classifier and a sentiment analysis procedure under a multivariate forecasting scheme is presented. It is shown, through an empirical evaluation of the results, that the integration of emotion representations related to non-Greek news concerning global current affairs improves the predictions.

Dermatograms are pivotal in the early detection of skin cancer, a disease with significant mortality rates. This paper introduces a novel feature extraction method that captures irregularities in the boundaries of abnormal skin regions. Each raw dermatogram is converted into a binary mask image using an effective segmentation algorithm. The boundary of the lesion region is extracted from the mask. The boundary, together with the centroid of the lesion mask, is used to define a set of directional vectors. An Arc is defined using these directional vectors, and a new Arc feature is calculated based on the number of times the lesion boundary crosses the arc. The proposed Arc feature is evaluated using three standard skin lesion datasets: ISBI 2016, HAM10000, and PH2. Additionally, color features and Local Binary Pattern (LBP) features are implemented for comparison. Classical machine learning algorithms are employed to evaluate these features. Results indicate that for the ISBI 2016 and HAM10000 datasets, the Arc feature set demonstrates superior classification accuracy. In contrast, the PH2 dataset benefits more from the LBP feature. Comparative analysis with recent studies highlights the dependency of accuracy on datasets and classifiers, underscoring the necessity for models incorporating feature fusion and ensemble classifiers. The proposed method outperforms traditional color and texture features and shows competitive results against deep learning models, particularly in scenarios with limited computational resources. These findings suggest that the Arc feature is a promising approach for improving skin cancer detection, although further investigation is needed to fine-tune performance, optimize classifier selection, and explore feature fusion strategies.

Steam coal is the blood of China industry. Forecasting steam coal prices accurately and reliably is of great significance to the stable development of China’s economy. For the predictive model of existing steam coal prices, it is difficult to dig the law of nonlinearity of power coal price data and with poor stability. To address the problems that steam coal price features are highly nonlinear and models lack robustness, Laplacian kernel–log hyperbolic loss–Ridge regression (LK-LC-Ridge-Ensemble) model is proposed, which uses ensemble learning model for steam coal price prediction. First, in each sliding window, two kinds of correlation coefficient are employed to identify the optimal time interval, while the optimal feature set is selected to reduce the data dimension. Second, the Laplace kernel functions are adopted for constructing kernel Ridge regression (LK-Ridge), which boosts the capacity to learn nonlinear laws; the logarithmic loss function is introduced to form the LK-LC-Ridge to enhance the robustness. Finally, the prediction results of each single regression models are utilized to build a results matrix that is input into the meta-model SVR for ensemble learning, which further develops the model performance. Empirical results from three typical steam coal price datasets indicate that the proposed ensemble strategy is reliable for the model performance enhancement. Furthermore, the proposed model outperforms all single primitive models including accuracy of prediction results and robustness of model. Grouping cross-comparison between the different models suggests that the proposed ensemble model is more accurate and robust for steam coal price forecasting.

Advances in Unmanned Aerial Vehicles (UAVs), otherwise recognized as drones, have tremendous promise in improving the wide-ranging applications of the Internet of Things (IoT). UAV image classification using deep learning (DL) is an amalgamation to modernize data analysis, collection, and decision-making in a variety of sectors. IoT devices collect information in real time, while remote sensing captures data afar without direct contact. UAVs equipped with sensors offer high-quality images for classification tasks. DL techniques, especially the convolutional neural networks (CNNs), analyze data streams, extracting complicated features for the accurate classification of objects or environmental features. This synergy enables applications including urban planning and precision agriculture, fostering smarter disaster response, decision support systems, and efficient resource management. This paper introduces a novel Pyramid Channel-based Feature Attention Network with an Ensemble Learning-based UAV Image Classification (PCFAN-ELUAVIC) technique in an assisted remote sensing environment. The PCFAN-ELUAVIC technique begins with the contrast enhancement of the UAV images using the CLAHE technique. Following that, the feature vectors are derived by the use of the PCFAN model. Meanwhile, the hyperparameter tuning procedure is executed by the inclusion of a vortex search algorithm (VSA). For image classification, the PCFAN-ELUAVIC technique comprises an ensemble of three classifiers like long short-term memory (LSTM), graph convolutional networks (GCNs), and Hermite neural network (HNN). To exhibit the improved detection results of the PCFAN-ELUAVIC system, an extensive range of experiments are carried out. The experimental values confirmed the enhanced performance of the PCFAN-ELUAVIC model when compared to other techniques.

Data stream learning in non-stationary environments and skewed class distributions has been receiving more attention in machine learning communities. This paper proposes a novel ensemble classification method (ECSDS) for classifying data streams with skewed class distributions. In the proposed ensemble method, back-propagation neural network is selected as the base classifier. In order to demonstrate the effectiveness of our proposed method, we choose three baseline methods based on ECSDS and evaluate their overall performance on ten datasets from UCI machine learning repository. Moreover, the performance of incremental learning is also evaluated by these datasets. The experimental results show our proposed method can effectively deal with classification problems on non-stationary data streams with class imbalance.

Classification is one of the most important problems in data mining and machine learning. The quality and quantity of classification rules are two factors to influence the accuracy of classification. In this paper, we propose a new algorithm to enhance the final classification accuracy, called CMCM (Classification based on Multiple Classifier Models), which consists of two classification models. Model1 centers on the improvement of quality. The optimal attribute values are obtained as the first item of a classification rule from both the items and their complements. While in Model2, quantity is taken into consideration, so it constructs two candidate sets and uses the one-versus-many strategy to generate several rules at one time. The experiment results demonstrate that CMCM can achieve higher classification accuracy than the proposed classification approaches. CMCM can extract sufficient high-quality rules for imbalanced data. Meanwhile, it can also obtain sufficient latent information for classification.

Neuro-fuzzy techniques have been widely used in many applications due to their ability to generate interpretable fuzzy rules. Ensemble learning, on the other hand, is an emerging paradigm in artificial intelligence used to improve performance results by combining multiple single learners. This paper aims to develop and evaluate a set of homogeneous ensembles over four medical datasets using hyperparameter tuning of four neuro-fuzzy systems: adaptive neuro-fuzzy inference system (ANFIS), Dynamic evolving neuro-fuzzy system (DENFIS), Hybrid fuzzy inference system (HyFIS), and neuro-fuzzy classifier (NEFCLASS). To address the interpretability challenges and to reduce the complexity of high-dimensional data, the information gain filter was used to identify the most relevant features. After that, the performance of the neuro-fuzzy single learners and ensembles was evaluated using four performance metrics: accuracy, precision, recall, and f1 score. To decide which single learners/ensembles perform better, the Scott-Knott and Borda count techniques were used. The Scott-Knott first groups the models based on the accuracy to find the classifiers appearing in the best cluster, while the Borda count ranks the models based on all the four performance metrics without favoring any of the metrics. Results showed that: (1) The number of the combined single learners positively impacts the performance of the ensembles, (2) Single neuro-fuzzy classifiers demonstrate better or similar performance to the ensembles, but the ensembles still provide better stability of predictions, and (3) Among the ensembles of different models, ANFIS provided the best ensemble results.

To accurately and rapidly predict seismic responses, including the maximum displacement (MaxD) and maximum acceleration (MaxA), of the isolated structure considering the soil–structure interaction (SSI), five ensemble learning models, i.e. random forest (RF), gradient boosting regression tree (GBRT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM) and stacking model, are constructed. Firstly, a total of 96 000 nonlinear time history analyses of the isolated structure considering the SSI are conducted with the aid of OpenSees. The generated database is used for training and testing ensemble learning models. The ensemble learning models have 12 input variables in four categories, i.e. ground motion parameters, structural parameter, isolation parameters and soil parameter, and two output variables, i.e. MaxD and MaxA. The study shows that all ensemble learning models have excellent prediction performance for both training and testing datasets. The determination coefficients are larger than 0.96 and root-mean-square errors (RMSEs) are relatively small. Among the five ensemble learning models, the stacking model exhibits the best performance. In addition, the calculation method of feature importance score for the stacking model is provided. According to the feature importance analysis, the ground motion parameters have greater impact on seismic responses than other three categories of inputs. Finally, six ground motions are randomly selected to verify the generalization ability of the proposed ensemble learning models. The results show that the stacking model has a favorable generalization ability with relatively small prediction errors.

Background: Coronary artery disease (CAD) is one of the most representative cardiovascular diseases. Early and accurate prediction of CAD based on physiological measurements can reduce the risk of heart attack through medicine therapy, healthy diet, and regular physical activity. Methods:Four heart disease datasets from the UC Irvine Machine Learning Repository were combined and re-examined to remove incomplete entries, and a total of 822 cases were utilized in this study. Seven machine learning methods, including Naïve Bayes, artificial neural networks (ANNs), sequential minimal optimization (SMO), k-nearest neighbor (KNN), AdaBoost, J48, and random forest, were adopted to analyze the collected datasets for CAD prediction. By combining co-expressed observations and an ensemble voting mechanism, we designed and evaluated a new medical decision classifier for CAD prediction. The TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) algorithm was applied to determine the best prediction method for CAD diagnosis. Results: Features of systolic blood pressure, cholesterol, heart rate, and ST depression are considered to be the most significant differences between patients with and without CADs. We show that the prediction capability of seven machine learning classifiers can be enhanced by integrating combinations of observed co-expressed features. Finally, compared to the use of any single classifier, the proposed voting mechanism achieved optimal performance according to TOPSIS.