LogitBoost Algorithm Research Articles

Multi-Branch Line Fault Arc Detection Method Based on the Improved Northern Goshawk Optimization Adaptive Base Class LogitBoost Algorithm

In low-voltage AC distribution systems, when a series arc fault occurs in a branch with multiple loads operating in parallel, it will be significantly more difficult to identify. Existing arc fault detection methods make it difficult to effectively detect faults occurring in the lower-level branch. This study introduces a novel series arc fault detection approach based on the improved northern goshawk optimization adaptive base class LogitBoost (INGO-ABCLogitBoost) algorithm. Considering the zero-rest, intermittent, and random fluctuation and high-frequency features of the arc current, the zero-rest coefficient, discrete coefficient, harmonic amplitude, and wavelet entropy are proposed to establish the high-dimensional feature matrix of the arc current. The ReliefF feature selection algorithm is used to optimize feature quality and decrease feature dimensionality. Subsequently, the ABCLogitBoost fault detection model is proposed, with the INGO algorithm applied to optimize the model parameters, thus enhancing the model’s diagnostic capabilities. The efficacy of the proposed diagnostic model is validated through the construction of a multi-load arc simulation system. The simulation results show that the overall fault diagnosis accuracy of the proposed method reaches 99.01% and can effectively identify the fault load types, which helps to locate the fault location.

Comparative Study of Anemia Classification Algorithms for International and Newly CBC Datasets

Data generated from modern applications and the internet in healthcare is extensive and rapidly expanding. Therefore, one of the significant success factors for any application is understanding and extracting meaningful information using digital analytics tools. These tools will positively impact the application's performance and handle the challenges that can be faced to create highly consistent, logical, and information-rich summaries. This paper contains three main objectives: First, it provides several analytics methodologies that help to analyze datasets and extract useful information from them as preprocessing steps in any classification model to determine the dataset characteristics. Also, this paper provides a comparative study of several classification algorithms by testing 12 different classifiers using two international datasets to provide an accurate indicator of their efficiency and the future possibility of combining efficient algorithms to achieve better results. Finally, building several CBC datasets for the first time in Iraq helps to detect blood diseases from different hospitals. The outcome of the analysis step is used to help researchers to select the best system structure according to the characteristics of each dataset for more organized and thorough results. Also, according to the test results, four algorithms achieved the best accuracy (Logitboost, Random Forest, XGBoost, Multilayer Perceptron). Then use the Logitboost algorithm that achieved the best accuracy to classify these new datasets. In addition, as future directions, this paper helps to investigate the possibility of combining the algorithms to utilize benefits and overcome their disadvantages.

DETECTION OF ALZHEIMER'S DISEASE FROM ELECTROENCEPHALOGRAPHY (EEG) SIGNALS USING MULTITAPER AND ENSEMBLE LEARNING METHODS

Alzheimer's disease is a complex brain disease and is also the most common form of dementia that leads to impaired social and intellectual abilities. The disease only manifests itself with a simple forgetfulness, as the disease progresses, the patient forgets the recent events, cannot recognize his family members and close environment, and becomes in need of care in the last stage. Early detection is therefore crucial for medical intervention to prevent brain injury and prolong everyday functioning. In this study is aimed to detection of Alzheimer’s disease from EEG signals using the multitaper and ensemble learning methods. The dataset comprises of 24 healthy people and 24 Alzheimer's patients' EEG signals. 49 features were extracted by calculating the power spectral density (PSD) of the frequencies of the EEG signals between 1-49 Hz using the multitaper method. Then, the performances of AdaboostM1, Total Boost, Gentle Boost, Logit Boost, Robust Boost, and Bagging ensemble learning algorithms were compared. As a result of experiments, the Logit Boost algorithm has the highest performance. The algorithm has achieved a promising performance of 93.04% accuracy, 93.09% f1-score, 92.75% sensitivity, 93.43% precision, and 93.33% specificity.

Classification of hemiplegia through gait analysis and machine learning methods

Objective: Gait analysis is a method that is used for understanding normal walking and determining the stage of the disease as it affects
 walking. It is important to objectively determine the stage of the disease in order to decide interventions and treatment strategies. This
 study aims to determine the Brunnstrom Stage of the hemiplegic patients with an analysis of gait data.
 Patients and Methods: In the first part of the study, the gait signal data were taken from 28 post-stroke hemiplegic patients and
 7 healthy individuals with three-axis accelerometers. In the second part, new gait data were collected from 15 healthy individuals
 through an accelerometer on the anteroposterior axis.
 First the accelerometer signals were decomposed to Daubechies 5 (Db5) level six wavelets using MATLAB software. Subsequently,
 these attributes were classified through several classifier and machine learning algorithms on WEKA and MATLAB software packages
 to predict the stages of hemiplegia.
 Results: The highest accuracy rate in the prediction of hemiplegia stage was achieved with the LogitBoost algorithm on WEKA with
 91% for 35 samples, and 90% for 50 samples. This performance was followed by the RUSBoosted Trees algorithm on the MATLAB
 software with an accuracy of 86.1% correct prediction.
 Conclusion: The Brunnstrom Stage of hemiplegia can be predicted with machine learning algorithms with a good accuracy, helping
 physicians to classify hemiplegic patients into correct stages, monitor and manage their rehabilitation.

Automatic Recognition of Illegal Substations by Employing Logit-Boost Algorithm and LSTM With the Help of Different Landsat-8 OLI Image Spectral Band Parameters: A Case Study in Sason, Turkey

Automatic Recognition of Illegal Substations by Employing Logit-Boost Algorithm and LSTM With the Help of Different Landsat-8 OLI Image Spectral Band Parameters: A Case Study in Sason, Turkey

Stability Risk Assessment of Underground Rock Pillars Using Logistic Model Trees.

Pillars are important structural elements that provide temporary or permanent support in underground spaces. Unstable pillars can result in rock sloughing leading to roof collapse, and they can also cause rock burst. Hence, the prediction of underground pillar stability is important. This paper presents a novel application of Logistic Model Trees (LMT) to predict underground pillar stability. Seven parameters—pillar width, pillar height, ratio of pillar width to height, uniaxial compressive strength of rock, average pillar stress, underground depth, and Bord width—are employed to construct LMTs for rock and coal pillars. The LogitBoost algorithm is applied to train on two data sets of rock and coal pillar case histories. The two models are validated with (i) 10-fold cross-validation and with (ii) another set of new case histories. Results suggest that the accuracy of the proposed LMT is the highest among other common machine learning methods previously employed in the literature. Moreover, a sensitivity analysis indicates that the average stress, p, and the ratio of pillar width to height, r, are the most influential parameters for the proposed models.

Automatic Recognition of Illegal Substations by Employing Logit-Boost Algorithm and Different Landsat-8 Oli Image Spectral Band Parameters: A Case Study in Sason, Turkey

Automatic Recognition of Illegal Substations by Employing Logit-Boost Algorithm and Different Landsat-8 Oli Image Spectral Band Parameters: A Case Study in Sason, Turkey

K-Mean Clustering Analysis and Logistic Boosting Regression for Rock Facies Characterization and Classification in Zubair Reservoir in Luhais Oil Field, Southern Iraq

Identifying rock facies from petrophysical logs is a crucial step in the evaluation and characterization of hydrocarbon reservoirs. The rock facies can be obtained either from core analysis (lithofacies) or from well logging data (electrofacies). In this research, two advanced machine learning approaches were adopted for electrofacies identification and for lithofacies classification, both given the well-logging interpretations from a well in the upper shale member in Luhais Oil Field, southern Iraq. Specifically, the K-mean partitioning analysis and Logistic Boosting (Logit Boost) were conducted for electrofacies characterization and lithofacies classification, respectively. The dataset includes the routine core analysis of core porosity, core permeability, and measured discrete lithofacies along with the well-logging interpretations include (shale volume, water saturation and effective porosity) given the entire reservoir interval. The K-Mean clustering technique demonstrated good matching between the vertical sequence of identified electrofacies and the observed lithofacies from core description as attained 89.92% total correct percent from the confusion matrix table. The Logit Boost showed excellent matching between the recognized lithofacies from the core description and the predicted lithofacies through attained 98.26% total correct classification rate index from the confusion matrix table. The high accuracy of the Logit Boost algorithm comes from taking into account the non-linearity between the lithofacies and petrophysical properties in the classification process. The high degree of lithofacies classification by Logit Boost in this research can be considered in a similar procedure at all sandstone reservoirs to improve the reservoir characterization. The complete facies identification and classification were implemented with the programming language R, the powerful open-source statistical computing language.

Fluorescence and Reflectance Spectroscopy for Early Detection of Different Mycorrhized Plantain Plants

<p class="1Body">Sustainable agriculture with use of Arbuscular Mycorrhizal Fungi (AMF) is an emerging farm management that improves crops nutrient and water use efficiency. Decision making on the effect of AMF is still dependent on agronomic diagnosis which is long, tedious, expensive and destructive. This study demonstrates the applicability of proximal fluorescence and reflectance spectroscopy for evaluating and detecting at early stage distinct types of mycorrhized plantain from two cultivars (<em>Musa paradisiaca</em>).</p><p class="1Body">Visible-near infrared (400-1000 nm) reflectance and fluorescence data were collected from control and three levels mycorrhized plants designed in randomized and complete block under greenhouse conditions. Two spectral measurements at a week interval were performed on plant leaves by using an USB spectrometer mounted with an Arduino-based LED driver clip.</p>A new normalized reflectance water NWI5 index shows with Datt5 alone highly significant differences at P<0.001 respectively for Orishele and fhia21 cultivars. dNIRmin920_980, NDVI3 and GI reflectance index are significant at P<0.01. Seven other reflectance and 3 fluorescence indices ANTH, FRF_R and NBI_R are significant at P<0.05. The two first principal components for each cultivar spectral features explaining 94.1 % of variance were used to build predictive classification models. LogitBoost algorithm indicates accuracy of 90.27% on stratified cross-validation and 87.5% on test split. Our results confirm that fluorescence and reflectance spectroscopy is a valuable tool for early assessment of mycorrhization success rate evaluation and pattern recognition. They also show promise for the development of non-destructive and cost-effective detectors in monitoring crops under biofertilizers with arbuscular mycorrhizae.

Ensemble Learning in Data Mining of Fetal Cardiotocograms

ReliefF feature selection and LogitBoost ensemble learning method are employed in the data mining procedure of 2126 fetal cardiotocograms (CTGs). Based on 10 critical features selected by ReliefF and the full 21 features, LogitBoost algorithm almost outperforms the other three ensemble learning methods of Stacking, Bagging and AdaBoostM1 in ACC (%) and AUC in classification, and the ACC (%) and AUC of LogitBoost algorithm are achieved to 94.45% and 0.977 based on the critical features from ReliefF.

An improved multiclass LogitBoost using adaptive-one-vs-one

LogitBoost is a popular Boosting variant that can be applied to either binary or multi-class classification. From a statistical viewpoint LogitBoost can be seen as additive tree regression by minimizing the Logistic loss. Following this setting, it is still non-trivial to devise a sound multi-class LogitBoost compared with to devise its binary counterpart. The difficulties are due to two important factors arising in multiclass Logistic loss. The first is the invariant property implied by the Logistic loss, causing the optimal classifier output being not unique, i.e. adding a constant to each component of the output vector won't change the loss value. The second is the density of the Hessian matrices that arise when computing tree node split gain and node value fittings. Oversimplification of this learning problem can lead to degraded performance. For example, the original LogitBoost algorithm is outperformed by ABC-LogitBoost thanks to the latter's more careful treatment of the above two factors. In this paper we propose new techniques to address the two main difficulties in multiclass LogitBoost setting: (1) we adopt a vector tree model (i.e. each node value is vector) where the unique classifier output is guaranteed by adding a sum-to-zero constraint, and (2) we use an adaptive block coordinate descent that exploits the dense Hessian when computing tree split gain and node values. Higher classification accuracy and faster convergence rates are observed for a range of public data sets when compared to both the original and the ABC-LogitBoost implementations. We also discuss another possibility to cope with LogitBoost's dense Hessian matrix. We derive a loss similar to the multi-class Logistic loss but which guarantees a diagonal Hessian matrix. While this makes the optimization (by Newton descent) easier we unfortunately observe degraded performance for this modification. We argue that working with the dense Hessian is likely unavoidable, therefore making techniques like those proposed in this paper necessary for efficient implementations.

Interpretation of Social Touch on an Artificial Arm Covered with an EIT-based Sensitive Skin

During social interaction humans extract important information from tactile stimuli that improves their understanding of the interaction. The development of a similar capacity in a robot will contribute to the future success of intuitive human–robot interactions. This paper presents experiments on the classification of social touch on a full-sized mannequin arm covered with touch-sensitive artificial skin. The flexible and stretchable sensitive skin was implemented using electrical impedance tomography. A classifier based on the LogitBoost algorithm was used to classify six emotions and six social messages transmitted by humans when touching the artificial arm. Experimental results show that classification of social touch can be achieved with accuracies comparable to those achieved by humans.

Applying additive logistic regression to data derived from sensors monitoring behavioral and physiological characteristics of dairy cows to detect lameness

The hypothesis was that sensors currently available on farm that monitor behavioral and physiological characteristics have potential for the detection of lameness in dairy cows. This was tested by applying additive logistic regression to variables derived from sensor data. Data were collected between November 2010 and June 2012 on 5 commercial pasture-based dairy farms. Sensor data from weigh scales (liveweight), pedometers (activity), and milk meters (milking order, unadjusted and adjusted milk yield in the first 2min of milking, total milk yield, and milking duration) were collected at every milking from 4,904 cows. Lameness events were recorded by farmers who were trained in detecting lameness before the study commenced. A total of 318 lameness events affecting 292 cows were available for statistical analyses. For each lameness event, the lame cow’s sensor data for a time period of 14d before observation date were randomly matched by farm and date to 10 healthy cows (i.e., cows that were not lame and had no other health event recorded for the matched time period). Sensor data relating to the 14-d time periods were used for developing univariable (using one source of sensor data) and multivariable (using multiple sources of sensor data) models. Model development involved the use of additive logistic regression by applying the LogitBoost algorithm with a regression tree as base learner. The model’s output was a probability estimate for lameness, given the sensor data collected during the 14-d time period. Models were validated using leave-one-farm-out cross-validation and, as a result of this validation, each cow in the data set (318 lame and 3,180 nonlame cows) received a probability estimate for lameness. Based on the area under the curve (AUC), results indicated that univariable models had low predictive potential, with the highest AUC values found for liveweight (AUC=0.66), activity (AUC=0.60), and milking order (AUC=0.65). Combining these 3 sensors improved AUC to 0.74. Detection performance of this combined model varied between farms but it consistently and significantly outperformed univariable models across farms at a fixed specificity of 80%. Still, detection performance was not high enough to be implemented in practice on large, pasture-based dairy farms. Future research may improve performance by developing variables based on sensor data of liveweight, activity, and milking order, but that better describe changes in sensor data patterns when cows go lame.

Spam detection using Random Boost

This paper proposes two alternative methods of random projections and compares their performance for robust and efficient spam detection when trained using a small number of examples. Robustness refers to learning and adaptation leading to a high level of performance despite data variability, while efficiency is concerned with (i) the complexity of the detection method employed; and (ii) the amount of training resources used for training and retraining. The first method, Random Project, employs a random projection matrix to produce linear combinations of input features, while the second method, Random Boost, employs random feature selection to enhance the performance of the Logit Boost algorithm. Random Boost is, in fact, a combination of Logit Boost and Random Forest. Experimental results, using TREC and CEAS as challenging spam benchmark sets, show that the Random Boost method significantly improves the performance of the spam filter compared to the Logit Boost algorithm (e.g., a 5% increase in AUC, which is the area under the Receiver Operating Characteristic curve), and yields similar classification accuracy compared to the Random Forest method but using only one fourth the runtime complexity of the Random Forest algorithm. Additionally, the Random Boost algorithm also reduces training time by two orders of magnitude compared to Logit Boost, which becomes important during retraining on the ever changing data streams, including adapting to adversarial tactics and “noise” injected by spammers.

Stratified Normalization LogitBoost for Two-Class Unbalanced Data Classification

The research on unbalanced data classification is a hot topic in recent years. LogitBoost algorithm is an adaptive algorithm that can get much higher prediction precision. But in the face of unbalanced data, this algorithm could produce a large minority class prediction error. In this article, we propose an improved LogitBoost algorithm named BLogitBoost, based on a stratified normalization method which normalizes within class sampling probability first, then normalizes between classes. The experiments on simulation data and empirical data show that the new algorithm can reduce the minority class prediction error significantly.

Development and validation of a novel molecular biomarker diagnostic test for the early detection of sepsis

IntroductionSepsis is a complex immunological response to infection characterized by early hyper-inflammation followed by severe and protracted immunosuppression, suggesting that a multi-marker approach has the greatest clinical utility for early detection, within a clinical environment focused on Systemic Inflammatory Response Syndrome (SIRS) differentiation. Pre-clinical research using an equine sepsis model identified a panel of gene expression biomarkers that define the early aberrant immune activation. Thus, the primary objective was to apply these gene expression biomarkers to distinguish patients with sepsis from those who had undergone major open surgery and had clinical outcomes consistent with systemic inflammation due to physical trauma and wound healing.MethodsThis was a multi-centre, prospective clinical trial conducted across four tertiary critical care settings in Australia. Sepsis patients were recruited if they met the 1992 Consensus Statement criteria and had clinical evidence of systemic infection based on microbiology diagnoses (n = 27). Participants in the post-surgical (PS) group were recruited pre-operatively and blood samples collected within 24 hours following surgery (n = 38). Healthy controls (HC) included hospital staff with no known concurrent illnesses (n = 20). Each participant had minimally 5 ml of PAXgene blood collected for leucocyte RNA isolation and gene expression analyses. Affymetrix array and multiplex tandem (MT)-PCR studies were conducted to evaluate transcriptional profiles in circulating white blood cells applying a set of 42 molecular markers that had been identified a priori. A LogitBoost algorithm was used to create a machine learning diagnostic rule to predict sepsis outcomes.ResultsBased on preliminary microarray analyses comparing HC and sepsis groups, a panel of 42-gene expression markers were identified that represented key innate and adaptive immune function, cell cycling, WBC differentiation, extracellular remodelling and immune modulation pathways. Comparisons against GEO data confirmed the definitive separation of the sepsis cohort. Quantitative PCR results suggest the capacity for this test to differentiate severe systemic inflammation from HC is 92%. The area under the curve (AUC) receiver operator characteristics (ROC) curve findings demonstrated sepsis prediction within a mixed inflammatory population, was between 86 and 92%.ConclusionsThis novel molecular biomarker test has a clinically relevant sensitivity and specificity profile, and has the capacity for early detection of sepsis via the monitoring of critical care patients.

LogitBoost classifier for discriminating thermophilic and mesophilic proteins

A novel classifier, the so-called LogitBoost classifier, was introduced to discriminate the thermophilic and mesophilic proteins according to their primary structures. When the 20-amino acid composition was chosen as the feature vector, the overall accuracy of the self-consistency check and a five-fold cross-validation procedure was 97.0% and 86.6%, respectively. To test if the method was also applicable to a wide range of biological targets, an independent testing dataset was also used. The method based on LogitBoost algorithm has achieved an overall classification accuracy of 88.9%. According to the three different validation check approaches, it was demonstrated that LogitBoost outperformed AdaBoost and performed comparably with RBF neural network and support vector machine. The influence of protein size on discrimination was addressed.

Induction of descriptive fuzzy classifiers with the Logitboost algorithm

Recently, Adaboost has been compared to greedy backfitting of extended additive models in logistic regression problems, or “Logitboost". The Adaboost algorithm has been applied to learn fuzzy rules in classification problems, and other backfitting algorithms to learn fuzzy rules in modeling problems but, up to our knowledge, there are not previous works that extend the Logitboost algorithm to learn fuzzy rules in classification problems.In this work, Logitboost is applied to learn fuzzy rules in classification problems, and its results are compared with that of Adaboost and other fuzzy rule learning algorithms. Contradicting the expected results, it is shown that the basic extension of the backfitting algorithm to learn classification rules may produce worse results than Adaboost does. We suggest that this is caused by the stricter requirements that Logitboost demands to the weak learners, which are not fulfilled by fuzzy rules. Finally, it is proposed a prefitting based modification of the Logitboost algorithm that avoids this problem