Decision Tree Neural Network Research Articles

GSpyNetTree: a signal-vs-glitch classifier for gravitational-wave event candidates

Despite achieving sensitivities capable of detecting the extremely small amplitude of gravitational waves (GWs), LIGO and Virgo detector data contain frequent bursts of non-Gaussian transient noise, commonly known as ‘glitches’. Glitches come in various time-frequency morphologies, and they are particularly challenging when they mimic the form of real GWs. Given the higher expected event rate in the next observing run (O4), LIGO-Virgo GW event candidate validation will require increased levels of automation. Gravity Spy, a machine learning tool that successfully classified common types of LIGO and Virgo glitches in previous observing runs, has the potential to be restructured as a compact binary coalescence (CBC) signal-vs-glitch classifier to accurately distinguish between glitches and GW signals. A CBC signal-vs-glitch classifier used for automation must be robust and compatible with a broad array of background noise, new sources of glitches, and the likely occurrence of overlapping glitches and GWs. We present GSpyNetTree, the Gravity Spy Convolutional Neural Network Decision Tree: a multi-CNN classifier using CNNs in a decision tree sorted via total GW candidate mass tested under these realistic O4-era scenarios.

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.

Evaluation efficiency of hybrid deep learning algorithms with neural network decision tree and boosting methods for predicting groundwater potential

Delineation of the groundwater’s potential zones is a growing phenomenon worldwide due to the high demand for fresh groundwater. Therefore, the identification of potential groundwater zones is an important tool for groundwater occurrence, protection, and management purposes. More specifically, in arid and semi-arid regions, groundwater is one of the most important natural resources as it supplies water during the drought period. The present research study focused on the delineation of potential groundwater zones in Saveh City, the northern part of the Markazi Province in Iran. The groundwater potential mapping was prepared using hybrid deep learning and machine learning algorithm of the boosted tree (BT), artificial neural network (ANN), deep learning neural network (DLNN), deep learning tree (DLT), and deep boosting (DB). This study was carried out by using fourteen groundwater potential conditioning factors and 349 each for springs and non-springs points. The performance of each model was validated through statistical analysis of sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), and receiver operating characteristic (ROC)-area under curve (AUC) analysis. The validation result showed that the success rate of AUC is very good for the DB model (0.87–0.99) and other models are also good i.e. BT (0.81–0.90), ANN (0.77–0.82), DLNN (0.84–0.86), and DLT (0.83–0.91). Among the several factors used in this study altitude, rainfall, distance to fault and soil types are the more important conditioning factors for groundwater potential modeling. Finally, all the models in this study had high efficiency in groundwater potential mapping, but it is recommended to use the Deep Boost model due to the better results in future studies. The result of this work will be useful to planners for optimal use and future planning of groundwater.

Improved algorithm of decision tree based on neural network

Decision trees have been applied to solve many data mining problems due to their superior learning and classification capabilities, and have achieved good results. However, for dealing with big data and complex model problems, decision trees show insufficient accuracy and overfitting. In order to solve these problems, neural network is introduced as a decision tree node, and an improved algorithm based on neural network decision tree is proposed. In the neural network decision tree, the classifier learning consists of two stages: the first stage uses a heuristic algorithm with reduced uncertainty to divide the big data, and stops the growth of the decision tree until the node dividing ability is below a certain threshold; in the second stage, the neural network is used to classify the decision-making leaf node with generalization ability. The experimental results show that compared with the traditional classification learning algorithm, the algorithm has a higher accuracy rate and it can determine the size of decision tree through structural adaptation for the classification problem of identifying big data and complex patterns.

Prediction of transition from mild cognitive impairment to Alzheimer's disease based on a logistic regression-artificial neural network-decision tree model.

To develop a logistic regression model, artificial neural network (ANN) model and decision tree (DT) model for the progression of mild cognitive impairment (MCI) to Alzheimer's disease (AD) to compare the performance of the three models. A total of 425 patients with MCI were screened from the original cohort. The actual follow up included 361 patients, with AD as the outcome variable. Three kinds of prediction models were developed: a logistic regression model, ANN model and DT model. The performance of all three models was measured with accuracy, sensitivity, positive predictive value and area under the receiver operating characteristic curve. A total of 121 patients with MCI developed AD, and the average conversion rate was 9.49% per year. The ANN model had higher accuracy (89.52 ± 0.36%), area under the receiver operating characteristic curve (92.08 ± 0.12), sensitivity (82.11 ± 0.42%) and positive predictive value (75.26 ± 0.86%) than the other two models. The first five important predictors of the ANN model were, in order, ADL score, age, urine AD-associated neuronal thread protein, alcohol consumption and smoking. For the DT model, they were age, activities of daily living score, family history of dementia, urine AD-associated neuronal thread protein and alcohol consumption. For the logistic regression model, they were age, sex, activities of daily living score, alcohol consumption and smoking. The logistic regression, ANN and DT models performed well at predicting the transition from MCI to AD with ideal stability. However, the ANN model had the best predictive value. Increased age, activities of daily living score, urine AD-associated neuronal thread protein, alcohol consumption, smoking and sex were important factors. Geriatr Gerontol Int 2021; 21: 43-47.

Dynamic game pricing model of car-free carrier platform based on GA-BP theory

With the rapid development of China’s logistics industry, transport demand continues to be strong, and how to scientifically give platform pricing to achieve rapid transactions and reduce operating costs is a major problem. Firstly, in order to solve this problem, after collecting data from the car-free carrier company P, due to the excessive influence factors of the data, it is necessary to clean the data and normalize the data. Six indexes that affect the pricing are screened out by the grey correlation model, which are total mileage, total line cost, vehicle length, vehicle tonnage, transportation grade and the time needed for plan execution. In order to verify the rationality, we do multiple linear regression to the data. Secondly, in order to solve the nonlinear fitting problem, the improved initial value GA-BP model is introduced to perform data fitting and prediction, it is concluded that the first pricing is to achieve the required accuracy at n=94, and the error is less than 0.1% when the number of iterations no more than 100, which verifies the model is suitable for such problems. Lastly, In order to simulate the dynamic pricing process of the platform, it is required to predict three times of pricing. Based on the first pricing and known data, a neural network decision tree prediction model based on game theory is established. The dichotomy is applied to the value of information entropy to measure the proportion of price adjustment, and a conclusion is drawn. When the information entropy value I>26.7204, the second price adjustment decreases by 3.87%, and the third price adjustment decreases by 0.39% .When the information entropy value I>26.7204, the second price adjustment decreases by 1.64%, and the third price adjustment decreases by 1.27%.

A Novel Intelligent System for Brain Tumor Diagnosis Based on a Composite Neutrosophic-Slantlet Transform Domain for Statistical Texture Feature Extraction.

Discrete wavelet transform (DWT) is often implemented by an iterative filter bank; hence, a lake of optimization of a discrete time basis is observed with respect to time localization for a constant number of zero moments. This paper discusses and presents an improved form of DWT for feature extraction, called Slantlet transform (SLT) along with neutrosophy, a generalization of fuzzy logic, which is a relatively new logic. Thus, a novel composite NS-SLT model has been suggested as a source to derive statistical texture features that used to identify the malignancy of brain tumor. The MR images in the neutrosophic domain are defined using three membership sets, true (T), false (F), and indeterminate (I); then, SLT was applied to each membership set. Three statistical measurement-based methods are used to extract texture features from images of brain MRI. One-way ANOVA has been applied as a method of reducing the number of extracted features for the classifiers; then, the extracted features are subsequently provided to the four neural network classification techniques, Support Vector Machine Neural Network (SVM-NN), Decision Tree Neural Network (DT-NN), K-Nearest Neighbor Neural Network (KNN-NN), and Naive Bayes Neural Networks (NB-NN), to predict the type of the brain tumor. Meanwhile, the performance of the proposed model is assessed by calculating average accuracy, precision, sensitivity, specificity, and Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve. The experimental results demonstrate that the proposed approach is quite accurate and efficient for diagnosing brain tumors when the Gray Level Run Length Matrix (GLRLM) features derived from the composite NS-SLT technique is used.

Bayesian regularized neural network decision tree ensemble model for genomic data classification

ABSTRACTMachine learning techniques have been widely applied to solve the classification problem of highly dimensional and complex data in the field of bioinformatics. Among them, Bayesian regularized neural network (BRNN) became one of the popular choices due to its robustness and ability to avoid over fitting. On the other hand, Bayesian approach applied to neural network training offers computational burden and increases its time complexity. This restricts the use of BRNN in an on-line machine learning system. In this article, a Bayesian regularized neural network decision Tree (BrNdT) ensemble model, is proposed to combat high computational time complexity of a classifier model. The key idea behind the proposed ensemble methodology is to weigh and combine several individual classifiers and apply majority voting decision scheme to obtain an efficient classifier which outperforms each one of them. The simulation results show that the proposed method achieves a significant reduction in time complexity and maintains high accuracy over other conventional techniques.

Arabic script web page language identifications using decision tree neural networks

In this paper, we propose a hybrid approach of Arabic scripts web page language identification based on decision tree and ARTMAP approaches. We use the decision tree approach to find the general identities of a web document, be it an Arabic script-based or a non-Arabic-based. Then, we use the selected representations of identified pages from the decision tree approach as an input to the ARTMAP neural network for further verification of the diversity of languages detected by the algorithm. From our initial experiments, we found that, although the decision tree approach may achieve a higher accuracy than ARTMAP, the former may not be as reliable as the ARTMAP approach if the language used is extended to other types of Arabic script web documents in different languages (e.g., Urdu, Arabic, Persian, etc.). Therefore, we propose this hybrid decision tree-ARTMAP approach in order to improve the performance of the Arabic script language identification on web documents in a variety of languages. The result shows that the proposed approach has outperformed both decision tree and the default ARTMAP approaches.

ANN-DT: an algorithm for extraction of decision trees from artificial neural networks

Although artificial neural networks can represent a variety of complex systems with a high degree of accuracy, these connectionist models are difficult to interpret. This significantly limits the applicability of neural networks in practice, especially where a premium is placed on the comprehensibility or reliability of systems. A novel artificial neural-network decision tree algorithm (ANN-DT) is therefore proposed, which extracts binary decision trees from a trained neural network. The ANN-DT algorithm uses the neural network to generate outputs for samples interpolated from the training data set. In contrast to existing techniques, ANN-DT can extract rules from feedforward neural networks with continuous outputs. These rules are extracted from the neural network without making assumptions about the internal structure of the neural network or the features of the data. A novel attribute selection criterion based on a significance analysis of the variables on the neural-network output is examined. It is shown to have significant benefits in certain cases when compared with the standard criteria of minimum weighted variance over the branches. In three case studies the ANN-DT algorithm compared favorably with CART, a standard decision tree algorithm.