Neural Network Tree Research Articles

Comparative study on peak power prediction methods during start-up and power-up of heat pipe reactor based on neural network and decision tree

The effective control of start-up and power-up processes is essential if thermal power reactors are operated autonomously. The decision-making processes of an autonomous operating system must anticipate and suppress unwanted power fluctuations in a timely manner. The rapid and accurate predictions of the peak power to be expected in a surge can help an autonomous operating system to make more appropriate and timely decisions. This paper compares the predictive performance of the neural network with that of the decision tree, where both are based on peak power datasets for start-up and power-up processes. We explain how the datasets are constructed, i.e., by coupling the Monte Carlo sampling methods with the analysis of the system analysis procedure. We introduce a framework for the comprehensive evaluation of both prediction algorithms. We conclude that the neural network has better intrinsic predictive accuracy than the decision tree for the control of start-up and power-up processes. With further expansion of the datasets, it is expected that the predictive advantage of the neural network will become more evident. However, predictive accuracy is not the only criterion by which different predictive algorithms can be compared. The poor interpretability and high training costs of the neural network limit its application in practice while the accuracy of the decision tree can be improved via ensemble strategies. Therefore, deciding an algorithm to use is a complex selection process that should reflect the constraints of the specific application.

The Research on Complex Lithology Identification Based on Well Logs: A Case Study of Lower 1st Member of the Shahejie Formation in Raoyang Sag

Lithology identification is the basis for sweet spot evaluation, prediction, and precise exploratory deployment and has important guiding significance for areas with low exploration degrees. The lithology of the shale strata, which are composed of fine-grained sediments, is complex and varies regularly in the vertical direction. Identifying complex lithology is a typical nonlinear classification problem, and intelligent algorithms can effectively solve this problem, but different algorithms have advantages and disadvantages. Compared were the three typical algorithms of Fisher discriminant analysis, BP neural network, and classification and regression decision tree (C&RT) on the identification of seven lithologies of shale strata in the lower 1st member of the Shahejie Formation (Es1L) of Raoyang sag. Fisher discriminant analysis method is linear discriminant, the recognition effect is poor, the accuracy is 52.4%; the accuracy of the BP neural network to identify lithology is 82.3%, but it belongs to the black box and can not be visualized; C&RT can accurately identify the complex lithology of Es1L, the accuracy of this method is 85.7%, and it can effectively identify the interlayer and thin interlayer in shale strata.

BADM-Net: Hierarchical Classification Network for Identifying Anomalous Trends in Bridge Monitoring Data Patterns

SHM systems have been widely implemented in long-span bridges, and seas of field measurement data have been accumulated. Due to the imperfect sensors, data transmission and acquisition, various anomalies inevitably exist in the SHM data, which may lead to unreliable structural condition assessment. Thus, an effective approach for detecting data anomalies is highly desirable. Due to the imbalanced data, some anomalous patterns are undertrained in popular end-to-end deep neural network models, resulting in a reduction in detection precision. In this paper, a hierarchical classification model with deep neural network tree is proposed for imbalanced data. The DNN tree contains three levels: (1) CNN to divide seven types of data into four categories (134, 2, 5, 67), denoted as C4; (2) two DNNs to classify to two classes separately (1, 34, 6, 7), denoted as D2D2; (3) DNNs to classify to two classes (3, 4). So, the DNN tree is presented as C4_D2D2_D2. The DNN tree is an open framework and can be defined based on the data characteristics. In the data processing, three data sets are built for training, namely single-channel data set, dual-channel data set and statistical data set. To validate our work, we considered the effects of balanced and imbalanced training sets and training ratios. The results show that our model can detect the multi-pattern anomalies of SHM data efficiently with 95.5% high accuracy. Besides, the proportion of abnormal data classified to normal data has been reduced, especially 3-minor. This model successfully solves the problem in a simple and easy to understand way, which has certain reference significance for the bridge structure anomaly judgment in the future.

An Interpretable Station Delay Prediction Model Based on Graph Community Neural Network and Time-Series Fuzzy Decision Tree

High-speed train delay prediction has always been one of the important research issues in the railway dispatching. Accurate and interpretable delay prediction can enable staff to implement preventive measures and scheduling decisions in advance, and guide relevant departments to cooperate in completing complex transportation tasks, so as to improve rail transit operations, service quality, and the efficiency of train operation. This article proposes a new interpretable model based on graph community neural network and time-series fuzzy decision tree. This model can well capture the influence of spatiotemporal characteristics, train community structure, and multifactor in high-speed train station delay prediction. Besides, the time series fuzzy decision tree based on multiobjective optimization and reduced error pruning can mine potential decision rules to improve the model's interpretability, transparency, and high reliability. Finally, we prove that the prediction effect of the proposed model is superior than the other seven state-of-the-art models and our model is interpretable.

Thermal error modeling of machine tool spindle through an ensemble approach

Thermal error compensation of machine tool is cost-effective than other methods. Towards this, data-driven machine learning (ML) algorithms have been used to produce accurate prediction models. However, ML models have limitations, such as overfitting, requiring large data etc. In present work, a hybrid model is proposed by exploiting the linear regression (LR), support vector machine (SVM), neural network (NN), and decision tree (DT) models. For this purpose, the optimum weights to each constituent model is identified by cosine similarity maximization. The developed models are validated against the experimental data. The prediction results with optimized weight are compared with equal weights and the root means square error (RMSE) for both methods are 1.8879 and 2.8978, respectively. The RMSE shows that the hybrid model produces good accuracy for both small and large data sets compared to individual models.

MLHandwrittenRecognition: Handwritten Digit Recognition using Machine Learning Algorithms

Handwritten digit recognition has remained a topic of interest to computer vision scientists. Its origination precedes the emergence of the machine as it is a crucial component of the digital transformation of the majority of institutions in numerous fields. With the uprising of machine models, choosing a satisfactory and fit algorithm for this multi-class (0-9) classification problem became challenging. This paper aims to compare seven machine learning algorithms in terms of their performance metrics in recognizing handwritten digits employing two datasets. The - Nearest Neighbors (kNN), Support Vector Machine (SVM), Logistic Regression, Neural Network, Random Forest (RF), Naive Bayes, and Decision Tree models are accordingly evaluated concerning the Area Under the Curve (AUC), accuracy (ACC), F1-score (F1), precision (PREC), and recall (REC). The widely used Modified National Institute of Standards and Technology database (MNIST) dataset and the Handwritten Digit Classification dataset (HDC) have been the providers of the images on which this research is conducted. The results confirm that the Neural Networks model is a great classifier for this problem; however, it presents similar results to other machine learning classifiers in several cases. Therefore, this paper does not provide an absolute choice of a classifier for the handwritten digit recognition problem but rather explains the reason behind the performance of each model.

Rain Discrimination with Machine Learning Classifiers for Opportunistic Rain Detection System Using Satellite Micro-Wave Links.

In the climate change scenario the world is facing, extreme weather events can lead to increasingly serious disasters. To improve managing the consequent risks, there is a pressing need to have real-time systems that provide accurate monitoring and possibly forecasting which could help to warn people in the affected areas ahead of time and save them from hazards. The oblique earth-space links (OELs) have been used recently as a method for real-time rainfall detection. This technique poses two main issues related to its indirect nature. The first one is the classification of rainy and non-rainy periods. The second one is the determination of the attenuation baseline, which is an essential reference for estimating rainfall intensity along the link. This work focuses mainly on the first issue. Data referring to eighteen rain events were used and have been collected by analyzing a satellite-to-earth link quality and employing a tipping bucket rain gauge (TBRG) properly positioned, used as reference. It reports a comparison among the results obtained by applying four different machine learning (ML) classifiers, namely the support vector machine (SVM), neural network (NN), random forest (RF), and decision tree (DT). Various data arrangements were explored, using a preprocessed version of the TBRG data, and extracting two different sets of characteristics from the microwave link data, containing 6 or 12 different features, respectively. The achieved results demonstrate that the NN classifier has outperformed the other classifiers.

E2H Distance-Weighted Minimum Reference Set for Numerical and Categorical Mixture Data and a Bayesian Swap Feature Selection Algorithm

Generally, when developing classification models using supervised learning methods (e.g., support vector machine, neural network, and decision tree), feature selection, as a pre-processing step, is essential to reduce calculation costs and improve the generalization scores. In this regard, the minimum reference set (MRS), which is a feature selection algorithm, can be used. The original MRS considers a feature subset as effective if it leads to the correct classification of all samples by using the 1-nearest neighbor algorithm based on small samples. However, the original MRS is only applicable to numerical features, and the distances between different classes cannot be considered. Therefore, herein, we propose a novel feature subset evaluation algorithm, referred to as the “E2H distance-weighted MRS,” which can be used for a mixture of numerical and categorical features and considers the distances between different classes in the evaluation. Moreover, a Bayesian swap feature selection algorithm, which is used to identify an effective feature subset, is also proposed. The effectiveness of the proposed methods is verified based on experiments conducted using artificially generated data comprising a mixture of numerical and categorical features.

Comparing Artificial Neural Network and Decision Tree Algorithm to Predict Tides at Tanjung Priok Port

Tanjung Priok Port is an international port in Indonesia located in Tanjung Priok, North Jakarta. For all activities carried out at Tanjung Priok Port to run smoothly, this research was made which aims to predict the height of tides using the Artificial Neural Network (ANN) and Decision Tree methods with a quantitative approach. Artificial Neural Network (ANN) is a technique inspired by the way the biological nervous system works, namely in brain cells in processing information received by humans. while Decision Tree is also known as a decision tree which is an algorithm for building a decision hierarchy structure. The process of making a Decision Tree starts from the Root Node to the Leaf Node which is done recursively. This research was conducted to predict the height of tides in January 2018 - June 2018. By using both methods that have been computed, the ANN method produces a smaller MSE value than the Decision Tree method. The ANN method produces an MSE value of 0.003727983. While the Decision Tree method produces an MSE value of 0.009870259. If the dataset used has larger amount of data and the architecture of each algorithm is more complex, then the calculation results obtained will be more accurate.

Analysis Factors Affecting Egyptian Inflation Based on Machine Learning Algorithms

<abstract><p>Given that inflation is one of the most important problems facing the Egyptian economy, determining the factors affecting it is very important. Thus, we aim to use machine learning algorithms like Support vector machine (SVM), K-nearest neighbor (KNN), Random Forest (R.F.), Neural Network (ANN), Gradient boosting (G.B.) and decision tree (D.T.) to determine the accurate algorithm and analyze the factors affecting on Egypt inflation. The study found that the G.B. algorithm is the most accurate among the used algorithms and showed that The major significant variables determining inflation in Egypt are the exchange rate (30.5%), gross fixed formation (24.5%) and government expenditure (12.3%). We also found a positive relationship between the inflation rate and government expenditure, money supply, gross domestic product (GDP) growth, gross fixed formation, foreign direct investment, GDP per capita and exchange rate. Furthermore, there is a negative relationship between the inflation rate, household expenditure and the external trade balance.</p></abstract>

Sea Surface Target Detection Using Global False Alarm Controllable Adaptive Boosting Based on Correlation Features

In the complex marine environment, traditional detectors based on the classifiers cannot guarantee the global false alarm control. In this paper, we propose a detector based on dual channel convolutional neural network (DC-CNN) and global false alarm controllable adaptive boosting tree (GFAC-A), which is shortened as DC-CNN-GFAC-A. DC-CNN focuses on the correlation features of radar echoes in time domain and frequency domain, better use multi-dimensional features and show a better feature extraction ability. Through the application of GFAC-A, the false alarm rate is introduced into the algorithm combining decision tree and AdaBoost to achieve the high-performance detection of global false alarm controllable for high-dimensional features. It is heuristics. The combination of DC-CNN and GFAC-A solves the disadvantage that current classifiers cannot meet the conditions of good performance and low false alarm. First, the connectivity information is obtained by using the frequency domain amplitude characteristics of radar echo data, and the recursive information is obtained by using the nonlinear recursive time series characteristics. And the dual-channel datasets of targets and clutter are built. Then, DC-CNN is built to extract and fuse high-dimensional features to obtain feature vectors of targets and clutter. Besides, the performance comparison of different neural network model combinations is carried out. Finally, compared with the traditional threshold-controllable classifiers, the proposed GFAC-A classifier achieves the high detection performance under the global controlled false alarm. The results show that DC-CNN-GFAC-A can achieve 96.491% detection accuracy when the false alarm rate is 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> , which is superior to other detections.

Applications of Neural Network-Based Plan-Cancer Method for Primary Diagnosis of Mesothelioma Cancer.

"Malignant mesothelioma (MM)" is an uncommon although fatal form of cancer. The proper MM diagnosis is crucial for efficient therapy and has significant medicolegal implications. Asbestos is a carcinogenic material that poses a health risk to humans. One of the most severe types of cancer induced by asbestos is "malignant mesothelioma." Prolonged shortness of breath and continuous pain are the most typical symptoms of the condition. The importance of early treatment and diagnosis cannot be overstated. The combination "epithelial/mesenchymal appearance of MM," however, makes a definite diagnosis difficult. This study is aimed at developing a deep learning system for medical diagnosis MM automatically. Otherwise, the sickness might cause patients to succumb to death in a short amount of time. Various forms of artificial intelligence algorithms for successful "Malignant Mesothelioma illness" identification are explored in this research. In relation to the concept of traditional machine learning, the techniques support "Vector Machine, Neural Network, and Decision Tree" are chosen. SPSS has been used to analyze the result regarding the applications of Neural Network helps to diagnose MM.

Prediction of Foot Ulcers Using Artificial Intelligence for Diabetic Patients at Cairo University Hospital, Egypt.

In Egypt, diabetic foot ulcers markedly contribute to the morbidity and mortality of diabetic patients. Accurately predicting the risk of diabetic foot ulcers could dramatically reduce the enormous burden of amputation. The aim of this study is to design an artificial intelligence-based artificial neural network and decision tree algorithms for the prediction of diabetic foot ulcers. A case-control study design was utilized to fulfill the aim of this study. The study was conducted at the National Institute of Diabetes and Endocrine Glands, Cairo University Hospital, Egypt. A purposive sample of 200 patients was included. The tool developed and used by the researchers was a structured interview questionnaire including three parts: Part I: demographic characteristics; Part II: medical data; and Part III: in vivo measurements. Artificial intelligence methods were used to achieve the aim of this study. The researchers used 19 significant attributes based on medical history and foot images that affect diabetic foot ulcers and then proposed two classifiers to predict the foot ulcer: a feedforward neural network and a decision tree. Finally, the researchers compared the results between the two classifiers, and the experimental results showed that the proposed artificial neural network outperformed a decision tree, achieving an accuracy of 97% in the automated prediction of diabetic foot ulcers. Artificial intelligence methods can be used to predict diabetic foot ulcers with high accuracy. The proposed technique utilizes two methods to predict the foot ulcer; after evaluating the two methods, the artificial neural network showed a higher improvement in performance than the decision tree algorithm. It is recommended that diabetic outpatient clinics develop health education and follow-up programs to prevent complications from diabetes.

Data-Efficient Reinforcement Learning with Data Augmented Episodic Memory

Reinforcement learning has been implemented to model a task by doing the task repeatedly to get the maximum results based on the reward and punishment policy. It has been implemented in the game and agent-based modelling. In the game, the game agent or Non-Players character can be modelled using several techniques to achieve the goal (e.g. reinforcement learning, deep neural network and Monte Carlo Tree Search). Deep neural networks and Monte Carlo Tree Search, two more sophisticated techniques in reinforcement learning algorithms, assisted the present reinforcement learning in resolving more challenging issues. However, this area has two challenges: the minimum number of data to model and generalization to different environments. Determining a minimum number of data required by the architecture to train the model is quite a cumbersome task to be applied to real- world jobs and situations since it demands substantial data to be explicitly provided and trial-and-error re-configuration. This work proposes a Data-Efficient Reinforcement Learning model by augmenting the data and implementing episodic memory. To illustrate the effectiveness of the proposed model, this research compares it to several models, such as the Deep Q-Network (DQN) with episodic memory to the same model with data augmentation and episodic memory. The model adds to the observations before being stored in the agent's memory, causing the agent to use the same logic and take the same action in comparable situations. The outcome demonstrates that the augmented model can surpass the fundamental model in speed (with an improvement of 50% quicker).

Human brain-inspired chemical artificial intelligence tools for the analysis and prediction of the anion-sensing characteristics of an imidazole-based luminescent Os(II)-bipyridine complex.

Neural network and decision tree-based soft computing techniques are implemented in this work for the thorough analysis of the multichannel anion-sensing characteristics of an Os(II)-bipyridine complex derived from imidazole-4,5-bis(benzimidazole) ligand. With the aid of three imidazole NH protons in its outer coordination sphere, a substantial change in the spectral response as well as OsII/OsIII potential is made possible upon treating with anions of varying basicity. Initial hydrogen bonding between NH protons and anions and thereafter complete proton transfer from the complex backbone probably take place in the process. The deprotonation of the complex by specific anions and restoration to its original form by acid is also reversible. The responsiveness of the new compound is complex enough to imitate multiple sophisticated binary and ternary Boolean logic (BL) functions (NOT logic, combinational logic, traffic signal, set-reset flip-flop logic, and ternary NOR logic) by employing its spectral and redox outputs upon the action of suitable anions and acid in a proper sequence. Executing sensing investigations on altering the amount of the anions within a widespread range is often time-consuming and tedious. To overcome the lacuna, we implemented multiple soft computing techniques, viz., fuzzy logic (FL), artificial neural networks (ANNs), adaptive neuro-fuzzy inference system (ANFIS), and decision tree (DT) regression, for the thorough analysis and prediction of the experimentally observed results. The outcomes obtained from different techniques were compared among themselves as well as with the experimental data and utilized for the proper modeling of the anion-sensing behaviors of the complex.

Detection Collision Flows in SDN Based 5G Using Machine Learning Algorithms

The rapid advancement of wireless communication is forming a hyper-connected 5G network in which billions of linked devices generate massive amounts of data. The traffic control and data forwarding functions are decoupled in software-defined networking (SDN) and allow the network to be programmable. Each switch in SDN keeps track of forwarding information in a flow table. The SDN switches must search the flow table for the flow rules that match the packets to handle the incoming packets. Due to the obvious vast quantity of data in data centres, the capacity of the flow table restricts the data plane’s forwarding capabilities. So, the SDN must handle traffic from across the whole network. The flow table depends on Ternary Content Addressable Memorable Memory (TCAM) for storing and a quick search of regulations; it is restricted in capacity owing to its elevated cost and energy consumption. Whenever the flow table is abused and overflowing, the usual regulations cannot be executed quickly. In this case, we consider low-rate flow table overflowing that causes collision flow rules to be installed and consumes excessive existing flow table capacity by delivering packets that don’t fit the flow table at a low rate. This study introduces machine learning techniques for detecting and categorizing low-rate collision flows table in SDN, using Feed Forward Neural Network (FFNN), K-Means, and Decision Tree (DT). We generate two network topologies, Fat Tree and Simple Tree Topologies, with the Mininet simulator and coupled to the OpenDayLight (ODL) controller. The efficiency and efficacy of the suggested algorithms are assessed using several assessment indicators such as success rate query, propagation delay, overall dropped packets, energy consumption, bandwidth usage, latency rate, and throughput. The findings showed that the suggested technique to tackle the flow table congestion problem minimizes the number of flows while retaining the statistical consistency of the 5G network. By putting the proposed flow method and checking whether a packet may move from point A to point B without breaking certain regulations, the evaluation tool examines every flow against a set of criteria. The FFNN with DT and K-means algorithms obtain accuracies of 96.29% and 97.51%, respectively, in the identification of collision flows, according to the experimental outcome when associated with existing methods from the literature.

Convolutional neural network tree species identification based on tree-ring radial section image features.

Convolutional neural networks can automatically identify tree species based on the images of structural features of tree-rings samples. In this study, we used a tree-ring image dataset for different species to achieve tree-ring based species automatic identification with high accuracy by four convolutional neural network models (LeNet, AlexNet, GoogLeNet, and VGGNet), aiming to determine the identification accuracy of the models, clarify the species misidentification during the automatic processes, and explore the identification differences among the models. The results showed that tree species identification derived from the trained convolutional neural network models was reliable, with the GoogLeNet and LeNet showed the highest (96.7%) and lowest (66.4%) identification accuracy. The tree species identifications using different models were highly consistent. Quercus mongolica and Abies nephrolepis showed the highest (100% for AlexNet) and lowest identification accuracy, respectively. Misidentification could occur among tree species with similar tree-ring structure. The identification accuracy of the models was higher at family and genus levels than that at the species level. The identification accuracy of broadleaf tree species was higher than that of coniferous trees due to distinct radial structure among broadleaf species. Overall, our method achieved a high accuracy for tree species identification, and provided a fast, convenient, and automatic tree species identification by detecting specific tree-ring properties with convolutional neural network.

Default Prediction for Loan Lenders Using Machine Learning Algorithms

Credit loans are considered most essential aspect of most financial institutions. All loan mortgagees or lenders are demanding to identify out effective commercial and business approaches to encourage customers to apply their credit loans. There are numerous business patrons who act negatively after their requests got approval. To avert this condition, lenders have to discover some techniques to forecast customer’s behaviors. This resulted to the usage of machine learning algorithms by the financial lending institutions for accessing loan applicants. Despite advancements in automating decision-based loan systems, most existing models do not consider the “early loan repayment” attribute as a factor in resolving this prediction error. In reality, the amendment for preliminary loan reimbursement in model building is obligatory, since a larger numbers of timely loan reimbursement observed during the loan period, reduces default rate. For effective model’s comparison based on accuracy and minimum errors of prediction, six supervised machine learning algorithms i.e. Random Forest, Artificial Neural Network, Classification and Regression Tree, Support Vector Machine, Logistic Regression, and Naïve Bayes were adopted to develop a default prediction models which include the early loan repayment attribute. The models were trained and tested on a loan dataset consisting of attributes with, and without early loan repayment attribute and were evaluated using five performance metrics. The results of the performance evaluation show that models that account for early loan repayment have higher accuracy, recall, precision, Root Mean Square Error and Receiver Operative Characteristics curve values than models trained without the early loan repayment attribute. The Random forest model proofed to be the best predictive model having 93% accuracy, 11% RMSE, 90% precision, 89% recall and 81% ROC value over others models.

Machine learning-assisted ensemble analysis for the prediction of urinary tract infection in elderly patients with ovarian cancer after cytoreductive surgery

Urinary tract infection (UTI) is a common type of postoperative infection following cytoreductive surgery for ovarian cancer, which severely impacts the prognosis and quality of life of patients. To develop a machine learning assistant model for the prevention and control of nosocomial infection. A total of 674 elderly patients with ovarian cancer who were treated at the Department of Gynaecology at Jingzhou Central Hospital between January 31, 2016 and January 31, 2022 and met the inclusion criteria of the study were selected as the research subjects. A retrospective analysis of the postoperative UTI and related factors was performed by reviewing the medical records. Five machine learning-assisted models were developed using two-step estimation methods from the candidate predictive variables. The robustness and clinical applicability of each model were assessed using the receiver operating characteristic curve, decision curve analysis and clinical impact curve. A total of 12 candidate variables were eventually included in the UTI prediction model. Models constructed using the random forest classifier, support vector machine, extreme gradient boosting, and artificial neural network and decision tree had areas under the receiver operating characteristic curve ranging from 0.776 to 0.925. The random forest classifier model, which incorporated factors such as age, body mass index, catheter, catheter intubation times, blood loss, diabetes and hypoproteinaemia, had the highest predictive accuracy. These findings demonstrate that the machine learning-based prediction model developed using the random forest classifier can be used to identify elderly patients with ovarian cancer who may have postoperative UTI. This can help with treatment decisions and enhance clinical outcomes.

Quality Control Model of Ore Processing Based on Elman Neural Network

In this paper, by establishing Elman neural network and decision tree model, the quality control of ore processing is studied and reasonably predicted.Predict the product quality and the temperature corresponding to the target quality. Elman neural network model is established, training set and output set are input for training, and test set is input into neural network to obtain the relative error of real value and training value; After verification, the relative error between our real value and training value is within 100% ± 5%, and the model is accurate; Finally, the input value is substituted to obtain the forecast data.Establish the mathematical model of product qualification rate. The decision tree model is established by using SPSS, and the decision tree system is constructed by using the four factors of system setting temperature, raw ore parameters, process data and final ore quality. Through inspection, our R 2 is 0.999991, close to 1, and the model fitting is good. Compared with other parameters, it can be found that the model has good sensitivity and high accuracy. Finally, the input data is sent into the model to get the final prediction results.