Decision Tree Forest Research Articles

Study of Forecasting and Estimation Methodology of Oilfield Development Cost Based on Machine Learning

Due to the oil price fluctuations in recent decades, international and national oil companies have developed programs of strategically oriented development and assets optimization. Many companies have also promptly promoted opportunities for oilfield joint ventures. Therefore, a fast and accurate assessment methodology of oilfield assets, including planning and costs assessment, is expected to be proposed. Oilfield development cost assessment can be dynamically affected by a number of factors, including oilfield internal indexes and macroeconomic indexes. Based on a machine learning algorithm and combining mathematical and statistical methodology, the Microsoft Azure machine learning studio has been used for modeling the oilfield development cost. The proposed method has adopted three algorithms: a neural network, a boosted decision tree, and a decision forest. Results showed that the boosted decision tree and decision forest algorithms can achieve PFI ranking with stable training results. The results of the machine training model have been analyzed, and they showed that the permutation feature importance (PFI) model can provide reasonable scientific and technical support for oil companies and help to attain more effective and accurate estimation and prediction of oilfield assets.

Automatic solution for solar cell photo-current prediction using machine learning

In this paper, we discuss the prediction of future solar cell photo-current generated by the machine learning algorithm. For the selection of prediction methods, we compared and explored different prediction methods. Precision, MSE and MAE were used as models due to its adaptable and probabilistic methodology on model selection. This study uses machine learning algorithms as a research method that develops models for predicting solar cell photo-current. We create an electric current prediction model. In view of the models of machine learning algorithms for example, linear regression, Lasso regression, K Nearest Neighbors, decision tree and random forest, watch their order precision execution. In this point, we recommend a solar cell photocurrent prediction model for better information based on resistance assessment. These reviews show that the linear regression algorithm, given the precision, reliably outperforms alternative models in performing the solar cell photo-current prediction Iph

Behavior of decision forest classification in dynamic manufacturing systems

Managers of manufacturing systems are constantly looking for ways to predict future production conditions. Due to the system’s complexity, modelling is effortful and never completed. So-called random forests of decision trees seem to be a promising machine-learning tool to forecast key figures of manufacturing systems. The selection of data to teach such classifiers significantly influences the quality of the prediction. However, quality of data and prediction decreases in case of a dynamic system. This paper deals with one possible way of data handling for decision forests in changing manufacturing system.

Assessment of Advanced Artificial Intelligence Techniques for Streamflow Forecasting in Jhelum River Basin

| Streamflow forecasting is a crucial hydrological variable. In the current study, the Artificial Intelligence (AI) based techniques: TB (Tree Boost), DTF Decision Tree Forest, SDT Single Decision Tree and conventional Multilayer Perceptron Neural Networks (MLPNN) are used for predicting streamflow of Jhelum River basin. The dataset was divided into two sections, i.e., training dataset (1971-2000); and testing dataset (2001-12). The tendency investigation was done by the Sen’s slope and Mann–Kendall (MK). Decreasing trends annually and seasonally found in MK and Sen’s Slope tests. The highest decreasing trend of -2.23 was observed in Autumn at Narran station, while the lowest change of -0.09 annually observed at Garhi Habibullah station at 95% of the significance level. The flow duration curves (FDCs) of all basin stations showed that DTF performed better and is more effective than other AI techniques. R2, RMSE, and NSE assessed the performance evaluation. DTF was more efficient AI techniques with the average evaluation parameters R2, NSE, and RMSE are 0.998, 0.992, and 382 m3/sec. The assessment revealed that DTF has potential and may be considered as an alternative method for streamflow forecasting.

Tax Default Prediction Using Feature Transformation-Based Machine Learning

This study proposes to address the economic significance of unpaid taxes by using an automatic system for predicting a tax default. Too little attention has been paid to tax default prediction in the past. Moreover, existing approaches tend to apply conventional statistical methods rather than advanced data analytic approaches, including state-of-the-art machine learning methods. Therefore, existing studies cannot effectively detect tax default information in real-world financial data because they fail to take into account the appropriate data transformations and nonlinear relationships between early-warning financial indicators and tax default behavior. To overcome these problems, this study applies diverse feature transformation techniques and state-of-the-art machine learning approaches. The proposed prediction system is validated by using a dataset showing tax defaults and non-defaults at Finnish limited liability firms. Our findings provide evidence for a major role of feature transformation, such as logarithmic and square-root transformation, in improving the performance of tax default prediction. We also show that extreme gradient boosting and the systematically developed forest of multiple decision trees outperform other machine learning methods in terms of accuracy and other classification performance measures. We show that the equity ratio, liquidity ratio, and debt-to-sales ratio are the most important indicators of tax defaults for 1-year-ahead predictions. Therefore, this study highlights the essential role of well-designed tax default prediction systems, which require a combination of feature transformation and machine learning methods. The effective implementation of an automatic tax default prediction system has important implications for tax administration and can assist administrators in achieving feasible government expenditure allocations and revenue expansions.

THE STUDY USING ENSEMBLE LEARNING FOR RECOMMENDING BETTER FUTURE INVESTMENTS

Generally, House estimation record addresses the summarized esteem changes of private housing. While at a single-family house cost desire, it needs more exact procedure reliant on the spot, house type, size, structure year, close by improvements, and some various parts which could impact house demand and deftly. With limited dataset and data incorporates, a sensible and composite data pre-taking care of, creative component planning methodology is assessed in this paper. People are careful when they are endeavouring to buy another house with their money related plans and market strategies. The objective of the paper is to measure the sensible house costs for non-house holders reliant on their financial courses of action and their desires. By analysing the earlier item, entry ranges and besides alerts enhancements, guessed costs will be evaluated. The paper includes expectations utilizing diverse Regression procedures like Ridge, LASSO, Random Forest, SVM (support-vector machine), KNN (k-nearest neighbours), Ada Boost Regression, Stacking (decision tree, lasso and random forest), Decision Tree. House estimation figure on an instructive file has been done by using all the recently referenced systems to find the best among them. The reason of this paper is to help the vendor with assessing the selling cost of a house perfectly and to assist people with foreseeing the time slap to store up a house. A part of the related segments that influence the cost were furthermore taken into examinations, for instance, states of being, thought, area and territory, etc.

Machine learning predictivity applied to consumer creditworthiness

Credit risk evaluation has a relevant role to financial institutions, since lending may result in real and immediate losses. In particular, default prediction is one of the most challenging activities for managing credit risk. This study analyzes the adequacy of borrower’s classification models using a Brazilian bank’s loan database, and exploring machine learning techniques. We develop Support Vector Machine, Decision Trees, Bagging, AdaBoost and Random Forest models, and compare their predictive accuracy with a benchmark based on a Logistic Regression model. Comparisons are analyzed based on usual classification performance metrics. Our results show that Random Forest and Adaboost perform better when compared to other models. Moreover, Support Vector Machine models show poor performance using both linear and nonlinear kernels. Our findings suggest that there are value creating opportunities for banks to improve default prediction models by exploring machine learning techniques.

Intelligent socio-economic status prediction system using machine learning models on Rajahmundry A.P., SES dataset

Background: Developing economic and social systems and assuring the efficiency of economic and social processes is the major task for the government of any country. Predictable machine learning (ML) models are used for analyzing data sets that allow more efficient enterprise management. Now a day, the research on Socio-Economic Status (SES) and Machine Learning (ML) is very crucial to find socio-economic inequalities, and take further actions that are preventions, protections, and suppressions. Objectives: The mainobjective of this research is to understand the Socio Economic System issues and predicting SES levels on particular area like Rajahmundry, AP, India using statistical analysis and machine learning methodologies. Methods: In this, we analyze the data that is collected from Rajahmundry (Rajamahandravaram),Andhra Pradesh, India with 48 feature attributes (dimensions), and one target four class attribute (poor, rich, middle, upper-middle ). The SES levels like poor, rich, middle, and upper-middle classes are predicted by 5 ML algorithms. Findings: In this paper, we conduct the statistical analysis of each attribute, and analyze and compare the performance accuracies using confusion matrix, performance parameter (classification accuracy, Precision,Recall, and F1) values and receive operating characteristic (ROC) under AUC values of five efficient ML algorithms like Naïve Bayes, Decision Trees (DTs), k-NN, SVM (kernel RBF) and Random Forest (RF). We observed that the RF algorithm showed better results when compared with other algorithms for the Rajahmundry AP SES dataset. The RF algorithm performs 97.82% of classification accuracy (CA) and time is taken for model construction 0.41 seconds. The next superior performed ML model is DTs with 96.67% of CA and 0.16 seconds for model construction. Novelty: Comprehensive analysis indicates that the novel AP SES Dataset with empirical statistical analysis gives the good results and predicts the SES levels with RF model is very effective. Keywords: Machine Learning; socio-economic status; Rajahmundry;household; poverty

Maximising the ability of stimulus-frequency otoacoustic emissions to predict hearing status and thresholds using machine-learning models

Objective This study aimed to maximise the ability of stimulus-frequency otoacoustic emissions (SFOAEs) to predict hearing status and thresholds based on machine-learning models. Design SFOAE data and audiometric thresholds were collected at octave frequencies from 0.5 to 8 kHz. Support vector machine, k-nearest neighbour, back propagation neural network, decision tree, and random forest algorithms were used to build classification models for status identification and to develop regression models for threshold prediction. Study sample About 230 ears with normal hearing and 737 ears with sensorineural hearing loss. Results All classification models yielded areas under the receiver operating characteristic curve of 0.926–0.994 at 0.5–8 kHz, superior to the previous SFOAE study. The regression models produced lower standard errors (8.1–12.2 dB, mean absolute errors: 5.53–8.97 dB) as compared to those for distortion-product and transient-evoked otoacoustic emissions previously reported (8.6–19.2 dB). Conclusions SFOAEs using machine-learning approaches offer promising tools for the prediction of hearing capabilities, at least at 0.5–4 kHz. Future research may focus on further improvements in accuracy and reductions in test time to improve clinical utility.

A feature fusion technique for improved non-intrusive load monitoring

Load identification is an essential step in Non-Intrusive Load Monitoring (NILM), a process of estimating the power consumption of individual appliances using only whole-house aggregate consumption. Such estimates can help consumers and utility companies improve load management and save power. Current state-of-the-art methods for load identification generally use either steady state or transient features for load identification. We hypothesize that these are complementary features and so a hybrid combination of them will result in an improved appliance signature. We propose a novel hybrid combination that has the advantage of being low-dimensional and can thus be easily integrated with existing classification models to improve load identification. Our improved hybrid features are then used for building appliance identification models using Naive Bayes, KNN, Decision Tree and Random Forest classifiers. The proposed NILM methodology is evaluated for robustness in changing environments. An automated data collection setup is established to capture 7 home appliances aggregate data under varying voltages. Experimental results show that our proposed feature fusion based algorithms are more robust and outperform steady state and transient feature-based algorithms by at least +9% and +15% respectively.

MLDroid—framework for Android malware detection using machine learning techniques

This research paper presents MLDroid—a web-based framework—which helps to detect malware from Android devices. Due to increase in the popularity of Android devices, malware developers develop malware on daily basis to threaten the system integrity and user’s privacy. The proposed framework detects malware from Android apps by performing its dynamic analysis. To detect malware from real-world apps, we trained our proposed framework by selecting features which are gained by implementing feature selection approaches. Further, these selected features help to build a model by considering different machine learning algorithms. Experiment was performed on 5,00,000 plus Android apps. Empirical result reveals that model developed by considering all the four distinct machine learning algorithms parallelly (i.e., deep learning algorithm, farthest first clustering, Y-MLP and nonlinear ensemble decision tree forest approach) and rough set analysis as a feature subset selection algorithm achieved the highest detection rate of 98.8% to detect malware from real-world apps.

Capacity Random Forest for Correlative Multiple Criteria Decision Pattern Learning

The Choquet capacity and integral is an eminent scheme to represent the interaction knowledge among multiple decision criteria and deal with the independent multiple sources preference information. In this paper, we enhance this scheme’s decision pattern learning ability by combining it with another powerful machine learning tool, the random forest of decision trees. We first use the capacity fitting method to train the Choquet capacity and integral-based decision trees and then compose them into the capacity random forest (CRF) to better learn and explain the given decision pattern. The CRF algorithms of solving the correlative multiple criteria based ranking and sorting decision problems are both constructed and discussed. Two illustrative examples are given to show the feasibilities of the proposed algorithms. It is shown that on the one hand, CRF method can provide more detailed explanation information and a more reliable collective prediction result than the main existing capacity fitting methods; on the other hand, CRF extends the applicability of the traditional random forest method into solving the multiple criteria ranking and sorting problems with a relatively small pool of decision learning data.

Research on video flame detection algorithm based on improved DS evidence theory

The detection and prevention of flame is of great value to protect people’s lives and property safety. At present, most flame detection methods use a single classifier and have achieved some results. However, a single classification algorithm has poor adaptability to fire detection in a variety of complex situations. Therefore, a multi-classifier fusion flame detection algorithm is proposed based on Dempster-Shafer (DS) evidence theory is proposed. In short, firstly four classifiers are used to classify the same flame feature, and the four classification results are fused to make preliminary decision. The four classifiers include support vector machine (SVM), K-nearest neighbor (KNN), decision tree (DT) and random forest (RF). Second, three complementary flame features are chosen, namely color, texture and shape changes. Finally, the preliminary decision results of the three features are fused to obtain the final classification result. It should be noted that when different classifiers have strong conflicts on the classification result of the same feature, the fusion rule of DS evidence theory will be invalid. To solve this problem, the DS evidence theory is improved. For the experiment, the public flame videos are collected to construct a data set including different complex scenes for algorithm verification, where the frame rate of the video is 15 or 24 frames/s and the resolution is 320 × 240. The experimental results show that there is a strong complementary among the results of different single classifier. The multi-classifier fusion algorithm can achieve better classification performance and robust performance than the single classifier by integrating the results of each classifier, and its average detection rate reaches 93.08%. In addition, for the changes of different environments, the proposed method has higher adaptability and stability than other state-of-art methods.

Detecting falls and estimation of daily habits with depth images using machine learning algorithms.

Different approaches have been proposed in the literature to detect the fall of an elderly person. In this paper, we propose a fall detection method based on the classification of parameters extracted from depth images. Three supervised learning methods are compared: decision tree, K-Nearest Neighbors (K-NN) and Random Forests (RF). The methods have been tested on a database of depth images recorded in a nursing home over a period of 43 days. The Random Forests based method yields the best results, achieving 93% sensitivity and 100% specificity when we restrict our study around the bed. Furthermore, this paper also proposes a 37 days follow-up of the person, to try and estimate his or her daily habits.

Analysis and Prediction Blood Pressure and Disease by Applying Decision Tree, Naïve Base and Random Forest algorithms

Analysis and Prediction Blood Pressure and Disease by Applying Decision Tree, Naïve Base and Random Forest algorithms

Improving Decision Tree Forest using Preprocessed Data

Random forest is one of the best techniques in data mining for classification. It not only improves accuracy of classification but performing best for various data types. Data mining researchers concentrated on improving random tree forest by constructing trees by using various methods. In this paper, we are improving decision forest by applying various preprocessing techniques. Decision tree forest is created by using bootstrapped samples. Trees created using preprocessed data improves not only accuracy of classification but also improves time required to construct forest. Experiments are carried out on various UCI data sets to show better performance of our proposed system.

Detection of Phishing Websites using an Efficient Feature-Based Machine Learning Framework

Phishing is a cyber-attack which is socially engineered to trick naive online users into revealing sensitive information such as user data, login credentials, social security number, banking information etc. Attackers fool the Internet users by posing as a legitimate webpage to retrieve personal information. This can also be done by sending emails posing as reputable companies or businesses. Phishing exploits several vulnerabilities effectively and there is no one solution which protects users from all vulnerabilities. A classification/prediction model is designed based on heuristic features that are extracted from website domain, URL, web protocol, source code to eliminate the drawbacks of existing anti-phishing techniques. In the model we combine some existing solutions such as blacklisting and whitelisting, heuristics and visual-based similarity which provides higher level security. We use the model with different Machine Learning Algorithms, namely Logistic Regression, Decision Trees, K-Nearest Neighbours and Random Forests, and compare the results to find the most efficient machine learning framework.

GrpClassifierEC: a novel classification approach based on the ensemble clustering space

BackgroundAdvances in molecular biology have resulted in big and complicated data sets, therefore a clustering approach that able to capture the actual structure and the hidden patterns of the data is required. Moreover, the geometric space may not reflects the actual similarity between the different objects. As a result, in this research we use clustering-based space that convert the geometric space of the molecular to a categorical space based on clustering results. Then we use this space for developing a new classification algorithm.ResultsIn this study, we propose a new classification method named GrpClassifierEC that replaces the given data space with categorical space based on ensemble clustering (EC). The EC space is defined by tracking the membership of the points over multiple runs of clustering algorithms. Different points that were included in the same clusters will be represented as a single point. Our algorithm classifies all these points as a single class. The similarity between two objects is defined as the number of times that these objects were not belong to the same cluster. In order to evaluate our suggested method, we compare its results to the k nearest neighbors, Decision tree and Random forest classification algorithms on several benchmark datasets. The results confirm that the suggested new algorithm GrpClassifierEC outperforms the other algorithms.ConclusionsOur algorithm can be integrated with many other algorithms. In this research, we use only the k-means clustering algorithm with different k values. In future research, we propose several directions: (1) checking the effect of the clustering algorithm to build an ensemble clustering space. (2) Finding poor clustering results based on the training data, (3) reducing the volume of the data by combining similar points based on the EC.Availability and implementationThe KNIME workflow, implementing GrpClassifierEC, is available at https://malikyousef.com

Arabic literal amount sub-word recognition using multiple features and classifiers

Bank check processing is an important application of document analysis and recognition. Recognising the literal amounts from the check images is challenging and an open research problem. In this paper, we present our work on Arabic bank check literal amounts' sub-word recognition using four sets of features and three classifiers namely: support vector machine (SVM), neural network (NN), and decision tree forest (DTF) classifiers. In addition, we investigated two different approaches for classifier fusion. We tested our system on the CENPARMI database of Arabic bank check images. Our recognition results outperform previous published results on the same database.

IDENTIFY OF FRAUDULENT FINANCIAL OPERATIONS USING THE MACHINE LEARNING ALGORITHM

Current work is devoted to the problem of automatic detection of fraudulent financial transactions. The article describes the causes of fraudulent transactions their typical attributes, as well as the basic principle of detection. The concepts of fraudulent and honest transactions are defined. Examples of algorithms for determining suspicious financial transactions in antifraud systems are given. Modern approaches to monitoring and detecting cases of fraud in remote banking systems are considered. The positive and negative aspects of each approach are described. Particular attention is paid to the problem of optimal recognition of transaction classes in highly unbalanced data. Methods for solving the problem of unbalanced data are considered. The choice of means for evaluating the operation of the machine learning model is justified considering the specifics of data distribution. As a solution, we propose an approach based on the use of ensemble classifiers in conjunction with balanced sampling algorithms, the key feature of which is to create a balanced sample not for the entire classifier, but for each student in the ensemble separately. Based on data on fraud in the field of bank credit cards, a comparison is made and the best classifier is selected among such ensemble algorithms as random forest, adaptive boosting and bagging of decision trees. To create balanced subsets of evaluators of ensemble algorithms, the algorithm of random insufficient sampling is used. To search for the optimal parameters of the classifiers, the random search algorithm on the grid is used. The results of experimental comparison of the selected methods are presented. The advantages of the proposed approach are analyzed, and the boundaries of its applicability are discussed.