scholarly journals Research on and Application of Tunnel Structure Defects Prediction Using Machine Learning Methods

2021 ◽  
Author(s):  
Bo Shi ◽  
Hui Su ◽  
Xu Du ◽  
Bao Jiao ◽  
Lin Wang
Keyword(s):  
Machine Learning ◽  
Rapid Development ◽  
Development Trend ◽  
Tunnel Structure ◽  
Underground Engineering ◽  
Learning Methods ◽  
Factors Affecting ◽  
Machine Learning Methods ◽  
Tree Algorithms ◽  
Subway Tunnels

With the rapid development of underground engineering in China, more metro tunnels are being constructed, the mileage of subway tunnels is increasing, and the corresponding problems of tunnel structure diseases are becoming more prominent. At present, the treatment of tunnel structural diseases mainly relies on manual inspection and identification, and research on defects prediction is still lacking. Because of the complexity of the factors affecting tunnel structure diseases, it is difficult to analyze the causes and development trend of the diseases comprehensively by manual analysis. Fortunately, machine learning methods have gained popularity in classification and regression tasks in recent decades. Many algorithms, such as decision tree algorithms, the random forest algorithm, and XGBoost, have been applied in fields including finance, engineering, and transportation. This study aimed to analyze the prediction effect of machine learning models by feeding 68055 segment lining rings of six subway lines in a city. According to the disease records from 2014 to 2016 and the corresponding convergence and characteristic data, defect conditions in 2017 were predicted and compared with real defect conditions in 2017. The accuracy rates and F1 values of the predicted results were all above 80%. The prediction results can help tunnel maintenance departments and relevant government regulators make auxiliary decisions to control tunnel structure diseases, and can help them focus on the tunnel interval of severe diseases to clarify the development trend of tunnel disease.

scholarly journals Advances in Blast-Induced Impact Prediction—A Review of Machine Learning Applications

Minerals ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 601
Author(s):  
Nelson K. Dumakor-Dupey ◽  
Sampurna Arya ◽  
Ankit Jha
Keyword(s):  
Machine Learning ◽  
Prediction Models ◽  
Academic Research ◽  
Empirical Models ◽  
Rock Breakage ◽  
Environmental Implications ◽  
Learning Methods ◽  
Factors Affecting ◽  
Impact Prediction ◽  
Machine Learning Methods

Rock fragmentation in mining and construction industries is widely achieved using drilling and blasting technique. The technique remains the most effective and efficient means of breaking down rock mass into smaller pieces. However, apart from its intended purpose of rock breakage, throw, and heave, blasting operations generate adverse impacts, such as ground vibration, airblast, flyrock, fumes, and noise, that have significant operational and environmental implications on mining activities. Consequently, blast impact studies are conducted to determine an optimum blast design that can maximize the desirable impacts and minimize the undesirable ones. To achieve this objective, several blast impact estimation empirical models have been developed. However, despite being the industry benchmark, empirical model results are based on a limited number of factors affecting the outcomes of a blast. As a result, modern-day researchers are employing machine learning (ML) techniques for blast impact prediction. The ML approach can incorporate several factors affecting the outcomes of a blast, and therefore, it is preferred over empirical and other statistical methods. This paper reviews the various blast impacts and their prediction models with a focus on empirical and machine learning methods. The details of the prediction methods for various blast impacts—including their applications, advantages, and limitations—are discussed. The literature reveals that the machine learning methods are better predictors compared to the empirical models. However, we observed that presently these ML models are mainly applied in academic research.

scholarly journals Email Spam Detection using Ensemble Methods

2019 ◽  
Vol 8 (3) ◽  
pp. 4148-4153
Keyword(s):  
Machine Learning ◽  
System Performance ◽  
Research Work ◽  
Ensemble Methods ◽  
Spam Detection ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Tree Algorithms ◽  
Email Spam ◽  
F Measure

The swiftly growth of spam email has escalated the need to upgrade the existing spam detection and filtration methods. There is the existence of several machine learning methods for the classification and detection of email spam but these lacks in some cases. In this research work ensemble methods are adapted to detect the email spam. The machine learning methods of Multinomial Naïve Bayes and J48 Decision Tree algorithms are considered and ensembled. The considered ensemble methods are bagging and boosting. The experimentation is conducted on the dataset of CSDMC2010 Spam corpus. The results for the considered dataset are evaluated using individual classifiers, bagging, and boosting ensemble approaches. The system performance is accessed in terms of precision, recall, f-measure, and accuracy. The experimental outcomes indicates the distinguish results for the detection of email spam using ensemble methods.

Machine Learning for Mass Spectrometry Data Analysis in Proteomics

2020 ◽  
Vol 17 ◽  
Author(s):  
Juntao Li ◽  
Kanglei Zhou ◽  
Bingyu Mu
Keyword(s):  
Machine Learning ◽  
Mass Spectrometry ◽  
Support Vector Machine ◽  
Data Analysis ◽  
Rapid Development ◽  
Point Of View ◽  
Mass Spectrometry Data ◽  
Support Vector ◽  
Learning Methods ◽  
Machine Learning Methods

: With the rapid development of high-throughput techniques, mass spectrometry has been widely used for largescale protein analysis. To search for the existing proteins, discover biomarkers, and diagnose and prognose diseases, machine learning methods are applied in mass spectrometry data analysis. This paper reviews the applications of five kinds of machine learning methods to mass spectrometry data analysis from an algorithmic point of view, including support vector machine, decision tree, random forest, naive Bayesian classifier and deep learning.

scholarly journals Modeling and interpreting hydrological responses of sustainable urban drainage systems with explainable machine learning methods

2020 ◽  
Author(s):  
Yang Yang ◽  
Ting Fong May Chui
Keyword(s):  
Machine Learning ◽  
Management Practices ◽  
Hydrological Processes ◽  
Rainfall Time Series ◽  
Hydrograph Separation ◽  
Learning Methods ◽  
Drainage Systems ◽  
Factors Affecting ◽  
Machine Learning Methods ◽  
Hydrological Responses

Abstract. Sustainable drainage systems (SuDS) are decentralized stormwater management practices that mimic the natural drainage processes. Their modeling is often challenged by insufficient data and unknown factors affecting the hydrological processes. This study uses machine learning methods to model directly the correlation between hydrological responses and rainfalls at fine temporal scales in two catchments of different sizes. A feature engineering method is developed to extract useful information from rainfall time series and is used in combination with a nested cross-validation procedure to derive high-quality models and to estimate their generalization errors. The SHAP method is adopted to explain the basis of each prediction, which is then used for estimating catchment response time and hydrograph separation. The explanations of the predictions provide valuable insights into the models’ behavior and the involved hydrological processes. Thus, interpreting machine learning models is found as a useful way to study catchment hydrology.

scholarly journals Potential for Vertical Heterogeneity Prediction in Reservoir Basing on Machine Learning Methods

Geofluids ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-12 ◽  
Author(s):  
Hongqing Song ◽  
Shuyi Du ◽  
Ruifei Wang ◽  
Jiulong Wang ◽  
Yuhe Wang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Neural Networks ◽  
Rapid Development ◽  
Petroleum Industry ◽  
Production Data ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Vertical Heterogeneity ◽  
Dynamic Production

With the rapid development of computer technology, some machine learning methods have begun to gradually integrate into the petroleum industry and have achieved some achievements, whether in conventional or unconventional reservoirs. This paper presents an alternative method to predict vertical heterogeneity of the reservoir utilizing various deep neural networks basing on dynamic production data. A numerical simulation technique was adopted to obtain the required dataset, which contains dynamic production data calculated under different heterogeneous reservoir conditions. Machine learning models were established through deep neural networks, which learn and capture the characteristics better between dynamic production data and reservoir heterogeneity, so as to invert the vertical permeability. On the basis of model validation, the results show that machine learning methods have excellent performance in predicting heterogeneity with the RMSE of 12.71 mD, which effectively estimated the permeability of the entire reservoir. Moreover, the overall AARD of the predictive result obtained by the CNN method was controlled at 11.51%, revealing the highest accuracy compared with BP and LSTM neural networks. And the permeability contrast, an important parameter to characterize heterogeneity, can be predicted precisely as well, with a derivation of below 10%. This study proposed a potential for vertical heterogeneity prediction in reservoir basing on machine learning methods.

scholarly journals Dataset segmentation considering the information about impact factors

2021 ◽  
pp. 29-38
Author(s):  
Iliya Lebedev
Keyword(s):  
Machine Learning ◽  
Machine Learning Algorithms ◽  
Impact Factors ◽  
Learning Methods ◽  
Factors Affecting ◽  
Machine Learning Methods ◽  
Multiple Data ◽  
Long Run ◽  
Data Value

Introduction: The application of machine learning methods involves the collection and processing of data which comes from the recording elements in the offline mode. Most models are trained on historical data and then used in forecasting, classification, search for influencing factors or impacts, and state analysis. In the long run, the data value ranges can change, affecting the quality of the classification algorithms and leading to the situation when the models should be constantly trained or readjusted taking into account the input data. Purpose: Development of a technique to improve the quality of machine learning algorithms in a dynamically changing and non-stationary environment where the data distribution can change over time. Methods: Splitting (segmentation) of multiple data based on the information about factors affecting the ranges of target variables. Results: A data segmentation technique has been proposed, based on taking into account the factors which affect the change in the data value ranges. Impact detection makes it possible to form samples based on the current and alleged situations. Using PowerSupply dataset as an example, the mass of data is split into subsets considering the effects of factors on the value ranges. The external factors and impacts are formalized based on production rules. The processing of the factors using the membership function (indicator function) is shown. The data sample is divided into a finite number of non-intersecting measurable subsets. Experimental values of the neural network loss function are shown for the proposed technique on the selected dataset. Qualitative indicators (Accuracy, AUC, F-measure) of the classification for various classifiers are presented. Practical relevance: The results can be used in the development of classification models of machine learning methods. The proposed technique can improve the classification quality in dynamically changing conditions of the functioning.

scholarly journals PREDICTING TIME BEFORE THE NEXT ORDER IN THE ONLINE STORE, BASED ON MACHINE LEARNING METHODS

2020 ◽  
Author(s):  
Olena Piskunova ◽  
◽  
Rostyslav Klochko ◽  
Keyword(s):  
Machine Learning ◽  
Comparative Analysis ◽  
Random Forest ◽  
Rapid Development ◽  
Confusion Matrix ◽  
Individual Characteristics ◽  
Support Vector ◽  
Learning Methods ◽  
Online Store ◽  
Machine Learning Methods

Due to the rapid development of e-commerce and increased competition in the retail market of Ukraine, companies are forced to look for new ways to grow their business. One of the options is to optimize business processes, in particular to increase the efficiency of marketing activities. Predicting consumer behavior is one of the most effective methods of optimizing marketing budgets by building processes based on the individual characteristics of each client. The aim of the study was to predict the behavior of online store customers, namely the time before the next order, based on machine learning methods and a comparative analysis of the effectiveness of different modeling algorithms. Five classification algorithms were implemented: linear discriminant analysis, сlassification and regression trees, random forest, support vector machine, k - nearest neighbors and comparative analysis of their efficiency was performed. Given the peculiarities of customer behavior for forecasting time to the next order, it is proposed to consider the following time intervals in the future when the customer makes the next order: up to two months, two to six months, six to fifteen months, and without order. Predicting such intervals allows us to identify customers who are more likely to make the next purchase and focus our advertising budgets on them, or build a customer experience management strategy: activate customers who have left, offer discounts to customers who are going to leave. Peculiarities of classification models quality assessment on the basis of the “confusion matrix” according to the forecasting accuracy indicators “Accuracy”, “F1”, “Recall” and “Precision” is considered. The study allowed us to give preference to the model of classification "random forest". A tenfold cross-validation was used to improve the quality of the simulation. The weighted accuracy of “F1” in the groups “Up to two months” and “two-six months” reached 62.5% and 64.1%, respectively. The developed model should reduce the influence of the human factor on the decision-making process in the construction of marketing strategies.

scholarly journals The Technique of Multi-aspect Evaluation and Categorization of Malicious Information Objects on the Internet

2019 ◽  
Vol 5 (3) ◽  
pp. 58-65 ◽  
Author(s):  
A. Branitskiy ◽  
I. Saenko
Keyword(s):  
Machine Learning ◽  
Information Technologies ◽  
Information Sources ◽  
Rapid Development ◽  
The Internet ◽  
Web Pages ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Multi Level ◽  
Information Objects

Under the influence of rapid development in the sphere of information technologies, rises the challenge related to detection of malicious information sources on the Internet. To solve this we can use machine learning methods as one of the most popular and powerful tools designed to identify dependencies between input (observed) data and output (desired) results. This article presents a methodology which is aimed at multi-level processing of input data about malicious information objects on the Internet and providing their multi-aspect assessment and categorization using machine learning methods. The purpose of the investigation is to improve the efficiency of the detecting process of malicious information on the Internet using the examples of Web-pages classification.

Sign in / Sign up

Export Citation Format

Share Document