scholarly journals Rainrate Estimation from FY-4A Cloud Top Temperature for Mesoscale Convective Systems by Using Machine Learning Algorithm

2021 ◽  
Vol 13 (16) ◽  
pp. 3273
Author(s):  
Ping Lao ◽  
Qi Liu ◽  
Yuhao Ding ◽  
Yu Wang ◽  
Yuan Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Mesoscale Convective Systems ◽  
Support Vector ◽  
Positive Contribution ◽  
Machine Learning Technique ◽  
Convective Systems ◽  
Learning Technique ◽  
Mesoscale Convective ◽  
Cloud Top Temperature

Satellite rainrate estimation is a great challenge, especially in mesoscale convective systems (MCSs), which is mainly due to the absence of a direct physical connection between observable cloud parameters and surface rainrate. The machine learning technique was employed in this study to estimate rainrate in the MCS domain via using cloud top temperature (CTT) derived from a geostationary satellite. Five kinds of machine learning models were investigated, i.e., polynomial regression, support vector machine, decision tree, random forest, and multilayer perceptron, and the precipitation of Climate Prediction Center morphing technique (CMORPH) was used as the reference. A total of 31 CTT related features were designed to be the potential inputs for training an algorithm, and they were all proved to have a positive contribution in modulating the algorithm. Random forest (RF) shows the best performance among the five kinds of models. By combining the classification and regression schemes of the RF model, an RF-based hybrid algorithm was proposed first to discriminate the rainy pixel and then estimate its rainrate. For the MCS samples considered in this study, such an algorithm generates the best estimation, and its accuracy is definitely higher than the operational precipitation product of FY-4A. These results demonstrate the promising feasibility of applying a machine learning technique to solve the satellite precipitation retrieval problem.

scholarly journals A Method for Identifying Midlatitude Mesoscale Convective Systems in Radar Mosaics. Part I: Segmentation and Classification

2018 ◽  
Vol 57 (7) ◽  
pp. 1575-1598 ◽  
Author(s):  
Alex M. Haberlie ◽  
Walker S. Ashley
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Probability Of Detection ◽  
Machine Learning Algorithms ◽  
Mesoscale Convective Systems ◽  
Gradient Boosting ◽  
Moist Convection ◽  
Probabilistic Prediction ◽  
Convective Systems ◽  
Mesoscale Convective

AbstractThis research evaluates the ability of image-processing and select machine-learning algorithms to identify midlatitude mesoscale convective systems (MCSs) in radar-reflectivity images for the conterminous United States. The process used in this study is composed of two parts: segmentation and classification. Segmentation is performed by identifying contiguous or semicontiguous regions of deep, moist convection that are organized on a horizontal scale of at least 100 km. The second part, classification, is performed by first compiling a database of thousands of precipitation clusters and then subjectively assigning each sample one of the following labels: 1) midlatitude MCS, 2) unorganized convective cluster, 3) tropical system, 4) synoptic system, or 5) ground clutter and/or noise. The attributes of each sample, along with their assigned label, are used to train three machine-learning algorithms: random forest, gradient boosting, and “XGBoost.” Results using a testing dataset suggest that the algorithms can distinguish between MCS and non-MCS samples with a high probability of detection and low probability of false detection. Further, the trained algorithm predictions are well calibrated, allowing reliable probabilistic classification. The utility of this two-step procedure is illustrated by generating spatial frequency maps of automatically identified precipitation clusters that are stratified by using various reflectivity and probabilistic prediction thresholds. These results suggest that machine learning can add value by limiting the amount of false-positive (non-MCS) samples that are not removed by segmentation alone.

scholarly journals An Ensemble Machine Learning Technique for Functional Requirement Classification

Symmetry ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 1601
Author(s):  
Nouf Rahimi ◽  
Fathy Eassa ◽  
Lamiaa Elrefaei
Keyword(s):  
Machine Learning ◽  
Requirement Engineering ◽  
Support Vector ◽  
Functional Requirement ◽  
Machine Learning Technique ◽  
Engineering Software ◽  
Ensemble Machine Learning ◽  
Learning Technique ◽  
Combined Models

In Requirement Engineering, software requirements are classified into two main categories: Functional Requirement (FR) and Non-Functional Requirement (NFR). FR describes user and system goals. NFR includes all constraints on services and functions. Deeper classification of those two categories facilitates the software development process. There are many techniques for classifying FR; some of them are Machine Learning (ML) techniques, and others are traditional. To date, the classification accuracy has not been satisfactory. In this paper, we introduce a new ensemble ML technique for classifying FR statements to improve their accuracy and availability. This technique combines different ML models and uses enhanced accuracy as a weight in the weighted ensemble voting approach. The five combined models are Naïve Bayes, Support Vector Machine (SVM), Decision Tree, Logistic Regression, and Support Vector Classification (SVC). The technique was implemented, trained, and tested using a collected dataset. The accuracy of classifying FR was 99.45%, and the required time was 0.7 s.

scholarly journals Mode Choice Prediction using Machine Learning Technique for A Door-to-Door Journey in Kuantan City

Mekatronika ◽  
2020 ◽  
Vol 2 (1) ◽  
pp. 73-78
Author(s):  
Nur Fahriza Mohd Ali ◽  
Ahmad Farhan Mohd Sadullah ◽  
Anwar P.P. Abdul Majeed ◽  
Mohd Azraai Mohd Razman ◽  
Rabiu Muazu Musa
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Mode Choice ◽  
Learning Models ◽  
Machine Learning Technique ◽  
Travel Mode Choice ◽  
Testing Data ◽  
Learning Technique ◽  
The City ◽  
Machine Learning Models

A door-to-door journey in a public transportation system is a notable concept that is practically being promoted among users to consider public transport as an important alternative. The door-to-door journey will integrate the travel segments starting from home to destination, including all visible amenities. Users’ preferences on the time travel of these key segments are necessary to be understood. In this case, Machine Learning technique has been seen as a robust computational advancement to forecast their travel mode choice. However, the most convenient model as the best predictor is still questionable. To address this issue, we employed some pre-eminent machine learning models, specifically Random Forest (RF), Naïve Bayes (NB), Logistic Regression (LR), k-Nearest Neighbor (kNN) as well as Support Vector Machine (SVM), to compare their travel mode choice prediction performance of users in the city of Kuantan. The data collection was conducted in Kuantan City via Revealed/Stated Preferences (RPSP) Survey between 8:00 AM to 5:00 PM on weekdays. The data collected was split into a ratio of 80:20 for training and testing before evaluating them between the aforesaid models. The results depicted that the Random Forest could provide satisfactory classification accuracies for both training and testing data up to 68.3% and 61.3%, respectively, compared to the other evaluated machine learning models. In summary, Random Forest provides a good result in the training and testing data and is considered as the best predictor in this research to forecast users’ mode choice in the city of Kuantan.

scholarly journals Modeling surface roughness of point robot laser hardening, with emphasis on the surface

Politehnika ◽  
2021 ◽  
Vol 5 (1) ◽  
pp. 6-9
Author(s):  
Matej Babič
Keyword(s):  
Machine Learning ◽  
Surface Roughness ◽  
Intelligent System ◽  
Future Trend ◽  
Support Vector ◽  
Machine Learning Technique ◽  
Wear And Corrosion ◽  
Learning Technique ◽  
Wear And Corrosion Resistance ◽  
Laser Cell

The topic of Machine Learning is so popular that it is not only the future trend, but also the money tide. Machine learning technique and intelligent system methods are very popular in mechanical engineering. Robot laser surface hardening is one of the most promising techniques for surface modification of the microstructure of a material to improve wear and corrosion resistance. For predicting the surface roughness of the hardened specimens, the support vector machine and multiple regression is used. The aim of this paper is to present modeling roughness of point robot laser hardened specimens with different parameters of robot laser cell.

Density Based Clustering with Integrated One-Class SVM for Noise Reduction

2017 ◽  
Vol 6 (3) ◽  
pp. 199
Author(s):  
K. Nafees Ahmed ◽  
T. Abdul Razak
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Data Analysis ◽  
Noise Reduction ◽  
Spatial Clustering ◽  
Support Vector ◽  
Machine Learning Technique ◽  
Learning Classifier ◽  
Density Based Clustering ◽  
Learning Technique

<p>Information extraction from data is one of the key necessities for data analysis. Unsupervised nature of data leads to complex computational methods for analysis. This paper presents a density based spatial clustering technique integrated with one-class Support Vector Machine (SVM), a machine learning technique for noise reduction, a modified variant of DBSCAN called Noise Reduced DBSCAN (NRDBSCAN). Analysis of DBSCAN exhibits its major requirement of accurate thresholds, absence of which yields suboptimal results. However, identifying accurate threshold settings is unattainable. Noise is one of the major side-effects of the threshold gap. The proposed work reduces noise by integrating a machine learning classifier into the operation structure of DBSCAN. The Experimental results indicate high homogeneity levels in the clustering process.</p>

scholarly journals Validity of the EQ-5D-5L questionnaire among the general population of Poland

2020 ◽  
Author(s):  
Katarzyna Młyńczak ◽  
Dominik Golicki
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
General Population ◽  
Construct Validity ◽  
Psychometric Properties ◽  
Machine Learning Technique ◽  
Validity Assessment ◽  
Novel Approach ◽  
Learning Technique ◽  
Eq Vas

Abstract Purpose We aim to compare the psychometric properties of the EQ-5D-5L questionnaire with the EQ-5D-3L version and EQ VAS, based on a survey conducted in a sample representing the general adult population of Poland. Methods The survey comprised health-related quality of life (HRQoL) questionnaires: EQ-5D-5L, EQ VAS, SF-12 and EQ-5D-3L, together with demographic and socio-economic characteristics items. The EQ-5D index values were estimated based on a directly measured value set for Poland. The following psychometric properties were analysed: feasibility, distribution of responses, redistribution from EQ-5D-3L to EQ-5D-5L, inconsistencies, ceiling effects, informativity power and construct validity. We proposed a novel approach to the construct validity assessment, based on the use of a machine learning technique known as the random forest algorithm. Results From March to June 2014, 3978 subjects (aged 18–87, 53.2% female) were surveyed. The EQ-5D-5L questionnaire had a lower ceiling effect compared to EQ-5D-3L (38.0% vs 46.6%). Redistribution from EQ-5D-3L to EQ-5D-5L was similar for each dimension, and the mean inconsistency did not exceed 5%. The results of known-groups validation confirmed the hypothesis concerning the relationship between the EQ-5D index values and age, sex and occurrence of diabetes. Conclusions The EQ-5D-5L, in comparison with its EQ-5D-3L equivalent, showed similar or better psychometric properties within the general population of a country. We assessed the construct validity of the questionnaire with a novel approach that was based on a machine learning technique known as the random forest algorithm.

Sign in / Sign up

Export Citation Format

Share Document