scholarly journals The Development of a Quantitative Precipitation Forecast Correction Technique Based on Machine Learning for Hydrological Applications

Atmosphere ◽  
2020 ◽  
Vol 11 (1) ◽  
pp. 111 ◽  
Author(s):  
Chul-Min Ko ◽  
Yeong Yun Jeong ◽  
Young-Mi Lee ◽  
Byung-Sik Kim
Keyword(s):  
Machine Learning ◽  
Heavy Rainfall ◽  
Extreme Rainfall ◽  
Machine Learning Techniques ◽  
Precipitation Forecast ◽  
Machine Learning Technique ◽  
Rainfall Forecast ◽  
Quantitative Precipitation Forecast ◽  
Correction Technique ◽  
Learning Technique

This study aimed to enhance the accuracy of extreme rainfall forecast, using a machine learning technique for forecasting hydrological impact. In this study, machine learning with XGBoost technique was applied for correcting the quantitative precipitation forecast (QPF) provided by the Korea Meteorological Administration (KMA) to develop a hydrological quantitative precipitation forecast (HQPF) for flood inundation modeling. The performance of machine learning techniques for HQPF production was evaluated with a focus on two cases: one for heavy rainfall events in Seoul and the other for heavy rainfall accompanied by Typhoon Kong-rey (1825). This study calculated the well-known statistical metrics to compare the error derived from QPF-based rainfall and HQPF-based rainfall against the observational data from the four sites. For the heavy rainfall case in Seoul, the mean absolute errors (MAE) of the four sites, i.e., Nowon, Jungnang, Dobong, and Gangnam, were 18.6 mm/3 h, 19.4 mm/3 h, 48.7 mm/3 h, and 19.1 mm/3 h for QPF and 13.6 mm/3 h, 14.2 mm/3 h, 33.3 mm/3 h, and 12.0 mm/3 h for HQPF, respectively. These results clearly indicate that the machine learning technique is able to improve the forecasting performance for localized rainfall. In addition, the HQPF-based rainfall shows better performance in capturing the peak rainfall amount and spatial pattern. Therefore, it is considered that the HQPF can be helpful to improve the accuracy of intense rainfall forecast, which is subsequently beneficial for forecasting floods and their hydrological impacts.

scholarly journals Using Betts-Miller-Janjic convective parameterization scheme in H14-31 model to forecast heavy rainfall in Vietnam

2007 ◽  
Vol 29 (2) ◽  
pp. 83-97
Author(s):  
Vu Thanh Hang ◽  
Kieu Thi Xin
Keyword(s):  
Heavy Rainfall ◽  
Precipitation Forecast ◽  
Parameterization Scheme ◽  
Model Resolution ◽  
Convective Parameterization ◽  
Rainfall Forecast ◽  
Quantitative Precipitation Forecast ◽  
Convective Adjustment ◽  
Convective Parameterization Scheme ◽  
The Tropics

According to Krishnamurti, improvements of physical parameterizations will mainly affect simulations for the tropics [10]. The study of William A. Gallus Jr. showed that the higher the model resolution and more detailed convective parameterizations, the better the skill in quantitative precipitation forecast (QPF) in general [16]. The quality of precipitation forecast is so sensitive to convective parameterization scheme (CPS) used in the model as well as model resolution. The fact shows that for high resolution regional model like H14-31 CPS based on low-level moisture convergence as Tiedtke did not give good heavy rainfall forecast in Vietnam. In this paper we used the scheme of Betts-Miller-Janjic (BMJ) based on the convective adjustment toward tropical observationally structures in reality instead of Tiedtke in Hl4-31. Statistical verification results and verification using CRA method of Hl4-31 of two CPSs for seperated cases and for three rain seasons (2003-2005) shows that heavy rainfall forecast of Hl4-31/BMJ is better than one of H14-31/TK for Vietnam-South China Sea. CRA verification also shows that it is possible to say that heavy rainfall forecast skill of l-I14-31/BMJ in tropics is nearly similar to the skill of LAPS of Australia.

scholarly journals Classification of specialty coffees using machine learning techniques

2021 ◽  
Vol 10 (5) ◽  
pp. e13110514732
Author(s):  
Paulo César Ossani ◽  
Diogo Francisco Rossoni ◽  
Marcelo Ângelo Cirillo ◽  
Flávio Meira Borém
Keyword(s):  
Machine Learning ◽  
Supervised Classification ◽  
Consumer Acceptance ◽  
Machine Learning Techniques ◽  
Machine Learning Technique ◽  
Learning Techniques ◽  
Learning Technique ◽  
Evaluation Test ◽  
New Methodologies

Specialty coffees have a big importance in the economic scenario, and its sensory quality is appreciated by the productive sector and by the market. Researches have been constantly carried out in the search for better blends in order to add value and differentiate prices according to the product quality. To accomplish that, new methodologies must be explored, taking into consideration factors that might differentiate the particularities of each consumer and/or product. Thus, this article suggests the use of the machine learning technique in the construction of supervised classification and identification models. In a sensory evaluation test for consumer acceptance using four classes of specialty coffees, applied to four groups of trained and untrained consumers, features such as flavor, body, sweetness and general grade were evaluated. The use of machine learning is viable because it allows the classification and identification of specialty coffees produced in different altitudes and different processing methods.

scholarly journals Machine learning Techniques for Hotel Online Reputation

2019 ◽  
Vol 8 (9) ◽  
pp. 3004-3007
Keyword(s):  
Machine Learning ◽  
Online Reviews ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Machine Learning Technique ◽  
Learning Techniques ◽  
Learning Technique ◽  
Learning Reinforcement ◽  
Online Reputation ◽  
Fake Reviews

Now days when someone decide to book a hotel, previous online reviews of the hotels play a major role in determining the best hotel within the budget of the customer. Previous Online reviews are the most important motivation for the information that are used to analyse public opinion. Because of the high impact of the reviews on business, hotel owners are always highly concerned and focused about the customer feedback and past online reviews. But all reviews are not true and trustworthy, sometime few people may intentionally generate the fake reviews to make some hotel famous of to defame. Therefore it is essential to develop and propose the techniques for analysis of reviews. With the help of various machine learning techniques viz. Supervised machine learning technique, Text mining, Unsupervised machine learning technique, Semi-supervised learning, Reinforcement learning etc we may detect the fake reviews. This paper gives some notions of using machine learning techniques in analysis of past online reviews of hotels, Based on the observation it also suggest the optimal machine learning technique for a particular situation

scholarly journals Application of Machine Learning in Rhinology: A State of the Art Review

2020 ◽  
Vol 63 (8) ◽  
pp. 341-349
Author(s):  
Myeong Sang Yu
Keyword(s):  
Machine Learning ◽  
Big Data ◽  
Deep Learning ◽  
Medical Records ◽  
State Of The Art ◽  
Machine Learning Techniques ◽  
Machine Learning Technique ◽  
Learning Techniques ◽  
Learning Technique ◽  
Machine Learning Applications

The revolutionary development of artificial intelligence (AI) such as machine learning and deep learning have been one of the most important technology in many parts of industry, and also enhance huge changes in health care. The big data obtained from electrical medical records and digitalized images accelerated the application of AI technologies in medical fields. Machine learning techniques can deal with the complexity of big data which is difficult to apply traditional statistics. Recently, the deep learning techniques including convolutional neural network have been considered as a promising machine learning technique in medical imaging applications. In the era of precision medicine, otolaryngologists need to understand the potentialities, pitfalls and limitations of AI technology, and try to find opportunities to collaborate with data scientists. This article briefly introduce the basic concepts of machine learning and its techniques, and reviewed the current works on machine learning applications in the field of otolaryngology and rhinology.

scholarly journals The effects of applying filters on EEG signals for classifying developers’ code comprehension

2021 ◽  
Vol 19 (6) ◽  
pp. 584-602
Author(s):  
Lucian Jose Gonçales ◽  
Kleinner Farias ◽  
Lucas Kupssinskü ◽  
Matheus Segalotto
Keyword(s):  
Machine Learning ◽  
Software Engineering ◽  
Random Forest ◽  
Random Forest Classifier ◽  
Machine Learning Techniques ◽  
Eeg Signals ◽  
Machine Learning Technique ◽  
Learning Techniques ◽  
Learning Technique ◽  
F Measure

EEG signals are a relevant indicator for measuring aspects related to human factors in Software Engineering. EEG is used in software engineering to train machine learning techniques for a wide range of applications, including classifying task difficulty, and developers’ level of experience. The EEG signal contains noise such as abnormal readings, electrical interference, and eye movements, which are usually not of interest to the analysis, and therefore contribute to the lack of precision of the machine learning techniques. However, research in software engineering has not evidenced the effectiveness when applying these filters on EEG signals. The objective of this work is to analyze the effectiveness of filters on EEG signals in the software engineering context. As literature did not focus on the classification of developers’ code comprehension, this study focuses on the analysis of the effectiveness of applying EEG filters for training a machine learning technique to classify developers' code comprehension. A Random Forest (RF) machine learning technique was trained with filtered EEG signals to classify the developers' code comprehension. This study also trained another random forest classifier with unfiltered EEG data. Both models were trained using 10-fold cross-validation. This work measures the classifiers' effectiveness using the f-measure metric. This work used the t-test, Wilcoxon, and U Mann Whitney to analyze the difference in the effectiveness measures (f-measure) between the classifier trained with filtered EEG and the classifier trained with unfiltered EEG. The tests pointed out that there is a significant difference after applying EEG filters to classify developers' code comprehension with the random forest classifier. The conclusion is that the use of EEG filters significantly improves the effectivity to classify code comprehension using the random forest technique.

Abstract 349: Prognostication for Out-of-Hospital Cardiogenic Cardiac Arrest Patients Using Advanced Machine Learning Technique

Circulation ◽  
2018 ◽  
Vol 138 (Suppl_2) ◽  
Author(s):  
Tomohisa Seki ◽  
Tomoyoshi Tamura ◽  
Masaru Suzuki
Keyword(s):  
Machine Learning ◽  
Cardiac Arrest ◽  
Study Data ◽  
Machine Learning Techniques ◽  
Operating Characteristics ◽  
Training Set ◽  
Machine Learning Technique ◽  
Test Set ◽  
Medical Institutions ◽  
Learning Technique

Introduction and Objective: Early prognostication for cardiogenic out-of-hospital cardiac arrest (OHCA) patients remain challenging. Recently, advanced machine learning techniques have been employed for clinical diagnosis and prognostication for various conditions. Therefore, in this study, we attempted to establish a prognostication model for cardiogenic OHCA using an advanced machine learning technique. Methods and Results: Data of a prospective multi-center cohort study of OHCA patients transported by an ambulance to 67 medical institutions in Kanto area of Japan between January 2012 and March 2013 was used in this study. Data for cardiogenic OHCA patients aged ≥18 years were retrieved and patients were grouped according to the time of calls for ambulances (training set: between January 1, 2012 and December 12, 2012; test set: between January 1, 2013 and March 31, 2013). From among 421 variables observed during the period between calls for ambulances and initial in-hospital treatments of cardiogenic OHCA, 38 prehospital factors or 56 prehospital factors and initial in-hospital factors were used for prognostication, respectively. Prognostication models for 1-year survival were established with random forest method, an advanced machine learning method that aggregates a series of decision trees for classification and regression. After 10-fold internal cross validation in the training set, prognostication models were validated using test set. Area under the receiver operating characteristics curve (AUC) was used to evaluate the prediction performance of models. Prognostication models trained with 38 variables or 56 variables for 1-year survival showed AUC values of 0.93±0.01 and 0.95±0.01, respectively. Conclusions: Prognostication models trained with advanced machine learning technique showed favorable prediction capability for 1-year survival of cardiogenic OHCA. These results indicate that an advanced machine learning technique can be applicable to establish early prognostication model for cardiogenic OHCA.

scholarly journals Machine learning modelling for predicting soil liquefaction susceptibility

2011 ◽  
Vol 11 (1) ◽  
pp. 1-9 ◽  
Author(s):  
P. Samui ◽  
T. G. Sitharam
Keyword(s):  
Machine Learning ◽  
Standard Penetration Test ◽  
Soil Liquefaction ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Machine Learning Technique ◽  
Ground Acceleration ◽  
Cyclic Stress Ratio ◽  
Liquefaction Susceptibility ◽  
Learning Technique

Abstract. This study describes two machine learning techniques applied to predict liquefaction susceptibility of soil based on the standard penetration test (SPT) data from the 1999 Chi-Chi, Taiwan earthquake. The first machine learning technique which uses Artificial Neural Network (ANN) based on multi-layer perceptions (MLP) that are trained with Levenberg-Marquardt backpropagation algorithm. The second machine learning technique uses the Support Vector machine (SVM) that is firmly based on the theory of statistical learning theory, uses classification technique. ANN and SVM have been developed to predict liquefaction susceptibility using corrected SPT [(N1)60] and cyclic stress ratio (CSR). Further, an attempt has been made to simplify the models, requiring only the two parameters [(N1)60 and peck ground acceleration (amax/g)], for the prediction of liquefaction susceptibility. The developed ANN and SVM models have also been applied to different case histories available globally. The paper also highlights the capability of the SVM over the ANN models.

scholarly journals Predicting On-Axis Rotorcraft Dynamic Responses Using Machine Learning Techniques

2019 ◽  
Author(s):  
Ryan Jackson ◽  
Michael Jump ◽  
Peter Green
Keyword(s):  
Machine Learning ◽  
Dynamic Response ◽  
Real Time ◽  
Computational Cost ◽  
Dynamic Responses ◽  
Training Data ◽  
Machine Learning Techniques ◽  
Machine Learning Technique ◽  
Learning Technique ◽  
Gp Model

Physical-law based models are widely utilized in the aerospace industry. One such use is to provide flight dynamics models for use in flight simulators. For human-in-the-loop use, such simulators must run in real-time. Due to the complex physics of rotorcraft flight, to meet this real-time requirement, simplifications to the underlying physics sometimes have to be applied to the model, leading to model response errors in the predictions compared to the real vehicle. This study investigated whether a machine-learning technique could be employed to provide rotorcraft dynamic response predictions, with the ultimate aim of this model taking over when the physics-based model's accuracy degrades. In the current work, a machine-learning technique was employed to train a model to predict the dynamic response of a rotorcraft. Machine learning was facilitated using a Gaussian Process (GP) non-linear autoregressive model, which predicted the on-axis pitch rate, roll rate, yaw rate and heave responses of a Bo105 rotorcraft. A variational sparse GP model was then developed to reduce the computational cost of implementing the approach on large data sets. It was found that both of the GP models were able to provide accurate on-axis response predictions, particularly when the input contained all four control inceptors and one lagged on-axis response term. The predictions made showed improvement compared to a corresponding physics-based model. The reduction of training data to one-third (rotational axes) or one-half (heave axis) resulted in only minor degradation of the GP model predictions.

scholarly journals An Algorithm for Deciding the Online Reputation of Hotels

2019 ◽  
Vol 8 (9) ◽  
pp. 2998-3003
Keyword(s):  
Machine Learning ◽  
Word Of Mouth ◽  
Online Reviews ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Machine Learning Technique ◽  
Detection Model ◽  
The People ◽  
Learning Technique ◽  
Online Reputation

After revolution in cell phone industry expansion and offering of promotional data packs by telecom companies like Reliance Jio, Airtel, Idea, Spice etc accessibility to the Internet has become very easy for the people. maximum people are now connected through social media viz. facebook, twitter, instagram etc. People are sharing their best and worst experiences for any brand. Various online review sites like Treebo, Yelp, Google Maps, and Tripadvisor OYO, Makemytrip, goibibo etc are used as an important source for the success of hotel businesses. Word of mouth has always been a powerful tool for marketing a business, Online reviews are today’s word of mouth marketing, and these can make or break your business; In this research paper it is proposed for analyzing online reviews about hotels our algorithm must able to detect and analyzing fake reviewers based on user, tweet, timestamp, IP, collision and manipulation concept as well as to develop optimal model (based on group theory) for detecting fake reviewers, Improvement in enhancing sentimental analysis and the review detection model which can be implemented on all positive or all negative reviews, also the algorithm must able to identify the best fit of four machine learning techniques: (supervised machine technique technique, text mining technique , support vector machine learning technique and Naïve bayes machine learning technique) for specify and verify the different parameters of classification of reviews. Algorithm must able to Quantify the results of above techniques and extract the parameters to analyze the Genuinity of reviews based on Location, Security, Price, Quality, Ambiance etc.

scholarly journals A Comparative Analysis and Prediction over Bitcoin Price Using Machine Learning Technique

2021 ◽  
pp. 31-41
Author(s):  
Meenu Gupta ◽  
◽  
◽  
Riya Srivastava
Keyword(s):  
Machine Learning ◽  
Machine Learning Techniques ◽  
Good Time ◽  
Time Duration ◽  
Research Papers ◽  
Machine Learning Technique ◽  
Advantages And Disadvantages ◽  
Learning Techniques ◽  
Learning Technique ◽  
Free People

Bitcoin is one of the primary computerized monetary forms to utilize peer innovation to work with moment installments. The free people and organizations who own the overseeing figuring control and take part in the bitcoin network—bitcoin miners— are accountable for preparing the exchanges on the blockchain and are persuaded by remunerations (the arrival of new bitcoin) and exchange charges paid in bitcoin. These excavators can be considered as the decentralized authority implementing the believability of the bitcoin network. New bitcoin is delivered to the excavators at a fixed yet occasionally declining rate. There is just 21 million bitcoin that can be mine altogether. As of January 30, 2021, there are around 18,614,806 bitcoin in presence and 2,385,193 bitcoin left to be mined. This paper will predict the nature of bitcoin price because, according to the reports of the past few years. The year 2020-present appeared to be a good time for bitcoin because, in this time duration, bitcoin has seen huge ups and downs. This paper will use various Machine Learning Techniques for the predictive analysis of bitcoin to accurately predict the price's nature. As the price of bitcoin depends upon various factors and these factors directly affect the price, i.e., multiple factors of bitcoin are dependent on each other. After analyzing the results from multiple research papers and review papers, we discovered each algorithm has its advantages and disadvantages while predicting the bitcoin value. Keeping in mind all the findings, we will find algorithms that predict the bitcoin price accurately and without fewer disadvantages. So, if we go as per assumptions, regression would be the best choice for predicting the bitcoin value, but there are others algorithms also. So, in this paper, we will see the results of the multiple algorithms and then choose the correct algorithm after analyzing the results of all the implemented algorithms. This paper also includes the implementation of the comparison charts with each algorithm so that it will be easy to analyze the findings of each algorithm.

Sign in / Sign up

Export Citation Format

Share Document