Detecting and Mitigating Adversarial Examples in Regression Tasks: A Photovoltaic Power Generation Forecasting Case Study

With data collected by Internet of Things sensors, deep learning (DL) models can forecast the generation capacity of photovoltaic (PV) power plants. This functionality is especially relevant for PV power operators and users as PV plants exhibit irregular behavior related to environmental conditions. However, DL models are vulnerable to adversarial examples, which may lead to increased predictive error and wrong operational decisions. This work proposes a new scheme to detect adversarial examples and mitigate their impact on DL forecasting models. This approach is based on one-class classifiers and features extracted from the data inputted to the forecasting models. Tests were performed using data collected from a real-world PV power plant along with adversarial samples generated by the Fast Gradient Sign Method under multiple attack patterns and magnitudes. One-class Support Vector Machine and Local Outlier Factor were evaluated as detectors of attacks to Long-Short Term Memory and Temporal Convolutional Network forecasting models. According to the results, the proposed scheme showed a high capability of detecting adversarial samples with an average F1-score close to 90%. Moreover, the detection and mitigation approach strongly reduced the prediction error increase caused by adversarial samples.

Download Full-text

Flood Stage Forecasting Using Machine-Learning Methods: A Case Study on the Parma River (Italy)

Water ◽

10.3390/w13121612 ◽

2021 ◽

Vol 13 (12) ◽

pp. 1612

Author(s):

Susanna Dazzi ◽

Renato Vacondio ◽

Paolo Mignosa

Keyword(s):

Machine Learning ◽

Short Term Memory ◽

Computational Time ◽

Support Vector ◽

Lead Times ◽

Efficiency Coefficient ◽

Forecast Horizon ◽

Training Time ◽

Forecasting System

Real-time river flood forecasting models can be useful for issuing flood alerts and reducing or preventing inundations. To this end, machine-learning (ML) methods are becoming increasingly popular thanks to their low computational requirements and to their reliance on observed data only. This work aimed to evaluate the ML models’ capability of predicting flood stages at a critical gauge station, using mainly upstream stage observations, though downstream levels should also be included to consider backwater, if present. The case study selected for this analysis was the lower stretch of the Parma River (Italy), and the forecast horizon was extended up to 9 h. The performances of three ML algorithms, namely Support Vector Regression (SVR), MultiLayer Perceptron (MLP), and Long Short-term Memory (LSTM), were compared herein in terms of accuracy and computational time. Up to 6 h ahead, all models provided sufficiently accurate predictions for practical purposes (e.g., Root Mean Square Error < 15 cm, and Nash-Sutcliffe Efficiency coefficient > 0.99), while peak levels were poorly predicted for longer lead times. Moreover, the results suggest that the LSTM model, despite requiring the longest training time, is the most robust and accurate in predicting peak values, and it should be preferred for setting up an operational forecasting system.

Download Full-text

A Feature-Based Analysis for Time-Series Classification of COVID-19 Incidence in Chile: A Case Study

Applied Sciences ◽

10.3390/app11157080 ◽

2021 ◽

Vol 11 (15) ◽

pp. 7080

Author(s):

Christopher Flores ◽

Carla Taramasco ◽

Maria Elena Lagos ◽

Carla Rimassa ◽

Rosa Figueroa

Keyword(s):

Short Term Memory ◽

National Level ◽

Point Of View ◽

Support Vector ◽

Mechanical Ventilators ◽

Feature Based ◽

Risk Categories ◽

Better Than

The 2019 Coronavirus disease (COVID-19) pandemic is a current challenge for the world’s health systems aiming to control this disease. From an epidemiological point of view, the control of the incidence of this disease requires an understanding of the influence of the variables describing a population. This research aims to predict the COVID-19 incidence in three risk categories using two types of machine learning models, together with an analysis of the relative importance of the available features in predicting the COVID-19 incidence in the Chilean urban commune of Concepción. The classification results indicate that the ConvLSTM (Convolutional Long Short-Term Memory) classifier performed better than the SVM (Support Vector Machine), with results between 93% and 96% in terms of accuracy (ACC) and F-measure (F1) metrics. In addition, when considering each one of the regional and national features as well as the communal features (DEATHS and MOBILITY), it was observed that at the regional level the CRITICAL BED OCCUPANCY and PATIENTS IN ICU features positively contributed to the performance of the classifiers, while at the national level the features that most impacted the performance of the SVM and ConvLSTM were those related to the type of hospitalization of patients and the use of mechanical ventilators.

Download Full-text

On modeling player fitness in training for team sports with application to professional rugby

International Journal of Sports Science & Coaching ◽

10.1177/1747954117694736 ◽

2017 ◽

Vol 12 (2) ◽

pp. 183-193 ◽

Cited By ~ 2

Author(s):

Matthew Revie ◽

Kevin J Wilson ◽

Rob Holdsworth ◽

Stuart Yule

Keyword(s):

Support Vector Machines ◽

Team Sports ◽

Professional Sports ◽

Support Vector ◽

Performance Models ◽

Sports Teams ◽

Vector Machines ◽

Using Data ◽

Explicit Measures

It is increasingly important for professional sports teams to monitor player fitness in order to optimize performance. Models have been put forward linking fitness in training to performance in competition but rely on regular measurements of player fitness. As formal tests for measuring player fitness are typically time-consuming and inconvenient, measurements are taken infrequently. As such, it may be challenging to accurately predict performance in competition as player fitness is unknown. Alternatively, other data, such as how the players are feeling, may be measured more regularly. This data, however, may be biased as players may answer the questions differently and these differences may dominate the data. Linear mixed methods and support vector machines were used to estimate player fitness from available covariates at times when explicit measures of fitness were unavailable. Using data provided by a professional rugby club, a case study was used to illustrate the application and value of these models. Both models performed well with R2 values ranging from 60% to 85%, demonstrating that the models largely captured the biases introduced by individual players.

Download Full-text

People Analytics of Semantic Web Human Resource Résumés for Sustainable Talent Acquisition

Sustainability ◽

10.3390/su11133520 ◽

2019 ◽

Vol 11 (13) ◽

pp. 3520 ◽

Cited By ~ 3

Author(s):

Sabina-Cristiana Necula ◽

Cătălin Strîmbei

Keyword(s):

Semantic Web ◽

Human Resource ◽

Data Science ◽

Support Vector ◽

K Nearest Neighbors ◽

Semantic Web Technologies ◽

Tree Algorithms ◽

Using Data ◽

Talent Acquisition

The purpose of this study was to define a data science architecture for talent acquisition. The approach was to propose analytics that derive data. The originality of this paper consists in proposing an architecture to work within the process of obtaining semantically enriched data by using data science and Semantic Web technologies. We applied the proposed architecture and developed a case study-based prototype that uses analytics techniques for résumé data integrated with Linked Data technologies. We conducted a case study to identify skills by applying classification via regression, k-nearest neighbors (k-NN), random forest, naïve Bayes, support vector machine, and decision tree algorithms to résumé data that we previously described with terms from publicly available ontologies. We labeled data from résumés using terms from existing human resource ontologies. The main contribution is the extraction of skills from résumés and the mining of data that was previously described with the Semantic Web.

Download Full-text

Flood Stage Forecasting at the Gurye-Gyo Station in Sumjin River Using LSTM-Based Deep Learning Models

Korean Society of Hazard Mitigation ◽

10.9798/kosham.2021.21.3.193 ◽

2021 ◽

Vol 21 (3) ◽

pp. 193-201

Author(s):

Jaewon Jung ◽

Hyelim Mo ◽

Junhyeong Lee ◽

Younghoon Yoo ◽

Hung Soo Kim

Keyword(s):

Deep Learning ◽

Short Term Memory ◽

Warning System ◽

Flood Forecasting ◽

Flood Damage ◽

Support Vector ◽

Forecasting Models ◽

Physically Based ◽

Forecasting And Warning ◽

Storm Rainfall

Instances of flood damage caused by extreme storm rainfall due to climate change and variability have been showing an increasing trend. Particularly, a flood forecasting and warning system has been recognized as an important nonstructural measure for flood damage reduction, including loss of life. Flood forecasting and warning have been performed by the forecasts of flood discharge and flood stage using the physically based rainfall-runoff models. However, recently, studies involving the application of a machine learning-based flood forecasting models, which addresses the limitations of extant physically based flood stage forecasting models, have been performed. We may require various case studies to determine more accurate methods. Therefore, this study performed the real-time forecasting of the river water level or stage at the Gurye station of the Sumjin river with lead times of 1, 3, and 6 h by applying a long short-term memory (LSTM)-based deep learning model. In addition, the applicability of the LSTM model was evaluated by comparing the results with those from widely used models based on support vector machine and multilayer perceptron. Consequently, we noted that the LSTM model exhibited a relatively better forecasting performance. Therefore, the applicability of the LSTM model should be extensively studied for flood forecasting applications.

Download Full-text

An improved short term load forecasting with ranker based feature selection technique

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-191568 ◽

2020 ◽

Vol 39 (5) ◽

pp. 6783-6800

Author(s):

Siva Sankari Subbiah ◽

Jayakumar Chinnappan

Keyword(s):

Feature Selection ◽

Short Term Memory ◽

Load Forecasting ◽

Support Vector ◽

Short Term ◽

Feature Selection Technique ◽

Forecasting Models ◽

Short Term Load Forecasting ◽

Utility Companies ◽

Electricity Load

The load forecasting is the significant task carried out by the electricity providing utility companies for estimating the future electricity load. The proper planning, scheduling, functioning, and maintenance of the power system rely on the accurate forecasting of the electricity load. In this paper, the clustering-based filter feature selection is proposed for assisting the forecasting models in improving the short term load forecasting performance. The Recurrent Neural Network based Long Short Term Memory (LSTM) is developed for forecasting the short term load and compared against Multilayer Perceptron (MLP), Radial Basis Function (RBF), Support Vector Regression (SVR) and Random Forest (RF). The performance of the forecasting model is improved by reducing the curse of dimensionality using filter feature selection such as Fast Correlation Based Filter (FCBF), Mutual Information (MI), and RReliefF. The clustering is utilized to group the similar load patterns and eliminate the outliers. The feature selection identifies the relevant features related to the load by taking samples from each cluster. To show the generality, the proposed model is experimented by using two different datasets from European countries. The result shows that the forecasting models with selected features produce better performance especially the LSTM with RReliefF outperformed other models.

Download Full-text

Using Machine Learning Algorithms on Prediction of Stock Price

Journal of Modeling and Optimization ◽

10.32732/jmo.2020.12.2.84 ◽

2020 ◽

Vol 12 (2) ◽

pp. 84-99

Author(s):

Li-Pang Chen

Keyword(s):

Machine Learning ◽

Stock Price ◽

Short Term Memory ◽

Machine Learning Algorithms ◽

Machine Learning Techniques ◽

Support Vector ◽

Short Term ◽

Learning Techniques ◽

Historical Database ◽

Long Short Term Memory

In this paper, we investigate analysis and prediction of the time-dependent data. We focus our attention on four different stocks are selected from Yahoo Finance historical database. To build up models and predict the future stock price, we consider three different machine learning techniques including Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN) and Support Vector Regression (SVR). By treating close price, open price, daily low, daily high, adjusted close price, and volume of trades as predictors in machine learning methods, it can be shown that the prediction accuracy is improved.

Download Full-text