Francisco Martínez-Álvarez

Neural networks to predict earthquakes in Chile

A new earthquake prediction system is presented in this work. This method, based on the application of artificial neural networks, has been used to predict earthquakes in Chile, one of the countries with larger seismic activity. The input values are related to the b-value, the Baths law, and the Omori-Utsus law, parameters that are strongly correlated with seismicity, as shown in solid previous works. Two kind of prediction are provided in this study: The probability that an earthquake of magnitude larger than a threshold value happens, and the probability that an earthquake of a limited magnitude interval might occur, both during the next five days in the areas analyzed. For the four Chiles seismic regions examined, with epicenters placed on meshes with dimensions varying from 0.5^ox0.5^o to 1^ox1^o, a prototype of neuronal network is presented. The prototypes predict an earthquake every time the probability of an earthquake of magnitude larger than a threshold is sufficiently high. The threshold values have been adjusted with the aim of obtaining as few false positives as possible. The accuracy of the method has been assessed in retrospective experiments by means of statistical tests and compared with well-known machine learning classifiers. The high success rate achieved supports the suitability of applying soft computing in this field and poses new challenges to be addressed.

Pattern recognition to forecast seismic time series

Earthquakes arrive without previous warning and can destroy a whole city in a few seconds, causing numerous deaths and economical losses. Nowadays, a great effort is being made to develop techniques that forecast these unpredictable natural disasters in order to take precautionary measures. In this paper, clustering techniques are used to obtain patterns which model the behavior of seismic temporal data and can help to predict medium-large earthquakes. First, earthquakes are classified into different groups and the optimal number of groups, a priori unknown, is determined. Then, patterns are discovered when medium-large earthquakes happen. Results from the Spanish seismic temporal data provided by the Spanish Geographical Institute and non-parametric statistical tests are presented and discussed, showing a remarkable performance and the significance of the obtained results.

Mining quantitative association rules based on evolutionary computation and its application to atmospheric pollution

This research presents the mining of quantitative association rules based on evolutionary computation techniques. First, a real-coded genetic algorithm that extends the well-known binary-coded CHC algorithm has been projected to determine the intervals that define the rules without needing to discretize the attributes. The proposed algorithm is evaluated in synthetic datasets under different levels of noise in order to test its performance and the reported results are then compared to that of a multi-objective differential evolution algorithm, recently published. Furthermore, rules from real-world time series such as temperature, humidity, wind speed and direction of the wind, ozone, nitrogen monoxide and sulfur dioxide have been discovered with the objective of finding all existing relations between atmospheric pollution and climatological conditions.

Determining the best set of seismicity indicators to predict earthquakes. Two case studies: Chile and the Iberian Peninsula

This work explores the use of different seismicity indicators as inputs for artificial neural networks. The combination of multiple indicators that have already been successfully used in different seismic zones by the application of feature selection techniques is proposed. These techniques evaluate every input and propose the best combination of them in terms of information gain. Once these sets have been obtained, artificial neural networks are applied to four Chilean zones (the most seismic country in the world) and to two zones of the Iberian Peninsula (a moderate seismicity area). To make the comparison to other models possible, the prediction problem has been turned into one of classification, thus allowing the application of other machine learning classifiers. Comparisons with original sets of inputs and different classifiers are reported to support the degree of success achieved. Statistical tests have also been applied to confirm that the results are significantly different than those of other classifiers. The main novelty of this work stems from the use of feature selection techniques for improving earthquake prediction methods. So, the information gain of different seismic indicators has been determined. Low ranked or null contribution seismic indicators have been removed, optimizing the method. The optimized prediction method proposed has a high performance. Finally, four Chilean zones and two zones of the Iberian Peninsula have been characterized by means of an information gain analysis obtained from different seismic indicators. The results confirm the methodology proposed as the best features in terms of information gain are the same for both regions.

An evolutionary algorithm to discover quantitative association rules in multidimensional time series

An evolutionary approach for finding existing relationships among several variables of a multidimensional time series is presented in this work. The proposed model to discover these relationships is based on quantitative association rules. This algorithm, called QARGA (Quantitative Association Rules by Genetic Algorithm), uses a particular codification of the individuals that allows solving two basic problems. First, it does not perform a previous attribute discretization and, second, it is not necessary to set which variables belong to the antecedent or consequent. Therefore, it may discover all underlying dependencies among different variables. To evaluate the proposed algorithm three experiments have been carried out. As initial step, several public datasets have been analyzed with the purpose of comparing with other existing evolutionary approaches. Also, the algorithm has been applied to synthetic time series (where the relationships are known) to analyze its potential for discovering rules in time series. Finally, a real-world multidimensional time series composed by several climatological variables has been considered. All the results show a remarkable performance of QARGA.

Combining pattern sequence similarity with neural networks for forecasting electricity demand time series

We present PSF-NN, a new approach for time series forecasting. It combines prediction based on sequence similarity with neural networks. PSF-NN first generates predictions using the PSF algorithm that are then refined by the neural network component, which also utilizes additional features. We evaluate the performance of PSF-NN using a time series of hourly electricity demands for the state of New South Wales in Australia for three years. The task is to predict an interval of future values simultaneously, i.e. the 24 demands for the next day, instead of predicting just a single future demand. The results showed that the combined PSF-NN approach provides accurate predictions, outperforming the original PSF algorithm and a number of baselines.

Partitioning-clustering techniques applied to the electricity price time series

Clustering is used to generate groupings of data from a large dataset, with the intention of representing the behavior of a system as accurately as possible. In this sense, clustering is applied in this work to extract useful information from the electricity price time series. To be precise, two clustering techniques, K-means and Expectation Maximization, have been utilized for the analysis of the prices curve, demonstrating that the application of these techniques is effective so to split the whole year into different groups of days, according to their prices conduct. Later, this information will be used to predict the price in the short time period. The prices exhibited a remarkable resemblance among days embedded in a same season and can be split into two major kind of clusters: working days and festivities.

A comparison of machine learning regression techniques for LiDAR-derived estimation of forest variables

Light Detection and Ranging (LiDAR) is a remote sensor able to extract three-dimensional information. Environmental models in forest areas have been benefited by the use of LiDAR-derived information in the last years. A multiple linear regression (MLR) with previous stepwise feature selection is the most common method in the literature to develop those models. MLR defines the relation between the set of field measurements and the statistics extracted from a LiDAR flight. Machine learning has emerged as a suitable tool to improve classic stepwise MLR results on LiDAR. Unfortunately, few studies have been proposed to compare the quality of the multiple machine learning approaches. This paper presents a comparison between the classic MLR-based methodology and regression techniques in machine learning (neural networks, support vector machines, nearest neighbour, ensembles such as random forests) with special emphasis on regression trees. The selected techniques are applied to real LiDAR data from two areas in the province of Lugo (Galizia, Spain). The results confirm that classic MLR is outperformed by machine learning techniques and concretely, our experiments suggest that Support Vector Regression with Gaussian kernels statistically outperforms the rest of the techniques.

LBF: A Labeled-Based Forecasting Algorithm and Its Application to Electricity Price Time Series

A new approach is presented in this work with the aim of predicting time series behaviors. A previous labeling of the samples is obtained utilizing clustering techniques and the forecasting is applied using the information provided by the clustering. Thus, the whole data set is discretized with the labels assigned to each data point and the main novelty is that only these labels are used to predict the future behavior of the time series, avoiding using the real values of the time series until the process ends. The results returned by the algorithm, however, are not labels but the nominal value of the point that is required to be predicted. The algorithm based on labeled (LBF) has been tested in several energy-related time series and a notable improvement in the prediction has been achieved.

Computational intelligence techniques for predicting earthquakes

Nowadays, much effort is being devoted to develop techniques that forecast natural disasters in order to take precautionary measures. In this paper, the extraction of quantitative association rules and regression techniques are used to discover patterns which model the behavior of seismic temporal data to help in earthquakes prediction. Thus, a simple method based on the k-smallest and k-greatest values is introduced for mining rules that attempt at explaining the conditions under which an earthquake may happen. On the other hand patterns are discovered by using a tree-based piecewise linear model. Results from seismic temporal data provided by the Spanishs Geographical Institute are presented and discussed, showing a remarkable performance and the significance of the obtained results.