José Fernando de Toledo, Patrícia Teixeira Leite Assano, H. Siqueira, R. Sacchi
{"title":"Performance comparison of machine learning models for streamflow forecasting","authors":"José Fernando de Toledo, Patrícia Teixeira Leite Assano, H. Siqueira, R. Sacchi","doi":"10.1109/LA-CCI48322.2021.9769829","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769829","url":null,"abstract":"In this work, the performance of three models for streamflow forecasting is compared, based on six scenarios that considered the phases of El Niño South Oscillation (ENSO) and Climate Indicators, for eight hydroelectric plants located in four regions of the Brazilian territory. The models addressed are Support Vector Regression, Extreme Learning Machine and Kernel Ridge Regression. The climatic variables used are the rainfall, the location of the Intertropical Convergence Zone (ITCZ) and occurrence data from the South Atlantic Convergence Zone (SACZ). The criterion for comparing the models considered the means and variances of the series of forecast errors for each Plant. The computational results indicated that the Kernel Ridge Regression model obtained the best results in most of the tested scenarios, including those that considered the use of Climate Indicators.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122332362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Probabilistic Multistep Time Series Forecasting Using Conditional Generative Adversarial Networks","authors":"Gerardo Zúñiga, G. Acuña","doi":"10.1109/LA-CCI48322.2021.9769836","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769836","url":null,"abstract":"Time series forecasting is a problem that has been studied for many years due to the impact it can have on the world economy and well-being. Predicting multiple future values is an especially complex problem due to the increasing error. This is why there is a need to design and evaluate more and better methods for this forecasting problem. The adversarial generative networks seem to have an excellent performance generating time series indistinguishable from real series. It has been shown that a probabilistic prediction of time series called ForGAN adversary generative network has been successfully used for one-step-ahead predictions. In this work, a modified architecture of ForGAN with multiple outputs is proposed in order to perform multiple-step-ahead predictions. We show by means of experiments using a real dataset that statistically significant improvement of multiple-step-ahead predictions with the proposed modified architecture of ForGAN compared with the use of the original ForGAN network is achieved, decreasing RMSE by 17.6% and CRPS by 17.3% when predicting 5 steps ahead.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132053107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Systematic Mapping of Smart Farming and Image Recognition in Agriculture","authors":"Bruno Gutierrez Ríos, G. C. Saldías","doi":"10.1109/LA-CCI48322.2021.9769828","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769828","url":null,"abstract":"Automation, Internet 4.0, the use of production models and Computational Intelligence techniques have been strongly related with agriculture in recent years. Considering agriculture as a fundamental human activity and also a key industry that is being strongly affected by the climate crisis, knowing how to propose solutions to its various problems through information technologies is of utmost importance. Consequently, this article presents a systematic mapping in order to identify the different applications of machine learning in agriculture, paying special attention to the tools used to acquire relevant information, such as drones and sensors, and to the models used to create solutions, such as Support Vector Machines and Artificial Neural Networks. The paper focuses on the use of image classifying models, evaluating the applications of artificial vision in agriculture, especially in detecting diseases. A comparative study carried out with different deep learning tools in order to identify plant diseases will be presented. The study shows the power of deep learning tools using transfer learning, evidencing that, in networks, these tools learn within few iterations, maintaining excellent levels of generalization, as shown by the validation results.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130970752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A protocol for Brain-Computer Interfaces based on Musical Notes Imagery","authors":"Anna Montevilla, Guillermo Sahonero-Alvarez","doi":"10.1109/LA-CCI48322.2021.9769845","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769845","url":null,"abstract":"The application of Brain-Computer Interfaces is expected to become a matter of daily life. For this purpose, several efforts are being developed to ensure that users can employ this technology without difficulties. A large amount of studies consider motor imagery, which implies the usage of sensorimotor rhythms produced when imaging motor actions. However, previous works have shown that from a sample of population, a portion of users (15~30%) is unable to efficiently control a BCI based on such paradigm. The roots of this issue have been partially located to different factors related to the training protocol that users follow to learn how to use the system. Thus, in order to extend the applicability of BCIs, training procedures must consider different approaches. Musical imagery is another mental task that may be used to control BCIs and requires users to have music related thoughts or imagine specific notes and even songs. In this work, we propose a protocol to explore the properties of Musical Imagery based training procedures. For this, we developed both offline and online experiments, where the last one consisted of 4 sessions. The data-processing steps include filtering the data using a FIR filter to later extract features using PCA, and classify such features with a multi-class SVM. Our results show that the offline classification is comparable to motor imagery based BCIs as the accuracy is between 80% to 95%. Moreover, we found that the online setup results point to up to 64% of accuracy for the third session with feedback.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121316986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Reducing the effort of Evolutionary Calibrator Using Opposite Information","authors":"Nicolás Rojas-Morales, M. Riff","doi":"10.1109/LA-CCI48322.2021.9769793","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769793","url":null,"abstract":"Metaheuristics have been successfully applied to solve complex real-world problems in many application domains. Their performance strongly depends on the values of their parameters. Many tuning algorithms have already been proposed to find a set of suitable values. However, the amount of computational time required to obtain these values is usually high. Our goal is to propose a collaborative strategy to help to reduce the configuration effort during the tuning process. Here, we introduce a novel initialization strategy that learns from poor quality configurations in a pre-processing phase. We evaluate our collaboration using the well-known Evolutionary Calibrator (Evoca). Moreover, we tune two different algorithms: the Ant Knapsack algorithm, using hard instances of the Multidimensional Knapsack Problem, and a Genetic Algorithm for solving landscapes that follow the NK model (N components and degree K). Evoca obtains promising results using our novel strategy, consuming less computational resources.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127294283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bryan Sagredo, Sonia Espanol-Jim'enez, Felipe A. Tobar
{"title":"Detection of blue whale vocalisations using a temporal-domain convolutional neural network","authors":"Bryan Sagredo, Sonia Espanol-Jim'enez, Felipe A. Tobar","doi":"10.1109/LA-CCI48322.2021.9769846","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769846","url":null,"abstract":"We present a framework for detecting blue whale vocalisations from acoustic submarine recordings. The proposed methodology comprises three stages: i) a preprocessing step where the audio recordings are conditioned through normalisation, filtering, and denoising; ii) a label-propagation mechanism to ensure the consistency of the annotations of the whale vocalisations, and iii) a convolutional neural network that receives audio samples. Based on 34 real-world submarine recordings (28 for training and 6 for testing) we obtained promising performance indicators including an Accuracy of 85.4% and a Recall of 93.5%. Furthermore, even for the cases where our detector did not match the ground-truth labels, a visual inspection validates the ability of our approach to detect possible parts of whale calls unlabelled as such due to not being complete calls.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126581458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alejandro Cuevas, Sebasti'an D. L'opez, D. Mandic, Felipe A. Tobar
{"title":"Bayesian autoregressive spectral estimation","authors":"Alejandro Cuevas, Sebasti'an D. L'opez, D. Mandic, Felipe A. Tobar","doi":"10.1109/LA-CCI48322.2021.9769834","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769834","url":null,"abstract":"Autoregressive (AR) time series models are widely used in parametric spectral estimation (SE), where the power spectral density (PSD) of the time series is approximated by that of the best-fit AR model, which is available in closed form. Since AR parameters are usually found via maximum-likelihood, least squares or the method of moments, AR-based SE fails to account for the uncertainty of the approximate PSD, and thus only yields point estimates. We propose to handle the uncertainty related to the AR approximation by finding the full posterior distribution of the AR parameters to then propagate this uncertainty to the PSD approximation by integrating out the AR parameters; we implement this concept by assuming two different priors over the model noise. Through practical experiments, we show that the proposed Bayesian autoregressive spectral estimation (BASE) provides point estimates that follow closely those of standard autoregressive spectral estimation (ASE), while also providing error bars. BASE is validated against ASE and the Periodogram on both synthetic and real-world signals.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"123 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131779330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"High-dimensional Multivariate Time Series Forecasting in IoT Applications using Embedding Non-stationary Fuzzy Time Series","authors":"H. V. Bitencourt, F. Guimarães","doi":"10.1109/LA-CCI48322.2021.9769792","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769792","url":null,"abstract":"In Internet of things (IoT), data is continuously recorded from different data sources and devices can suffer faults in their embedded electronics, thus leading to a high-dimensional data sets and concept drift events. Therefore, methods that are capable of high-dimensional non-stationary time series are of great value in IoT applications. Fuzzy Time Series (FTS) models stand out as data-driven non-parametric models of easy implementation and high accuracy. Unfortunately, FTS encounters difficulties when dealing with data sets of many variables and scenarios with concept drift. We present a new approach to handle high-dimensional non-stationary time series, by projecting the original high-dimensional data into a low dimensional embedding space and using FTS approach. Combining these techniques enables a better representation of the complex content of non-stationary multivariate time series and accurate forecasts. Our model is able to explain 98% of the variance and reach 11.52 of RMSE, 2.68 of MAE and 2.91% of MAPE.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125885837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Adaptive Gaussian Fuzzy Classifier for Real-Time Emotion Recognition in Computer Games","authors":"D. Leite, Volnei Frigeri, Rodrigo Medeiros","doi":"10.1109/LA-CCI48322.2021.9769842","DOIUrl":"https://doi.org/10.1109/LA-CCI48322.2021.9769842","url":null,"abstract":"Emotion recognition has become a need for more realistic and interactive machines and computer systems. The greatest challenge is the availability of high-performance algorithms to effectively manage individual differences and nonstationarities in physiological data, i.e., algorithms that customize models to users with no subject-specific calibration data. We describe an evolving Gaussian Fuzzy Classifier (eGFC), which is supported by a semi-supervised learning algorithm to recognize emotion patterns from electroencephalogram (EEG) data streams. We extract features from the Fourier spectrum of EEG data. The data are provided by 28 individuals playing the games ‘Train Sim World’, ‘Unravel’, ‘Slender The Arrival’, and ‘Goat Simulator’ – a public dataset. Different emotions prevail, namely, boredom, calmness, horror and joy. We analyze individual electrodes, time window lengths, and frequency bands on the accuracy of user-independent eGFCs. We conclude that both brain hemispheres may assist classification, especially electrodes on the frontal (Af3-Af4), occipital (O1-O2), and temporal (T7-T8) areas. We observe that patterns may be eventually found in any frequency band; however, the Alpha (8-13Hz), Delta (1-4Hz), and Theta (4-8Hz) bands, in this order, are more correlated with the emotion classes. eGFC has shown to be effective for real-time learning of EEG data. It reaches a 72.2% accuracy using a variable rule base, 10-second windows, and 1.8ms/sample processing time in a highly-stochastic time-varying 4-class classification problem.","PeriodicalId":431041,"journal":{"name":"2021 IEEE Latin American Conference on Computational Intelligence (LA-CCI)","volume":"843 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116423598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}