Browsing by Keyword "Feature selection"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item A Coral Reefs Optimization algorithm with Harmony Search operators for accurate wind speed prediction(2015-03-01) Salcedo-Sanz, Sancho; Pastor-Sanchez, Alvaro; Del Ser, Javier; Prieto, Luis; Geem, Zong-Woo; IAThis paper introduces a new hybrid bio-inspired solver which combines elements from the recently proposed Coral Reefs Optimization (CRO) algorithm with operators from the Harmony Search (HS) approach, which gives rise to the coined CRO-HS optimization technique. Specifically, this novel bio-inspired optimizer is utilized in the context of short-term wind speed prediction as a means to obtain the best set of meteorological variables to be input to a neural Extreme Learning Machine (ELM) network. The paper elaborates on the main characteristics of the proposed scheme and discusses its performance when predicting the wind speed based on the measures of two meteorological towers located in USA and Spain. The good results obtained in these experiments when compared to naïve versions of the CRO and HS algorithms are promising and pave the way towards the utilization of the derived hybrid solver in other optimization problems arising from diverse disciplines.Item Data Driven Performance Prediction in Steel Making(2022-01-18) Boto, Fernando; Murua, Maialen; Gutierrez, Teresa; Casado, Sara; Carrillo, Ana; Arteaga, Asier; Tecnalia Research & Innovation; FACTORY; CIRMETAL; PROMETALThis work presents three data-driven models based on process data, to estimate different indicators related to process performance in a steel production process. The generated models allow the optimization of the process parameters to achieve optimal performance and quality levels. A new approach based on ensembles has been developed with feature selection methods and four state-of-the-art regression approximations (random forest, gradient boosting, xgboost and neural networks). The results show that the proposed approach makes the prediction more stable reducing the variance for all cases, even in one case, slightly reducing the bias. Furthermore, from the four machine learning paradigms presented, random forest is the one with the best results in a quantitative way, obtaining a coefficient of determination of 0.98 as a maximum, depending on the target sub-process.Item Feature selection for hand pose recognition in human-robot object exchange scenario(Institute of Electrical and Electronics Engineers Inc., 2014-01-08) Rasines, Irati; Remazeilles, Anthony; Bengoa, Pedro M.Iriondo; Martinez Garcia, Herminio; Grau, Antoni; Robótica MédicaVision-based hand gesture recognition relies on the extraction of features describing the hand, and the appropriate set of features is usually selected in an empirical manner. We propose in this article a systematic selection of the best features to be considered. An iterative sequential forward feature selection (SFS) approach is proposed to combine the features with the highest recognition rate considering the Gaussian Mixture Modelling within the Expectation Maximization algorithm as classification technique. This approach has been tested with two different illustrative databases. The first one is related to human robot physical interaction and the hand postures considered correspond to key postures the human partner performs just before acquiring an object from the robot. The second database corresponds to the representation of the 10 first numbers of the American Sign Language. In both cases, the recognition rate obtained, measured through the F1 score metrics, is satisfactory (over 0,97), and demonstrates that the proposed technique could be applied to a very large field of applications.Item A feature selection method for author identification in interactive communications based on supervised learning and language typicality(2016-11-01) Villar-Rodriguez, Esther; Del Ser, Javier; Bilbao, Miren Nekane; Salcedo-Sanz, Sancho; Tecnalia Research & Innovation; Quantum; IAAuthorship attribution, conceived as the identification of the origin of a text between different authors, has been a very active area of research in the scientific community mainly supported by advances in Natural Language Processing (NLP), machine learning and Computational Intelligence. This paradigm has been mostly addressed from a literary perspective, aiming at identifying the stylometric features and writeprints which unequivocally typify the writer patterns and allow their unique identification. On the other hand, the upsurge of social networking platforms and interactive messaging have undoubtedly made the anonymous expression of feelings, the sharing of experiences and social relationships much easier than in other traditional communication media. Unfortunately, the popularity of such communities and the virtual identification of their users deploy a rich substrate for cybercrimes against unsuspecting victims and other forms of illegal uses of social networks that call for the activity tracing of accounts. In the context of one-to-one communications this manuscript postulates the identification of the sender of a message as a useful approach to detect impersonation attacks in interactive communication scenarios. In particular this work proposes to select linguistic features extracted from messages via NLP techniques by means of a novel feature selection algorithm based on the dissociation between essential traits of the sender and receiver influences. The performance and computational efficiency of different supervised learning models when incorporating the proposed feature selection method is shown to be promising with real SMS data in terms of identification accuracy, and paves the way towards future research lines focused on applying the concept of language typicality in the discourse analysis field.Item Wind power production forecasting using ant colony optimization and extreme learning machines(2017) Carrillo, Maria; Del Ser, Javier; Nekane Bilbao, Miren; Perfecto, Cristina; Camacho, David; IANowadays the energy generation strategy of almost every nation around the world relies on a strong contribution from renewable energy sources. In certain countries the relevance taken by wind energy is particularly high within its national production share, mainly due to its large-scale wind flow patterns. This noted potentiality of wind energy has so far attracted public and private funds to support the development of advanced wind energy technologies. However, the proliferation of wind farms makes it challenging to achieve a proper electricity balance of the grid, a problem that becomes further involved due to the fluctuations of wind generation that occur at different time scales. Therefore, acquiring a predictive insight on the variability of this renewable energy source becomes essential in order to optimally inject the produced wind energy into the electricity grid. To this end the present work elaborates on a hybrid predictive model for wind power production forecasting based on meteorological data collected at different locations over the area where a wind farm is located. The proposed method hybridizes Extreme Learning Machines with a feature selection wrapper that models the discovery of the optimum subset of predictors as a metric-based search for the optimum path through a solution graph efficiently tackled via Ant Colony Optimization. Results obtained by our approach for two real wind farms in Zamora and Galicia (Spain) are presented and discussed, from which we conclude that the proposed hybrid model is able to efficiently reduce the number of input features and enhance the overall model performance.