Wind Data Analysis and Data Imputation Using Classical and Machine Learning Techniques


  • Claudia Sánchez
  • Dr. Mario Graff


This tutorial presents how to use matplotlib, a python library, for exploring, visualizing, and understanding a wind data base. Based on the daily registers of wind speed and direction, we are going to analyze the wind behavior by months or hours. Besides, because some of the sensors could fail, for the imputation of missing data we are going to use classical interpolation techniques and machine learning tools.

Claudia Sánchez is finalizing her Ph.D. in Data Sciences from INFOTEC, México. She has a master's degree in Computation and Industrial Mathematics from CIMAT, México. Currently, she is a full-time researcher-professor at Universidad Panamericana campus Aguascalientes. She has taught Evolutionary Computation in both INFOTEC and Universidad Panamericana. Her main research interests include Machine Learning, Data Analysis, Evolutionary Computation, and Computer Vision.
Mario Graff is a Research Fellow of CONACYT commission to INFOTEC. He is part of the National System of Researches. He researches in the fields of Machine Learning, Evolutionary Computation, and Natural Language Processing (NLP). The research deals with the application of Evolutionary Computation, particularly Genetic Programming (GP), to supervised learning problems. In this research avenue, special attention has been put on the use of GP to solve text classification and sentiment analysis problems posed as a supervised learning task. In NLP, the interest is on developing multilingual text representations, also focused on those representations that facilitate transfer knowledge between languages. He obtained his Ph.D. from the School of Computer Science and Electronic Engineering at the University of Essex, U.K.