Extracción de conocimiento mediante ventanas de tiempo en variables atmosféricas

Industrialization and the rapid growth of urban areas are alarmingly increasing the presence of air pollutants. These pollutants affect the quality of life of people and present an opportunity for study is created to determine the atmospheric behavior and relationship between meteorological variables present in the environment. Prior to this, rolling windows of time were applied to remove anomalous data. Next, variables were identified, and the data was segmented through the X-means algorithm. Also, two clusters that represent the relationships between pairs of variables and the temporality of the time windows. As a result, an inverse correlation of-0.78 was found between the ozone and dew point variables within the hours of the working day.