This report details a time series forecasting analysis of air quality data, focusing on predicting relative humidity, temperature, and absolute humidity. The report utilizes a dataset of 9358 samples from chemical sensors in an Italian city, employing linear regression and the Weka time series model. It describes the dataset's attributes, including concentrations of various pollutants and meteorological data. The methodology involves converting the data into the arff format and applying linear regression to forecast temperature. The report discusses the time series model within Weka, including configuration options and evaluation metrics. The results indicate a high R^2 value for the temperature model, with errors increasing over time. The conclusion highlights the advantages of time series forecasting over linear regression for air quality prediction, particularly the accuracy achieved with the date frame of data in successive hours. References to relevant research are also included.