Creating predictive models to process large amounts of information is a major challenge for which the latest big data techniques provide solutions. In this paper, we present a methodology based on time series algorithms and multiplex networks focused on Big Data solutions. This methodology allows to process a large amount of information and obtain a way to group the most effective and useful information, allowing to solve problems with a large number of time-varying variables in an efficient way. This approach offers the possibility of predicting the evolution of housing prices using cab rides between different areas of the city. The methodology presented to combine all this information is based both on the original use of some unsupervised machine learning techniques and on the use of certain attributes of the time series and their representation as a complex multiplex network, achieving a very significant reduction in the dimensionality of the resulting data representation. The result is a forecast that reduces the representation of cab rides to a small dataset able to forecast real estate prices.
Read full abstract