2008 | OriginalPaper | Buchkapitel
Querying and Reasoning for Spatiotemporal Data Mining
verfasst von : G. Manco, M. Baglioni, F. Giannotti, B. Kuijpers, A. Raffaetà, C. Renso
Erschienen in: Mobility, Data Mining and Privacy
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In the previous chapters, we studied movement data from several perspectives: the application opportunities, the type of analytical questions, the modeling requirements, and the challenges for mining. Moreover, the complexity of the overall analysis process was pointed out several times. The analytical questions posed by the end user need to be translated into several tasks such as choose analysis methods, prepare the data for application of these methods, apply the methods to the data, and interpret and evaluate the results obtained. To clarify these issues, let us consider an example involving the following analytical questions:
Describe the collective movement behavior of the population (or a given subset) of entiti es during the whole time period (or a given interval)
Find the entity subsets and time periods with the collective movement behavior corresponding to a given pattern
Compare the collective movement behaviors of the entities on given time intervals
It is evident that there is a huge distance between these analytical questions and the complex computations needed to answer them. In fact, answering the above questions requires combining several forms of knowledge and the cooperation among solvers of different nature: we need spatiotemporal reasoning supporting deductive inferences along with inductive mechanisms, in conjunction with statistical methods.