Cluster analysis for functional data: Kinematic analysis of one leg hop of anterior cruciate ligament patients and uninjured controls
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
In this thesis we present an overview of functional data analysis. More specifically, we consider methods of aligning and clustering functions. We compare different clustering methods based on a kinematic data collected from a one leg hop performed by individuals with an anterior cruciate ligament (ACL) injury and uninjured controls. The clustering methods are also used to identify whether there are clusters of individuals with similar movement patterns. We investigate whether the ACL injured and uninjured controls and other variables recorded for each individual lead in significantly different classification of the clusters. The results show that different clustering methods provide various compositions of clusters. Furthermore, we found out that the significant clusters mostly depend on the length of the jump but also indications that some clusters differ between individuals with ACL injury and uninjured controls.
Place, publisher, year, edition, pages
2014. , 55 p.
IdentifiersURN: urn:nbn:se:umu:diva-85063OAI: oai:DiVA.org:umu-85063DiVA: diva2:691473