Abstract:
An approach is provided for state classification for a travel segment with multi-modal speed profiles. A traffic processing platform processes and/or facilitates a processing of probe data associated with at least one travel segment to determine that probe data indicates a plurality of speed profiles. The plurality of speed profiles represent one or more observed clusters of speed states. The traffic processing platform also determine that the at least one travel segment exhibits a multi-modality with respect to travel speed based, at least in part, on the plurality of speed profiles. The traffic processing platform then determines at least one likely sequence of speed states for traversing the at least one travel segment based, at least in part, on the one or more observed clusters of speed states and state transition probability information, wherein the state transition probability information represents one or more probabilities for transitioning among the plurality of speed states and causes, at least in part, a classification of at least one hidden state of the at least one travel segment based, at least in part, on the at least one likely sequence of speed states.