Context selection for linguistic data fusion

Reports generated by soldiers are common in time-critical military environments. Data fusion systems that attempt to process those reports must maintain the context for each set of observations to avoid inaccurate state estimates. This paper analyzes the selection and assignment of topical context under a Bayesian methodology. We present several techniques to decrease the hypothesis space and heuristics that apply specifically to military reporting environments. Using a data set consisting of semi-structured reports, we show that this approach allows accurate assignment of topical context even when the context is only implied rather than given explicitly.