Bibliography

(eng): The paper discuss the problem of wide context (temporal, spatial, local, objective, attribute oriented, relation oriented) as a tool to compensate and to decrease the uncertainty of data, classification and analytical process at all process to increase the information value of decision support. The contribution deals with a problem of creating the composed classifier with boosting architecture, whose components are composed of classifiers working with k - NN algorithm (k - th nearest neighbour).

(cze): Příspěvek je věnován problematice širokého kontextu z hlediska jeho možností kompenzovat neurčitost dat.