Permanent link to this recordhttp://hdl.handle.net/10754/561675
MetadataShow full item record
AbstractDecision trees appeared in 50-60s of the last century in theoretical computer science [14, 64, 80] and applications [24, 37]. Similar objects are also considered by natural and social sciences, for example, taxonomy keys  or questionnaires . Decision trees naturally represent identification and testing algorithms that specify the next test to perform based on the results of the previous tests. A number of particular formulations were generalized by Garey  as identification problem that is a problem of distinguishing objects described by a common set of attributes. More general formulation is provided by decision table framework [34, 65] where objects can have incomplete set of attributes and non-unique class labels. In that case, acquiring class label is enough to solve the problem: identifying a particular object is not required. In this context, decision trees found many applications in test theory [39, 45, 46, 81], fault diagnosis [14, 60, 72], rough set theory [61, 62], discrete optimization, non-procedural programming languages , analysis of algorithm complexity , computer vision , computational geometry . © Springer-Verlag Berlin Heidelberg 2011.
CitationChikalov, I. (2011). Introduction. Average Time Complexity of Decision Trees, 1–14. doi:10.1007/978-3-642-22661-8_1