Similarity Tests (Compression Tests)

The similarity test is composed of 8 exercises generated with binary strings of different length and different levels of edit errors (insertion, deletion or change). The strings were generated and checked by dynamic programming to ensure that they did not have a better correction path. The purpose of this test was to measure the ability of compression by trivial pattern matching.


Return to the Computational Measurement of Intelligence Page


© 1999-2000 José Hernández Orallo.