The KIAM System in the C@merata Task at MediaEval 2016

The KIAM system is an attempt to solve the C@merata task at MediaEval 2016 [4] using regular expressions. The aim of the KIAM system is to try the suitability of regular expressions for text recognition in a limited dictionary. The system answers natural language queries over musical scores. The C@merata project has existed since 2014 and there are 200 questions each year which a participant's system has to answer automatically. A question consists of a short noun phrase in English referring to a musical feature in the music score like "three consecutive thirds in the left hand in measures 22-30" together with a classical music score in MusicXML [2]. The required answer to this question is a set of one or more passages specifying exact places in the score relevant with input string.