Finding Repeated Flexible Relational Words in Sequences

Finding regularities in sequences is an important problem in various areas. Regularities are often words (in a strict or somewhat flexible meaning) "repeated" in the sequence, i.e. satisfying some constraints about their occurrence. In this paper we deal with relational values that express what relates two positions in a word or a sequence. Then a strict relational word is defined as the set of relational values corresponding to all pairs of positions within a subsequence. A relational word is flexible if the constraint on a relational value is to belong to a set of relational values. We present here an algorithm, called KMRCRelat, which is derived from a previous algorithm for identification of repeated flexible words. KMRCRelat find either the k-length or the longest repeated flexible relational words in a sequence or a set of sequences