Speedy, Mini and Totally Fuzzy: Three Ways for Fuzzy Sequential Patterns Mining

Most real world databases are constituted from historical and numerical data such as sensors, scientific or even demographic data. In this context, algorithms extracting sequential patterns, which are well adapted to the temporal aspect of the data, do not allow processing numerical information. Therefore the data are pre-processed to be transformed into a binary representation which leads to a loss of information. Algorithms have been proposed to process numerical data using intervals and particularly fuzzy intervals. With regards to the search of sequential patterns based on fuzzy intervals, both existing methods are incomplete either in the processing of sequences or in the support computation. Therefore this paper proposes three methods to mine fuzzy sequential patterns (SPEEDYFUZZY, MINIFUZZY and TOTALLYFUZZY). We detail these algorithms by highlighting their different fuzzification levels. Finally we have assessed them through different experiments carried out on several datasets.