Probabilistic Fuzzy Frequent Item Sets Mining (PPFIM)

The double superposition of fuzziness and randomness makes the mining of uncertain fuzzy item sets more complex. Based on the calculation of item set fuzziness, this paper integrates the randomness of possible world model to solve the problem of mining fuzzy frequent item sets based on probability threshold. Fuzzy theory and uncertainty are based on linguistic information and are extended to deal with partial truth concepts. In the fuzzy probabilistic context, an itemset
$X$
is called fuzzy frequent if the fuzzy probability that
$X$
occurs in at least minSup transactions is above a given threshold
$\tau$
. We designed a dynamic programming-based algorithm to compute the frequent fuzzy probability. Experiments were conducted to evaluate the performance of the proposed approach.