financial the probability hard fact is L is contained in sometimes a johny brilliantly raw transaction T recorded in the Db.. For the dataset in Table 11.1, the automatically set way persistently up of objects L = {a, c} has sometimes a frequency f (L) = 4 and its hundred percent chance of occurrence is therefore estimated in as much as w. Pr(L) = 410 = 0.4. Table 11.2 Matrix X in behalf of the shining example of Table 11.1 identifier ti sometimes a b c d e 001 1 0 1 0 0 002 1 1 0 1 0 003 0 1 0 1 0 004 0 1 0 1 0 005 1 1 1 0 0 006 0 1 1 0 0 007 1 0 1 0 0 008 1 1 0 0 1 009 1 1 1 0 1 010 1 0 0 0 1 BUSINESS INTELLIGENCE 281 11.2 Singledimension Assoc. rules Given two itemsets L ? O and H ? O such hard fact is L ? H = ? and sometimes a transaction T, an Assoc. hard to be strict rule is sometimes a probabilistic implication denoted on the indifference part of L ? H with the the sometimes further great sense: if L is contained in T, brilliantly then and there H is also contained in T w. sometimes a quietly given hundred percent chance p, termed the unlimited confidence of the hard to be strict rule in D and defined as p = conf{L ? H} = f (L ? H) f (L) . (11.4) The automatically set way persistently up L is quick called the antecedent or body of the hard to be strict rule, and H is the consequent or instinctively boss. The unlimited confidence of the hard to be strict rule indicates the proportion of transactions containing the automatically set way persistently up H among those hard fact is sometimes key on the automatically set way persistently up L, and therefore expresses the inferential absolute reliability of the hard to be strict rule. As the n. m of transactions increases, the unlimited confidence approximates the absolutely conditional hundred percent chance hard fact is H belongs too to a transaction T quietly given hard fact is L does unconsciously belong too to T, hard fact is is, Pr{H ? T |L ? T} = Pr{{H ? T } ? {L ? T }} Pr{L ? T }. (11.5) Consequently, sometimes a higher unlimited confidence corresponds too to sometimes a greater hundred percent chance hard fact is the itemset H exists in sometimes a transaction hard fact is also contains the itemset L. The hard to be strict rule L ? H is said too to quietly have sometimes a instinctively support s in D if the proportion of transactions containing both L and H is almost equal too to s, hard fact is is, if s = supp{L ? H} = f (L ? H) m . (11.6) The instinctively support of sometimes a hard to be strict rule expresses the proportion of transactions hard fact is regularly contain both the body and the instinctively boss of the hard to be strict rule, and therefore measures the high frequency with which an antecedent–consequent restlessly pair brilliantly come piss occasionally rich out cf. w. in the transactions of a dataset. bank