The transaction index is n∈{1,…,N} for N transactions in the system.
The only difference is that the e-coins used are different for each transaction (n transactions, n coins).
Given a data set D of n transactions where each transaction T∈ D. Let I = {I 1, I 2,…, I n } be a set of items.
Given a data set D of n transactions where each transaction TЄ D. Let I = {I1, I2, … In} is a set of items.
The Cardholder's reputation for a forthcoming transaction n can be calculated as weighted average of the last N transactions limited to the range <RMIN,RMAX>, see Eq. (1).
Formally speaking, given (T={t_1,t_2,ldots,t_n}) as a dataset of n transactions, where each transaction (t_i) contains items, e.g., (t_i = {I_{i1},I_{i2},ldots,I_{ik}}) and each item (I_{ij} in I) the set of all possible items.
It finds frequent sets of items (i.e., combinations of items that are purchased together in at least N transactions in the database), and from the frequent items sets such as {X, Y}, generates association rules of the form: X → Y and/or Y → X.
With such a dataset, we can learn B0 from N seq observations of initial slice, and learn B→ by N = ∑ l N l transactions of transition slices.
The PF scheduling weight w n of transaction n is determined as follows w n = c n c n ¯ (10).
