Abstract: A method of an attribution server. The method determines publishing channels for advertisements in a marketing campaign to analyze their marketing effectiveness for purchasable items using a processor and a memory of the attribution server. Data points are associated with users. A K-th order attribution model is constructed. Independent and dependent variables of the attribution model are associated with various types of marketing data. An observation matrix and a conversion vector are determined. A regression analysis is performed with refining steps. Insignificant second order cross terms of the attribution model are identified and removed. A modified K-th order attribution model is constructed. Another regression analysis is performed to find optimal model parameters.

Abstract: A method of encoding sequential data that allows encoding a subsequence of full sequences as a composite data symbol, wherein a subsequence is comprised of a maximum of one original data element, and a maximum of K original data elements. These composite data symbols, arranged sequentially, can then be used to train a machine learning model, and thus reduce complexity when a strict ordering within the context of the original data subsequences is not required, while still modeling synergies between the sequential data elements. Further, the method determines a set of related data elements to a composite symbol at the next time step, given the original subsequence.