| CPC G06N 20/00 (2019.01) [G06F 7/32 (2013.01); G06F 17/16 (2013.01); G06N 7/00 (2013.01); G06N 20/20 (2019.01)] | 17 Claims |
|
1. A loyalty extraction machine, comprising:
an input module configured to receive sample data (x);
a data collection module connected to the input module and configured to store a collection of data (Ω) from the input module, the collection of data (Ω) including a training set (Ωtr) and/or a test set (Ωtt);
a multiform separation engine connected to the data collection module and configured to generate 2m developed classifiers (ŷβ,1, . . . , ŷβ,m, ŷγ,1, . . . , ŷγm), m≥2;
wherein the 2m developed classifiers are combined to form a vector function V(x)=(ŷβ,1(x), . . . , ŷβ,m(x), ŷγ,1(x), . . . , ŷγ,m(x)), wherein an outcome of the vector function (V) is a 2m-dimensional vector and named an identity (I);
wherein, for generating the 2m developed classifiers (ŷβ,1, . . . , ŷβ,m, ŷγ,1, . . . , ŷγm), the multiform separation engine is configured to use m piecewise continuous functions to perform classification; the m piecewise continuous functions are respectively trained with the training set Ωtr through a training process;
wherein in the training process, m weighted cost functions Φβ,1, . . . , Φβ,m with a lighter weight β where 0<β<1 and m weighted cost functions ΦY,1, . . . , ΦY,m with a heavier weight γ where γ>1 provide performance measures for separating the training subsets Ωtr(1), . . . , Ωtr(m), and are used respectively to derive member functions f1β,j, . . . , fmβ,j, j=1, . . . , m, and f1γ,j, . . . fmγ,j, j=1, . . . , m;
a classifier combination module connected to the multiform separation engine and the data collection module and configured to combine the 2m developed classifiers (ŷβ,1, . . . , ŷβ,m, ŷγ,1, . . . , ŷγm), the developed classifiers (ŷβ,1, . . . , ŷβ,m, ŷγ,1, . . . , ŷγm) being trained with the training set (Ωtr);
an output module connected to the classifier combination module and configured to derive an output result after the sample data (x) is processed through the classifier combination module; and
a loyalty type indicator associated with the output module and configured to determine loyalty type of a sample data (x) by confirming a location of the lighter weight (β) and/or the heavier weight (γ) in an identity (I);
wherein the loyalty extraction machine is realized as hardware or software in separated circuit devices on a set of chips or an integrated circuit device on a single chip, and the loyalty extraction machine is implemented in a cloud server or a local computer.
|