Machine learning for marketing of branded consumer products
View Patent ↗A method including retrieving a product information from a database is provided. The method includes associating the product information with multiple classification values, forming a vector associated with a consumer product. The classification values form coordinates of the vector in a vector space that comprises multiple vectors associated with multiple consumer products. The method includes determining a cluster in the vector space, including at least one vector selected according to a relative distance within a cluster boundary. The method includes selecting a discriminator vector from a vector difference between a first vector in a first cluster in the vector space and a second vector in a second cluster in the vector space and identifying a new consumer product associated with a new vector that is formed by adding the discriminator vector to a third vector from the vector space, the third vector associated with a known consumer product.
1. A computer-implemented method comprising:
retrieving, in a server, a product information from a database, the product information associated with a consumer product;
associating, with a marketability scoring subsystem in the server, the product information with multiple classification values;
forming, with a neural network in the server, a vector associated with the consumer product, wherein each classification value forms a coordinate of the vector in a vector space that comprises multiple vectors associated with multiple consumer products;
determining, with the neural network in the server, a cluster in the vector space, the cluster comprising at least one vector selected according to a relative distance within a cluster boundary;
selecting, with the neural network in the server, a discriminator vector from a vector difference between a first vector in a first cluster in the vector space and a second vector in a second cluster in the vector space, wherein the discriminator vector is indicative of a direction in the vector space that classifies a desirable attribute of the consumer product;
identifying, with the marketability scoring subsystem, a modified consumer product associated with a new vector that is formed by adding or subtracting the discriminator vector to a third vector from the vector space, the third vector associated with a known consumer product, wherein the modified consumer product comprises the consumer product with the desirable attribute;
causing a display in a client device communicatively coupled to the server to display one of a chart, a value, or a graphic indicative of a modified attribute based on the discriminator vector; and
providing, to a graphical user interface in the client device, a controllable signal to receive a user input to predict a content and a value of a shopping basket of a consumer including the modified consumer product based on the modified attribute, wherein determining a cluster in vector space comprises expanding, with a neural network model, a dimensionality of the cluster in an orthogonal dimension based on the user input to the controllable signal.
2. The computer-implemented method of claim 1 , wherein associating the product information with multiple classification values comprises using the neural network in the server, wherein the neural network in the server is trained on a product information from the database.
3. The computer-implemented method of claim 1 , further comprising predicting a shopping basket content and a shopping basket value for a consumer based on the product information, a purchasing history of the consumer, and the cluster in the vector space.
4. The computer-implemented method of claim 1 , further comprising selecting the discriminator vector and the new vector based on an increased predicted value of a shopping basket that includes the modified consumer product.
5. The computer-implemented method of claim 1 , further comprising selecting a classification value based on a consumer history associated with the consumer product, the consumer history retrieved from a brand database or a retailer database.
6. The computer-implemented method of claim 1 , further comprising predicting a responsiveness of a consumer to the modified consumer product based on a responsiveness of the consumer to consumer products associated with vectors in a same cluster as the new vector.
7. The computer-implemented method of claim 1 , further comprising providing to a consumer an offer for a consumer product associated with a vector in the first cluster, wherein a shopping basket comprising the consumer product has a higher value than a shopping basket comprising the known consumer product.
8. The computer-implemented method of claim 1 , further comprising adjusting a price of the consumer product according to an expected value of a shopping basket for an aggregated group of consumers, the expected value based on a value of a shopping basket comprising the known consumer product when the consumer product and the known consumer product belong in a same cluster.
9. The computer-implemented method of claim 1 , further comprising determining a stock value of a retail inventory for the consumer product based on a responsiveness of an aggregated group of consumers to a different consumer product.
10. The computer-implemented method of claim 1 , further comprising selecting multiple consumer products belonging in different clusters.
11. A system, comprising:
a memory comprising instructions; and
one or more processors configured to execute an instruction to:
retrieve, in a server, a product information from a database, the product information associated with a consumer product;
associate, with a marketability scoring subsystem in the server, the product information with multiple classification values;
form, with a neural network in the server, a vector associated with the consumer product, the vector having a classification value as a coordinate in a vector space comprising multiple vectors associated with multiple consumer products;
determine, with the neural network in the server, a cluster in the vector space, the cluster comprising at least one vector selected according to a relative distance within a cluster boundary;
select, with the neural network in the server, a discriminator vector from a vector difference between a first vector in a first cluster in the vector space and a second vector in a second cluster in the vector space, wherein the discriminator vector is indicative of a direction in the vector space that classifies a desirable attribute of the consumer product;
identify, with the marketability scoring subsystem, a modified consumer product associated with a new vector that is formed by adding the discriminator vector to a third vector from the vector space, the third vector associated with a known consumer product, wherein the modified consumer product comprises the consumer product with the desirable attribute;
cause a display in a client device communicatively coupled to the server to display one of a chart, a value, or a graphic indicative of a modified attribute based on the discriminator vector; and
provide, to a graphical user interface in the client device, a controllable signal to receive a user input to predict a content and a value of a shopping basket of a consumer including the modified consumer product based on the modified attribute, wherein to determine a cluster in the vector space the one or more processors execute an instruction to expand, with a neural network model, a dimensionality of the cluster in an orthogonal dimension based on the user input to the controllable signal.
12. The system of claim 11 , wherein the one or more processors further execute instructions to predict a shopping basket content and a shopping basket value for a consumer based on the product information, a purchasing history of the consumer, and the cluster in the vector space.
13. The system of claim 11 , wherein the one or more processors further execute instructions to select the discriminator vector and the new vector based on an increased predicted value of a shopping basket that includes the modified consumer product.
14. The system of claim 11 , wherein the one or more processors further execute instructions to select a classification value based on a consumer history associated with the consumer product, the consumer history retrieved from a brand database or a retailer database.
15. The system of claim 11 , wherein the neural network in the server is trained on historical product information from the database.
16. The system of claim 11 , wherein the one or more processors further execute instructions to predict a responsiveness of a consumer to the modified consumer product based on a responsiveness of the consumer to consumer products associated with vectors in a same cluster as the new vector.