Systems and methods for providing personalized online content
A system for providing online-content includes a data-acquisition module configured to collect user data and product data. A configuration module determines a predictive algorithm and one or more filters based on received marketer-input. A recommendation engine generates recommendations in real-time, in response to a request. The recommendations are generated via the predictive algorithm, which generates scalar scores from the dot-product multiplication of at least one user-vector and at least one product-vector, the scalar score indicating the likelihood that a desired interaction will occur between the user and the product. The algorithm also determines a hierarchical list based on the scalar scores, and applies the filters to the hierarchical list so as to identify the recommendations. The recommendation engine can generate recommendations in two ways: recommend items based on users, and recommend users based on items. A rendering module generates a recommendation block based on the recommendations, which is transmitted to a user-computer in response to the request.
1. A server system for providing personalized online-content, the server system including one or more server computers, the server system comprising:
a data-acquisition module implemented via the one or more server computers, the data-acquisition module configured to collect user data and product data;
a configuration module implemented via the one or more server computers, the configuration module configured to determine a predictive algorithm and one or more filters based on received marketer-input wherein the configuration module includes a sentence-based user interface;
a recommendation engine implemented via the one or more server computers, the recommendation engine configured to generate one or more recommendations in real-time, in response to a request, wherein the recommendations are generated via the predictive algorithm, which includes the steps of:
generating scalar scores from the dot-product multiplication of at least one user-vector and at least one product-vector, the user-vector being generated from the user data, the product-vector being generated from the product data, and the scalar score indicating the likelihood that a desired interaction will occur between the at least one user and the at least one product,
determining a hierarchical list based on the scalar scores, and
applying the one or more filters to the hierarchical list so as to identify the one or more recommendations;
a rendering module implemented via the one or more server computers, the rendering module configured to generate a recommendation block based on the one or more recommendations, wherein the recommendation block is a digital graphical representation of a product; and
a network interface connecting the one or more server computers to a network, the network interface configured to transmit the generated recommendation block to a user-computer, coupled to the network, in response to the request.
2. The system of claim 1 , wherein the scalar scores are generated from the dot-product multiplication of one user-vector and a plurality of product-vectors such that the hierarchical list is of products, the one or more recommendations are for recommended products, and the recommendation block is generated from the product data associated with the one or more recommended products.
3. The system of claim 2 , wherein the hierarchical list is generated by comparing the scalar scores associated with each product.
4. The system of claim 1 , wherein the scalar scores are generated from the dot-product multiplication of a plurality of user-vectors and one product-vector such that the hierarchical list is of users, the one or more recommendations are for recommended users, and the recommendation block is generated from the user data associated with the one or more recommended users.
5. The system of claim 4 , wherein the hierarchical list is generated by comparing the scalar scores associated with each user.
6. The system of claim 1 , wherein the system further comprises:
a training engine configured to determine, independently of the request, the at least one product-vector from the product data, the at least one product-vector being stored in a memory and accessible to the recommendation engine.
7. A method for providing personalized online-content, the method comprising:
collecting user data and product data, via a data-acquisition module of one or more server computers;
determining a predictive algorithm and one or more filters based on received marketer-input, via a configuration module of one or more server computers, wherein the configuration module includes a sentence-based user interface;
receiving a request for one or more recommendations, via a recommendation engine of one or more server computers;
generating one or more recommendations in real-time, via the recommendation engine, in response to the request, wherein the one or more recommendations are generated via the predictive algorithm, which includes the steps of:
generating scalar scores from the dot-product multiplication of at least one user-vector and at least one product-vector, the user-vector being generated from the user data, the product-vector being generated from the product data, and the scalar score indicating the likelihood that a desired interaction will occur between the at least one user and the at least one product,
determining a hierarchical list based on the scalar scores, and
applying the one or more filters to the hierarchical list so as to identify the one or more recommendations;
generating a recommendation block based on the one or more recommendations, via a rendering module of one or more server computers; and
transmitting the generated recommendation block to a user-computer, coupled to a network, in response to the request, via a network interface connecting the one or more server computers to the network.
8. The method of claim 7 , wherein the scalar scores are generated from the dot-product multiplication of one user-vector and a plurality of product-vectors such that the hierarchical list is of products, the one or more recommendations are for recommended products, and the recommendation block is generated from the product data associated with the one or more recommended products.
9. The method of claim 8 , wherein the hierarchical list is generated by comparing the scalar scores associated with each product.
10. The method of claim 7 , wherein the scalar scores are generated from the dot-product multiplication of a plurality of user-vectors and one product-vector such that the hierarchical list is of users, the one or more recommendations are for recommended users, and the recommendation block is generated from the user data associated with the one or more recommended users.
11. The method of claim 10 , wherein the hierarchical list is generated by comparing the scalar scores associated with each user.
12. The method of claim 7 , wherein the method further comprises: determining, independently of the request, the at least one product-vector from the product data, the at least one product-vector being stored in a memory and accessible to the recommendation engine.
13. A non-transitory computer readable medium having instructions stored thereon, the execution of which by a computer causes the computer to execute a method for providing personalized online-content, the method comprising:
collecting user data and product data, via a data-acquisition module of the computer;
determining a predictive algorithm and one or more filters based on received marketer-input, via a configuration module of the computer, wherein the configuration module includes a sentence-based user interface;
receiving a request for one or more recommendations, via a recommendation engine of the computer;
generating one or more recommendations in real-time, via the recommendation engine, in response to the request, wherein the one or more recommendations are generated via the predictive algorithm, which includes the steps of:
generating scalar scores from the dot-product multiplication of at least one user-vector and at least one product-vector, the user-vector being generated from the user data, the product-vector being generated from the product data, and the scalar score indicating the likelihood that a desired interaction will occur between the at least one user and the at least one product,
determining a hierarchical list based on the scalar scores, and
applying the one or more filters to the hierarchical list so as to identify the one or more recommendations;
generating a recommendation block based on the one or more recommendations, via a rendering module of the computer, wherein the recommendation block is a digital graphical representation of a product; and
transmitting the generated recommendation block to a user-computer, coupled to a network, in response to the request, via a network interface connecting the one or more server computers to the network.
14. The non-transitory computer readable medium of claim 13 , wherein the scalar scores are generated from the dot-product multiplication of one user-vector and a plurality of product-vectors such that the hierarchical list is of products, the one or more recommendations are for recommended products, and the recommendation block is generated from the product data associated with the one or more recommended products.
15. The non-transitory computer readable medium of claim 14 , wherein the hierarchical list is generated by comparing the scalar scores associated with each product.
16. The non-transitory computer readable medium of claim 13 , wherein the scalar scores are generated from the dot-product multiplication of a plurality of user-vectors and one product-vector such that the hierarchical list is of users, the one or more recommendations are for recommended users, and the recommendation block is generated from the user data associated with the one or more recommended users.
17. The non-transitory computer readable medium of claim 16 , wherein the hierarchical list is generated by comparing the scalar scores associated with each user.
18. The non-transitory computer readable medium of claim 13 , wherein the method further comprises:
determining, independently of the request, the at least one product-vector from the product data, the at least one product-vector being stored in a memory and accessible to the recommendation engine.