Bayesian product recommendation engine
The invention provides a method of generating a recommendation for a product type. A plurality of product attributes associated with the product type is provided. A sampled set of consumer values is received. At least one user preference corresponding to the product attributes is received. A plurality of consumer values based on the at least one user preference and the sample set of consumer values is calculated, and at least one product recommendation is determined based on the calculated consumer values.
1. A method of generating a product recommendation for a product type, using a product recommendation engine including an application server, at least one graphical user interface, and a user computer, as well as a measured consumer values database, a product definition database, and a user database, the method comprising said customer computer performing the steps of:
receiving at the user interface a user selection of a product type;
retrieving from the product definition database a plurality of product attributes associated with the product type;
receiving a sampled set of consumer values from the measured consumer values database;
receiving at least one user preference corresponding to the product attributes from at least one of the user interface and the user database;
calculating a plurality of consumer values based on the at least one user preference and the sampled set of consumer values by determining a likelihood function based on the at least one user preference, calculating an initial Gaussian probability density function based on the sampled set of consumer values, calculating a posterior Gaussian probability density function based on the likelihood function and the initial Gaussian probability density function, and calculating a plurality of consumer values based on the posterior Gaussian probability density function;
determining at least one product recommendation based on the calculated consumer values; and
providing the at least one product recommendation to the user via the user interface.
2. The method of claim 1 wherein the product type comprises mobile vehicles.
3. The method of claim 1 wherein the product type is selected from the group consisting of digital cameras, personal computers, personal digital assistants, computer printers, cell phones, televisions, cameras, entertainment systems, power tools, MP3 players, DVD players, consumer electronic products, consumer appliances, consumer goods, consumer services, office equipment, business software, credit cards, home mortgages, car loans, health insurance, restaurants, telephone services, hotel rooms, airplane tickets, vacation packages, a consumer-valued product, and a consumer-valued service.
4. The method of claim 1 wherein the plurality of product attributes includes at least one attribute level associated with each product attribute.
5. The method of claim 1 wherein the user preference is received via a web interface.
6. The method of claim 1 wherein the likelihood function is characterized by a linear form and an additive Gaussian noise function with a variance based on the sampled set of consumer values.
7. The method of claim 1 wherein the posterior Gaussian probability density function is characterized by a mean vector and a precision matrix.
8. The method of claim 1 wherein determining at least one product recommendation comprises computing a total value for each of a set of products in a product-definition database, sorting the total values to produce an ordered list of total values, and associating a product in the product-definition database with the ordered list of total values.
9. The method of claim 1 further comprising:
displaying the last least one product recommendation based on the calculated consumer values.
10. The method of claim 1 further comprising:
generating a product type market study based on the plurality of calculated consumer values.
11. A non-transitory computer usable medium including a program to generate a product recommendation for a product type, comprising:
computer program code to providing a plurality of product attributes associated with the product type;
computer program code to receive a sampled set of consumer values;
computer program code to received at least one user preference corresponding to the product attributes;
computer program code to calculate a plurality of consumer values based on the at least one user preference and the sampled set of consumer value by determining a likelihood function based on the at least one user preference, calculating an initial Gaussian probability density function based on the sampled set of consumer values, calculating a posterior Gaussian probability density function based on the likelihood function and the initial Gaussian probability density function, and calculating a plurality of consumer values based on the posterior Gaussian probability density function; and
computer program code to determine at least one product recommendation based on the calculated consumer values.
12. The computer usable medium of claim 11 , further comprising:
computer program code to display the at least one product recommendation based on the calculated consumer values.
13. The computer usable medium of claim 11 , further comprising:
computer program code to receive a product-recommendation request; and
computer program code to display the at least one product recommendation based on the product-recommendation request and the calculated consumer values.
14. The computer usable medium of claim 11 , further comprising:
computer program code to generate a product type market study based on the plurality of calculated consumer values.