Back to results
US 6,507,851 B1Granted
Customer information retrieving method, a customer information retrieving apparatus, a data preparation method, and a database
AssigneeSony Corporation
InventorsYoshihito Fujiwara, Naotaka Ando
Life of this patent
Estimating the term…
Abstract
The improvement of accuracy is provided in extracting a rule in customers. A customer database DB1 is constructed by storing customer inner information concerning the mental side, which includes a value sense of each customer, and customer outer information which includes basic attributes consisting of at least sexuality and age of each customer.
Claims4
A database used for retrieving customer information comprising:
- a customer database storing a plurality of data items;
- a learning database storing data items common to said customer database and data items not stored in said customer database;
- and
- a virtual database storing inherent data items inherent to said learning database;
- wherein,
- said inherent data items are obtained by estimation using a learning model to said data items common to said customer database and said data items not stored in said customer database, said estimation being performed via a recursive solution based on a determination tree or via a back provocation algorithm;
- said data items of said customer database comprise responses to questions by customers listed in said customer database;
- and
- said data items in said learning database not stored in said customer database comprise responses to different questions by persons extracted at random or on a classification basis.
6 more elements
A computer-implemented method of generating a virtual database for retrieving customer information, comprising the steps of:
- generating a customer database storing a plurality of first data items for each of a plurality of customers;
- generating a learning database storing data items common to said customer database and second data items not stored in said customer database;
- and
- estimating inherent data items of said learning database and storing said estimated data items in a virtual database;
- wherein,
- said estimating is performed using a learning model to said first data items common to said customer database and said second data items not stored in said customer database, said estimating being performed via a recursive solution based on a determination tree or via a back provocation algorithm;
- said first data items of said customer database comprise responses to questions by customers listed in said customer database;
- and
- said second data items in said learning database not stored in said customer database comprise responses to different questions by persons extracted at random or on a classification basis.
6 more elements
Dependencies are inferred from the claim text.