Abstract:
PROBLEM TO BE SOLVED: To provide improved technology to update customer's signature or other types of records in a data base system. SOLUTION: The customer's signature or other types of records are updated by using event driving type estimation quantity based on a model of a time- dependent behavior in the data base system. In a preferable form of embodiment, the latest version of a record to be affected by prescribed transaction is retrieved from a memory of the system, the event driving type estimation quantity of at least one of transaction rate and period probability to the record are determined at least partially based on a dynamic Poisson timing model having the transaction rate on the basis of period corresponding to many related periods. The event driving type estimation quantity is allowed to be constituted so as to generate an estimated transaction rate λjin for a period j and a prescribed transaction n and after that to generate an estimated period probability λjm/Σkλkm for the period j setting πj,n. And an updated version of the record is allowed to be generated based on the event driving type estimation quantity.
Abstract:
Disclosed is a method for maintaining compact descriptions of the behavior o f customers in a database of information relating to customer transactions. Of all possible variables that might be used to characterize customer transactions, a limited number are selected on the basis of discriminating power to define customer profiles. Customers are classified according to distinguishing features of their profiles. A new customer can be initially assigned to a particular class based on data characterizing his first few transactions.
Abstract:
Disclosed is a method for maintaining compact descriptions of the behavior of customers in a database of information relating to customer transactions. Of all possible variables that might be used to characterize customer transactions, a limited number are selected on the basis of discriminating power to define customer profiles. Customers are classified according to distinguishing features of their profiles. A new customer can be initially assigned to a particular class based on data characterizing his first few transactions.
Abstract:
L'invention concerne un procédé permettant de mettre à jour des descriptions compactes du comportement de clients dans une base de données d'informations concernant les transactions des clients. Parmi toutes les variables qu'il est possible d'utiliser pour caractériser les transactions des clients, on en a sélectionnées un nombre restreint sur la base du pouvoir discriminatif propre à définir les profils des clients. Les clients sont classés conformément aux traits caractéristiques de leurs profils. Un nouveau client peut être initialement associé à une classe spécifique en fonction des données qui caractérisent quelques-unes de ses premières transactions.