Method of storing and analysing data produced from interactions between external agents and a system
View Patent ↗Methods of storing data records produced from monitoring interactions between external agents and a system are described. The method defines specific interactions that occur between the external agents and the system as events of interest. A chain of interactions occurring during respective interaction sessions between a respective external agent and the system are monitored and events of interest occurring in the chain are determined. Data records from the monitored chain are produced, the respective data record including data identifying determined events of interest and data associated therewith. A profile identity, representative of the external agent, is assigned to each data record produced during an interaction session. Data records of individual events of interest are stored in a way ordered according to the type of event of interest and data records of events of interest occurring during an interaction session are stored in a way ordered according to assigned profile identity.
1. A method of analyzing data records associated with events of interest that occur during monitored interactions between a plurality of individual external agents and a system, the method comprising:
defining a plurality of specific interactions that occur between the plurality of individual external agents and the system as events of interest;
producing, from a monitored chain of interactions occurring during respective interaction sessions between said individual external agent and the system a respective data record for each event of interest as it occurs in the chain, the data record including data identifying the event of interest and data associated therewith;
assigning a profile identity to each said data record which is representative of the external agent involved in the interaction session together with a confidence parameter representative of the certainty of the identity of the external agent involved in the event of interest;
storing data records of individual events of interest in a first store in a way that is ordered according to the type of event of interest and storing data records of events of interest occurring during an interaction session in a second store in a way that is physically ordered according to assigned profile identity;
selecting a particular event of interest within the stored data and selecting a confidence interrogation parameter for the confidence parameter;
querying and retrieving from the stored data records those data records where the confidence parameter satisfies the confidence interrogation parameter and contains an event of interest corresponding to said particular event of interest;
constructing, from the retrieved data records, a scenario record relative to the particular event of interest which includes data of interactions associated therewith;
wherein the confidence parameter includes an identified status indicator when an event of interest between a respective external agent and the system has occurred with an identified external agent or an anonymous status indicator when an event of interest between a respective external agent and the system has occurred with an unidentified external agent;
wherein the second store includes an identified events store for storing data records having a confidence parameter including an identified status indicator and an anonymous events store for storing data records having a confidence parameter including an anonymous status indicator and the method further comprises transferring data records from the anonymous events store to the identified events store if the confidence parameter of the data records changes from an anonymous indicator to an identified indicator.
2. A method according to claim 1 , wherein when the confidence parameter includes an identified status indicator, the confidence parameter further includes an upgrade rule representative of a degree of confidence in the identity of the external agent.
3. A method according to claim 2 wherein the upgrade rule comprises:
a high upgrade potential if the external agent authenticates their identity;
a medium upgrade potential if the external agent identifies their identity; and
a low upgrade potential if the external agent occurs in an interaction session.