Systems and methods for identifying and counting instances of temporal patterns
View Patent ↗Systems and methods for instance counting and for the identification of a temporal pattern in an event sequence. The system addresses the use of “earliest-first” and “no-reuse” policies as criteria for the correctness of counting. The system also achieves higher performance than conventional methods by utilizing incremental computation.
1. A computer implemented temporal data-mining system comprising:
an arrangement for counting and identifying instances of temporal patterns;
said counting and identifying arrangement comprising:
at least one component which identifies temporal pattern instances; and
an arrangement for caching events; and
said counting and identifying arrangement being adapted to implement at least one counting policy related to the identification of temporal pattern instances, wherein the at least one counting policy includes an earliest-first policy and a no-reuse policy.
2. The system according to claim 1 , wherein said counting and identifying arrangement employs a sequential algorithm for counting and identifying temporal pattern instances.
3. The system according to claim 2 , wherein the algorithm is adapted to maintain a local cache of stored events from a predetermined number or previous time slots.
4. The system according to claim 2 , wherein the algorithm employs a per-pattern data structure for keeping track of the state of each temporal pattern.
5. The system according to claim 2 , wherein the algorithm includes an indexing scheme for fast-retrieving temporal patterns related to an item.
6. A computer implemented method of facilitating temporal data-mining, said method comprising the steps of:
counting and identifying instances of temporal patterns;
said counting and identifying step comprising:
identifying temporal pattern instances;
caching events; and
implementing at least one counting policy related to the identification of temporal pattern instances:
said step of implementing at least one counting policy comprising implementing an earliest-first policy and a no-reuse policy.
7. The method according to claim 6 , wherein said counting and identifying step comprises applying a sequential algorithm for counting and identifying temporal pattern instances.
8. The method according to claim 7 , wherein said step of applying a sequential algorithm comprises maintaining a local cache of stored events from a predetermined number of previous time slots.
9. The method according to claim 7 , wherein said step of applying a sequential algorithm comprises employing a per-pattern data structure for keeping track of the state of each temporal pattern.
10. The method according to claim 7 , wherein said step of applying a sequential algorithm comprises employing an indexing scheme for fast-retrieving temporal patterns related to an item.
11. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for facilitating computer implemented data-mining, said method comprising the steps of:
counting and identifying instances of temporal patterns;
said counting and identifying step comprising:
identifying temporal pattern instances;
caching events; and
implementing at least one counting policy related to the identification of temporal patterns instances;
said step of implementing at least one counting policy comprising complementing an earliest-first policy and as no-reuse policy.