SYSTEM & METHOD

• US 20100131433A1
• Filed: 03/12/2008
• Published: 05/27/2010
• Est. Priority Date: 04/04/2007
• Status: Abandoned Application

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Majeed, Basim, Sanchez Garcia, Felix, Nauck, Detlef D.
• Application Number:
  US12/594,714
• Publication Number:
  US 20100131433A1
• Time in Patent Office: Days
• US Class Current: 706/5
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 7/02 using fuzzy logic computing arrangements based on biological models G06N3/00; computing arrangements using knowledge-based models G06N5/00

Abstract

The present invention relates to a system and a method for automatically determining the relationship between a plurality of tasks comprising a process and then detecting any abnormalities in the execution of the process.

Claims

1. A method of monitoring a process, the method comprising the steps of:

a) receiving a first process data set from one or more process logs, wherein the first process data set comprises a first plurality of data objects and each of the first plurality of data objects is associated one of a plurality of attributes;

b) identifying associations between each of the plurality of data attributes and one or more of the further data attributes to form an association network;

c) receiving a second process data set from one or more process logs, wherein the second process data set comprises a second plurality of data objects; and

d) comparing the second plurality of data objects with the association network.

2. A method according to claim 1, in which the association network is updated in accordance with the results of the comparison made in step d).

3. A method according to claim 1, comprising the further step of:

e) the first process data set is transformed such that and each of the first plurality of data objects is represented by a fuzzy logic function;

wherein step e) is carried out after step a) but before step b).

4. A method according to claim 1, comprising the further step of:

f) the first process data set is transformed such that a further plurality of the plurality of data objects are replaced by one or more composite data objects, wherein each composite data objects is generated by combining one ore more of the plurality of data objects;

wherein step f) is carried out after step a) but before step b).

5. A method according to claim 3, comprising the further step of

g) creating one or more composite fuzzy logic functions, the or each composite fuzzy logic functions comprising one ore more fuzzy logic functions, wherein step g) is carried out after step b) but before step c).

6. A method according to claim 1, wherein an alert is generated in accordance with the results of the comparison made in step d).

7. A computer program product, comprising computer executable code for performing a method according to claim 1.

8. A system comprising:

a learning component configured to receive process data from one or more process logs, a detection component and an association network wherein, in use, the system such that;

the learning component receives a first process data set from the one or more process logs, wherein the first process data set comprises a first plurality of data objects and each of the first plurality of data objects is associated one of a plurality of attributes;

the learning component identifies associations between each of the plurality of data attributes and one or more of the further data attributes to form the association network;

the detection component receives a second process data set from one or more process logs, wherein the second process data set comprises a second plurality of data objects; and

detection component compares the second plurality of data objects with the association network.