Diagnostic method and system utilizing historical event logging data
Disclosed is a diagnostic method and system including the processing of historical event logs generated by one or more devices. According to an exemplary embodiment, a diagnostic system includes an event log acquisition module, an event classification module classifying event logs acquired, and a diagnostic module generating a labeled version of the historical event log including labels provided by the classification module. The event classification module is trained using supervised machine learning techniques.
1. A computer-implemented method of processing a historical event log generated by an image processing device, the historical event log including a historical record of each occurrence of a plurality of events associated with an execution of one or both of hardware and software associated with the image processing device and the historical record including event data associated with each event, the method comprising:
a) an event log acquisition module acquiring all or part of the historical event log generated by the image processing device;
b) accessing an event classifier module to generate a label for each event included in the acquired historical event log with one of a plurality of predetermined labels, the event classifier module trained with a plurality of other historical event logs including manually labeled event data associated with an occurrence of a plurality of events recorded by one or more of the image processing device and other image processing devices and the event classifier module trained by extracting features from the plurality of other historical event logs predictive of failures and diagnostics associated with the one or more of the image processing device and other image processing devices, the extracted features including collocational features and bag-of-words features associated with event data included in the plurality of other historical event logs; and
c) a diagnostic module generating a labeled version of the historical event log, the labeled version of the historical event log including the historical record of each occurrence of the events associated with the image processing device and a respective label generated by the event classifier module for each occurrence of the events associated with the image processing device,
wherein the plurality of events are associated with a plurality of image processing device states associated with the image processing device, and the predetermined labels uniquely identify each of the plurality of image device states,
wherein the plurality of image processing device states are associated with a plurality of image processing device fault states, the fault states and the plurality of predetermined labels including two or more of error, no error, warning, okay state, critical, major, ordinary, and nonclassified, and
wherein the image processing device is one of a MFD (Multi-Function Device), a printer, a scanner, and a fax machine.
2. The computer-implemented method of processing a historical event log according to claim 1 , wherein steps a)-c) are performed as a remotely located web service operatively associated with the image processing device.
3. The computer-implemented method of processing a historical event log according to claim 1 , step a) comprising:
a1) an event log acquisition module acquiring all or part of the historical event log generated by the image processing device; and
a2) an event log parsing module processing the historical event log to remove events associated with one or more predefined event types.
4. The computer-implemented method of processing a historical event log according to claim 1 , wherein the event classifier module module generates feature vectors representative of each event and determines the label for each event based on the feature vector representative of the event.
5. The computer-implemented method of processing a historical event log according to claim 1 , wherein the event classifier module is trained using one of a Maximum Entropy method, a Nave Bayes method and a Decision Tree method.
6. The computer-implemented method of processing a historical event log according to claim 1 , wherein the image processing device generates a plurality of historical event logs including historical event logs generated by one or more sub-processes executed by the image processing device.
7. The computer-implemented method of processing a historical event log according to claim 1 , wherein a DMA (Device Management Application) operatively associated with the image processing device communicates with a remote web server application to initiate the execution of steps a)-c) by the remote web server.
8. A diagnostic system for processing a historical event log generated by an image processing device, the historical event log including a historical record of each occurrence of a plurality of events associated with an execution of one or both of hardware and software associated with the image processing device and the historical record including event data associated with each event, the diagnostic system comprising:
an event log acquisition module configured to acquire all or part of the historical event log generated by the image processing device;
an event classifier module configured to generate a label for each event included in the acquired historical event log with one of a plurality of predetermined labels, the event classifier module trained with a plurality of other historical event logs including manually labeled event data associated with an occurrence of a plurality of events recorded by one or more of the image processing device and other image processing device and the event classifier module trained by extracting features from the plurality of other historical event logs predictive of failures and diagnostics associated with the one or more of the image processing device and other image processing devices, the extracted features including collocational features and bag-of-words features associated with event data included in the plurality of other historical event logs; and
a diagnostic module configured to generate a labeled version of the historical event log, the labeled version of the historical event log including the historical record of each occurrence of the events associated with the image processing device and a respective label generated by the event classifier module for each occurrence of the events associated with the image processing device
wherein the plurality of events are associated with a plurality of image processing device states associated with the image processing device, and the predetermined labels uniquely identify each of the plurality of image device states,
wherein the plurality of image processing device states are associated with a plurality of image processing device fault states, the fault states and the plurality of predetermined labels including two or more of error, no error, warning, okay state, critical, major, ordinary, and nonclassified, and
wherein the image processing device is one of a MFD (Multi-Function Device), a printer, a scanner, and a fax machine.
9. The diagnostic system for processing a historical event log according to claim 8 , wherein the event log acquisition module, event classifier module and diagnostic module are remotely located in a web server operatively associated with the image processing device.
10. The diagnostic system for processing a historical event log according to claim 8 , comprising:
an event log parsing module configured to process the historical event log to remove events associated with one or more predefined event types.
11. The diagnostic system for processing a historical event log according to claim 8 , wherein the event classifier module is configured to generate feature vectors representative of each event and determine the label for each event based on the feature vector representative of the event.
12. The diagnostic system for processing a historical event log according to claim 8 , wherein the event classifier module is trained using one of a Maximum Entropy method, a Naïve Bayes method and a Decision Tree method.
13. The diagnostic system for processing a historical event log according to claim 8 , wherein the image processing device generates a plurality of historical event logs including historical event logs generated by one or more sub-processes executed by the image processing device.
14. The diagnostic system for processing a historical event log according to claim 8 , further comprising:
a DMA (Device Management Application) module operatively associated with the image processing device, the DMA configured to communicate with a remote web server application, the remote web server application including the event log acquisition module, the event classifier module and the diagnostic module.
15. A diagnostic system for processing a plurality of historical event logs generated by a plurality of networked image processing devices, each historical event log including a historical record of each occurrence of a plurality of events associated with an execution of one or both of hardware and software associated with a respective image processing device and the historical record including event data associated with each event, the diagnostic system comprising:
an event log acquisition module configured to acquire all or part of the plurality of historical event logs generated by the plurality of networked image processing devices;
an event classifier module configured to generate a label for each event included in the plurality of historical event logs with one of a plurality of predetermined labels, the event classifier module trained with a plurality of other historical event logs including manually labeled event data associated with an occurrence of a plurality of events recorded by one or more of the plurality of image processing devices and other image processing devices and the event classifier module trained by extracting features from the plurality of other historical event logs predictive of failures and diagnostics associated with the one or more of the image processing device and other image processing devices, the extracted features including collocational features and bag-of-words features associated with event data included in the plurality of other historical event logs; and
a diagnostic module configured to generate a labeled version of each of the plurality of historical event logs including the historical record of each occurrence of the events associated with the respective image processing device and a respective label generated by the event classifier module for each occurrence of the events associated with the respective image processing device,
wherein the plurality of events are associated with a plurality of image processing device states associated with the image processing device, and the predetermined labels uniquely identify each of the plurality of image device states,
wherein the plurality of image processing device states are associated with a plurality of image processing device fault states, the fault states and the plurality of predetermined labels including two or more of error, no error, warning, okay state, critical, major, ordinary, and nonclassified, and
wherein the image processing device is one of a MFD (Multi-Function Device), a printer, a scanner, and a fax machine.