Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3-11 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over SHUMPERT (US 20160342903 A1; cited previously) in view of BHATIA et al. (US 20210026722 A1; cited previously; hereinafter “BHATIA”) and Neumann (US 20210104300 A1).
Regarding claim 1, SHUMPERT teaches a diagnosis system for diagnosing presence or absence of an abnormality from data pieces collected in a factory (i.e., “systems and/or methods for dynamic anomaly detection in machine sensor data”; see [0001]; “in all areas of manufacturing and controlling”; see [0023]), the diagnosis system comprising:
diagnosing circuitry (i.e., “at least one processor and a memory coupled thereto”; see [0069]) to diagnose presence or absence of the abnormality by classifying, in accordance with a model defining a plurality of groups, the collected data pieces into at least one of the plurality of groups (i.e., “the shared learning and prediction component 506 predicts the class (normal or anomalous) of data instances as they arrive. It may do this by comparing the current instances with the model of normal behavior and identifying any significant differences”; see [0055]; “using a modified k-means cluster algorithm … Each said cluster has an associated class, with the class being one of an anomalous type class and a non-anomalous type class”; see [0035]);
extracting circuitry (i.e., “at least one processor and a memory coupled thereto”; see [0069]) to extract, from the collected data pieces, a candidate for a data piece to belong to a new group different from the plurality of groups (i.e., “if a determination is made that it is a new anomaly, then information about the anomaly is sent for review”; see [0060]) when the data-piece has a lower belonging level to any of the plurality of groups than others of the collected data pieces (i.e., “where the current instance is predicted to be the class of its nearest cluster (in terms of a multivariate distance measure to the centroid of the cluster), whose positions are determined during training. The current instance is predicted to be a new potential anomaly if it is nowhere near any of the existing clusters in certain example embodiments”; see [0080]),
generating circuitry (i.e., “at least one processor and a memory coupled thereto”; see [0069]) to generate candidate information indicating the new group from the candidate extracted by the extracting circuitry (i.e., “the shared learning and prediction component 506 outputs flagged instance records, along with the immediately preceding instances for context and review”; see [0133]; “The expert confirmed that this was indeed an anomaly, and so a new cluster was created”; see [0131]);
a receiver to groups (i.e., “An expert classification about the anomaly is received”; see [0061]); and
learning circuitry to learn a new model, (i.e., “If the anomaly is confirmed as being new (in step S420), then the knowledgebase 514 is updated (step S422), the shared model is updated (step S408) … The expert might confirm the instance as new type of anomaly or they might classify it as a new normal operating state”; see [0061]; “The expert confirmed that this was indeed an anomaly, and so a new cluster was created”; see [0131]), wherein
the diagnosing circuitry diagnoses and outputs in real time presence or absence of the abnormality with the new model after the learning circuitry learns the new model (i.e., “The class of the data that is received is predicted (step S404) using the shared learning and prediction component 506”; see [0055]; “The initial model is then updated using supervised learning techniques so that the model can learn to discriminate between the various classes of operation”; see [0064]; “Thus, in real-time as the data stream is received, the shared model's clusters are elaborated with a classification from various states of normal or abnormal behavior. Prediction is then straightforward”; see [0116]) and
SHUMPERT does not explicitly disclose (see only the underlined):
a new-group candidate storage to accumulate the candidate extracted by the extracting circuitry; and
generating circuitry to generate candidate information indicating the new group from the candidate accumulated in the new-group candidate storage when receiving, from a user, a generation instruction to generate the new group.
However, SHUMPERT further teaches outputting a plurality of flagged records for review (see [0133]). Since an expert may not be always available to review a flagged record when the record is flagged, it is obvious to accumulate the flagged records in a storage for review when the expert is available. Also, it is well-known to use a computer storage component to accumulate records.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify SHUMPERT to provide a new-group candidate storage to accumulate the candidate extracted by the extracting circuitry; and such that the generating circuitry is configured to generate candidate information indicating the new group from the candidate accumulated in the new-group candidate storage when receiving, from a user, a generation instruction to generate the new group, as claimed. The rationale would be to facilitate outputting the plurality of flagged records for, so that the expert can review the flagged records at his convenience.
SHUMPERT does not explicitly disclose (see only the underlined):
a receiver to display a screen indicating the candidate information together with the collected data pieces belonging to the plurality of groups in a two-dimensional graph illustrating the plurality of groups.
But SHUMPERT teaches:
a graphical representation indicating the candidate information together with the collected data pieces belonging to the plurality of groups in a two-dimensional graph illustrating the plurality of groups (i.e., see FIGs. 12-14 and [0130]-[0131] showing the accumulated candidate instances, and the existing instances with their clusters in 2-dimension).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify SHUMPERT to configure the receiver to display a screen indicating the candidate information together with the collected data pieces belonging to the plurality of groups in a two-dimensional graph illustrating the plurality of groups, as claimed. The rationale would be to provide a visual feedback regarding the exiting clusters and the candidate instances to assist the user in making a decision for creating a new cluster or classifying the instances.
SHUMPERT does not explicitly disclose:
wherein the abnormality includes at least one of manufacturing defects, breakage of mechanical components, software execution errors, or communication errors detected in real time in the factory.
But SHUMPERT further teaches:
wherein the abnormality includes a component failure (see [0014], [0025]) or a machine failure type (see [0062]); and
discovering new problems (see [0031]).
It is well-known that a component failure or a machine failure type can include a breakage of mechanical components.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify SHUMPERT by building knowledgebase (see [0062]) for identifying a component failure or a machine failure type, such that the abnormality includes at least one of manufacturing defects, breakage of mechanical components, software execution errors, or communication errors detected in real time in the factory, as claimed. The rationale would be to facilitate the identification of the type of abnormality.
SHUMPERT does not explicitly disclose (see only the underlined):
the belonging level being calculated based on a Gaussian distribution;
learning circuitry to learn a new model, based on a Gaussian mixture model (GMM), including the new group when the addition information received by the receiver indicates that the new group is to be added to the plurality of groups.
But BHATIA teaches:
using a Gaussian mixture model (GMM) for data clustering (i.e., “Another type of clustering algorithm is Gaussian mixture models (GMM), which uses statistical probability to determine which data belongs in a cluster of normal data, such that any data points outside of that cluster is deemed to be anomalous”; see [0034]).
And Neumann teaches:
the belonging level being calculated based on a Gaussian distribution (i.e., “computing the probability that each datapoint belongs to a particular cluster. . EM algorithm may assume that the closer a datapoint is to the Gaussian's center, the more likely it belongs to that cluster”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify SHUMPERT in view of BHATIA and Neumann to incorporate a Gaussian mixture model (GMM) for data clustering, such that the belonging level is calculated based on a Gaussian distribution and the learning circuitry is to learn a new model, based on a Gaussian mixture model (GMM), including the new group when the addition information received by the receiver indicates that the new group is to be added to the plurality of groups, as claimed. The rationale would be to utilize a known clustering technique for its known ability of data clustering.
Regarding claim 3, SHUMPERT further teaches: wherein
the generating circuitry generates a plurality of subgroups into which data pieces that are excluded from the extraction performed by the extracting circuitry and belong to one group of the plurality of groups are to be classified (i.e., “Using the unsupervised prediction techniques described below, some instances eventually will be flagged as potentially anomalous and sent for review. Once confirmed or denied, that information is fed back, and this step is repeated with the newly classified (and thus labeled) data”; see [0064]; note that a plurality of subgroups will be generated as data are continuously collected), and
the learning circuitry learns the new model including the plurality of subgroups (i.e., “The initial model is then updated using supervised learning techniques so that the model can learn to discriminate between the various classes of operation”; see [0064]).
Regarding claim 4, SHUMPERT further teaches: wherein
the receiver receives an instruction to change a group to which data pieces excluded from the extraction performed by the extracting circuitry belong (i.e., “Otherwise, the shared model is updated (step S408)”; see [0061]; note that the expert confirmation of not creating a new anomaly cluster triggers the update of the model for the exiting clusters), and
the learning circuitry learns the new model from the data pieces belonging to the group changed in accordance with the instruction (i.e., “Otherwise, the shared model is updated (step S408)”; see [0061]; note that the expert confirmation of not creating a new anomaly triggers the update of the model for the exiting clusters).
Regarding claim 5, the prior art applied to the preceding linking claim(s) teaches the features of the linking claim(s).
SHUMPERT does not explicitly disclose:
a plurality of diagnosis devices to diagnose presence or absence of the abnormality with the model; and
a learning device to learn the model,
wherein
each of the plurality of diagnosis devices includes
collecting circuitry to collect the data pieces in the factory, and
the diagnosing circuitry,
the learning device includes
the extracting circuitry to extract the candidate from the data pieces collected by the plurality of diagnosis devices, the receiver,
the learning circuitry, and
transmitting circuitry to transmit the new model learned by the learning circuitry to the plurality of diagnosis devices, and
the diagnosing circuitry in each of the plurality of diagnosis devices diagnoses presence or absence of the abnormality with the new model transmitted by the transmitting circuitry.
The additional features in the claim are the separation of the diagnosis and learning functionalities into a learning device and a plurality of diagnosis devices and the learning device transmitting the new model to the plurality of diagnosis devices for the plurality of diagnosis devices to diagnose abnormality respectively.
However, SHUMPERT teaches the diagnosis and the learning functionalities as discussed in claim 1 above. Also, it is well-known to provide a computer service from a server to a plurality clients using a network. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the features as claimed. The rationale would be to take advantages of the known client-server configuration for better sharing of computer resources and centralized management (e.g., training and updating the model at the server).
Regarding claim 6, the claim recites the same substantive limitations as claim 1 and is rejected by applying the same teachings.
Regarding claim 7, the claim recites the same substantive limitations as claim 1 and is rejected by applying the same teachings.
Regarding claim 8, the claim recites the same substantive further limitations as claim 4 and is rejected by applying the same teachings.
Regarding claim 9, the claim recites the same substantive further limitations as claim 5 and is rejected by applying the same teachings
Regarding claim 10, the claim recites the same substantive further limitations as claim 5 and is rejected by applying the same teachings.
Regarding claim 11, the claim recites the same substantive further limitations as claim 5 and is rejected by applying the same teachings.
Regarding claim 13, SHUMPERT further teaches:
wherein the diagnosing circuitry outputs a diagnosis result (i.e., “the operator is alerted accordingly (step S414)”; see [0059])
SHUMPERT does not explicitly disclose (see only the underlined):
wherein the diagnosing circuitry outputs a diagnosis result as at least one of a display on a display device, a light on a light emitting diode, and a sound via a buzzer.
However, it is well-known to present a result through a display of the result or outputting an indication of the result usings lights or sounds. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to configure the system such that the diagnosing circuitry outputs a diagnosis result as at least one of a display on a display device, a light on a light emitting diode, and a sound via a buzzer, as claimed. The rationale would be to help communicating the result to the operator.
Response to Arguments
The objections to the claims have been withdrawn in view of the amendment.
The rejections under 35 USC 101 have been withdrawn in view of the amendment. The Examiner notes that the additional elements are sufficient to make the claim a practical application by providing a specific GUI for interacting with a user and presenting candidate information, collected data pieces, and their clusters in a two-dimension graph.
Regarding 35 USC 103, Applicant argued: Claim 1 is amended to recite, in part, "extracting circuitry to extract, from the collected data pieces, a candidate for a data piece to belong to a new group different from the plurality of groups when the data piece has a lower belonging level to any of the plurality of groups than others of the collected data pieces, the belonging level being calculated based on a Gaussian distribution." … Shumpert merely describes identifying an new anomaly and then providing the new anomaly to a human operator (Shumpert, [0060]).There is no mention in Shumpert of how the new anomaly is determined nor of whether a new group is created by Shumpert's system to contain the new anomaly… Bhatia does not cure the deficiencies in Shumpert. Therefore, no combination of Shumpert and Bhatia discloses or suggests every feature recited in amended Claim1.
The Examiner respectfully submits that Applicant’s arguments have been considered but are moot because a new ground has been found further in view of Neumann, as discussed in the rejections above.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN C KUAN whose telephone number is (571)270-7066. The examiner can normally be reached M-F: 9:00AM-5:30PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Schechter can be reached on (571) 272-2302. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JOHN C KUAN/Primary Examiner, Art Unit 2857