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 .
DETAILED CORRESPONDENCE
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) is acknowledged.
Status of Claims
Claims 1 has been amended.
Claims 3, 6, 7 have been cancelled.
No claims have been added.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 2, 4, 5, 8 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite:
receiving an inspection report comprising original text from a user;
recognizing, in the original text, a first text entity representing damage a second text entity representing and cause;
testing the first and second text entities for consistency with each other using a training dataset provided with inspection reports containing text entities comprising consistent damage and cause pairs annotated by an equipment inspection specialist;
in accordance with the first and second text entities being inconsistent with each other, notifying the user that the first and second text entities are inconsistent with each other and presenting the user with one or more suggestions for editing the original text to make the first and second text entities consistent with each other;
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms according to API 571 standard for the first and second text entities, a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism according to API 571 standard and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms according to API 571 standard for the first and second text entities in the original text, notifying the user that the original text lacks the estimated deterioration mechanisms and presenting the user with one or more suggestions for editing the original text to include at least one of the estimated deterioration mechanisms according to API 571 standard based on output;
in accordance with the original text comprising at least one of the estimated deterioration mechanisms, estimating, based at least in part on output and output, recommended actions to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism according to API 571 standard;
in the absence of the estimated recommended actions to be taken to ensure integrity of the equipment in the original text, notifying the user that the original text lacks the estimated recommended actions to be taken to ensure integrity of the equipment and presenting a user with one or more suggestions for editing the original text with at least one of the estimated recommended actions to be taken to ensure integrity of the equipment; and
in instances when at least one of the recommended actions to be taken to ensure integrity of the equipment is in the original text, concluding the method.
The invention is directed towards the abstract idea of reviewing reports to identify discrepancies based on the collection and comparison of information and, based on a rule, identify options; collecting, analyzing, and displaying information and certain results of the collection and analysis, which corresponds to both “Mental Processes” and “Certain Methods of Organizing Human Activities” as it is directed towards steps that can be performed in the human mind and/or through the aid of pen and paper, e.g., having a human receive and read over a document and identify any errors, mistakes, missing information, or the like and, if any exist, make the necessary corrections based on known information, such as, but not limited to, writing down any corrections that need to be made to a submitted document. The claimed invention is also directed towards “Certain Methods of Organizing Human Activities” as it is directed to the fundamental economic principles or practices (mitigating risk) by ensuring that inspection reports are proper; managing persona behavior or relationships or interactions between people (ensuring that inspectors are following rules or instructions with regards to writing an inspection report); and legal interactions (ensuring that inspectors are fulfilling legal obligations with respect to properly reporting the inspection of an asset).
The limitations of:
receiving an inspection report comprising original text from a user;
recognizing, in the original text, a first text entity representing damage a second text entity representing and cause;
testing the first and second text entities for consistency with each other using a training dataset provided with inspection reports containing text entities comprising consistent damage and cause pairs annotated by an equipment inspection specialist;
in accordance with the first and second text entities being inconsistent with each other, notifying the user that the first and second text entities are inconsistent with each other and presenting the user with one or more suggestions for editing the original text to make the first and second text entities consistent with each other;
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms according to API 571 standard for the first and second text entities, a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism according to API 571 standard and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms according to API 571 standard for the first and second text entities in the original text, notifying the user that the original text lacks the estimated deterioration mechanisms and presenting the user with one or more suggestions for editing the original text to include at least one of the estimated deterioration mechanisms according to API 571 standard based on output;
in accordance with the original text comprising at least one of the estimated deterioration mechanisms, estimating, based at least in part on output and output, recommended actions to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism according to API 571 standard;
in the absence of the estimated recommended actions to be taken to ensure integrity of the equipment in the original text, notifying the user that the original text lacks the estimated recommended actions to be taken to ensure integrity of the equipment and presenting a user with one or more suggestions for editing the original text with at least one of the estimated recommended action to be taken to ensure integrity of the equipment; and
in instances when at least one of the recommended actions to be taken to ensure integrity of the equipment is in the original text, concluding the method,
are processes that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of a generic machine learning model. That is, other than reciting a generic machine learning model nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the generic machine learning model in the context of this claim encompasses a human can review a document to determine if any edits, corrections, or the like need to be made based on known information. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of a generic machine learning model, then it falls within the “Mental Processes” and “Certain Methods of Organizing Human Activities” groupings of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application. In particular, the claim only recites additional elements – a generic machine learning model to, presumably, performing operations that a human can perform in their mind and/or using pen and paper, i.e. reviewing a document to compare it against known information and determine whether any errors, mistakes, missing information, or the like exists. The generic machine learning model in the steps are recited at a high-level of generality (see additional discussion below) while also reciting that, presumably, the generic machine learning model is merely being applied to perform the steps that can be performed in the human mind and/or using pen and paper; "[use] of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.” Therefore, according to the MPEP, this is not solely limited to computers but includes other technology that, recited in an equivalent to “apply it,” is a mere instruction to perform the abstract idea on that technology (See MPEP 2106.05(f)) such that it amounts no more than mere instructions to apply the exception using a generic processor executing computer code stored on a computer medium.
Although the claim recites train and retraining a machine learning algorithm, the claims and specification fail to provide sufficient disclosure regarding an improvement to how a machine learning algorithm can be trained, but simply recites a high-level generic recitation that a machine learning algorithm is being trained. There is insufficient evidence from the specification to indicate that the use of the machine learning algorithm involves anything other than the generic application of a known technique in its normal, routine, and ordinary capacity or that the claimed invention purports to improve the functioning of the computer itself or the machine learning algorithm. None of the limitations reflects an improvement in the functioning of a computer, or an improvement to other technology or technical field, applies or uses a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, implements a judicial exception with, or uses a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, effects a transformation or reduction of a particular article to a different state or thing, or applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception.
Even training and applying a machine learning model using the various techniques recited in claim 7 or saving information for later retraining (claim 8) is simply application of a computer model, itself an abstract idea manifestation. Further, such training and applying of a model is no more than putting data into a black box machine learning operation. The nomination as being a machine learning model is a functional label, devoid of technological implementation and application details. The specification does not contend it invented any of these activities, or the creation and use of such machine learning models. In short, each step does no more than require a generic computer to perform generic computer functions. As to the data operated upon, "even if a process of collecting and analyzing information is 'limited to particular content' or a particular 'source,' that limitation does not make the collection and analysis other than abstract." SAP America, Inc. v. InvestPic LLC, 898 F.3d 1161, 1168 (Fed. Cir. 2018).
The Examiner asserts that the scope of the disclosed invention, as presented in the originally filed specification, is not directed towards the improvement of machine learning, but directed towards simply reviewing a document to determine if any corrections need to be made based on known information. The specification’s disclosure on machine learning is nothing more than a high general explanation of generic technology and applying it to the abstract idea. The Examiner asserts that the claimed invention fails to recite any iterative process being performed on the machine learning algorithm/model in order to demonstrate that the machine learning algorithm/model is being improved upon, i.e. a demonstration that would support an improvement upon machine learning technology. Referring to MPEP § 2106.05(f), the training and re-training are merely being used to facilitate the tasks of the abstract idea, which provides nothing more than a results-oriented solution that lacks detail of the mechanism for accomplishing the result and is equivalent to the words “apply it,” per MPEP § 2106.05(f). The Examiner asserts that in light of the 2024 Guidance Update on Patent Subject Matter Eligibility, Including on Artificial Intelligence, the claimed invention is analogous to Example 47, Claim 2.
Further, the combination of these elements is nothing more than a generic computing system with machine learning model(s). Because the additional elements are merely instructions to apply the abstract idea to a computer, as described in MPEP § 2106.05(f), they do not integrate the abstract idea into a practical application.
Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of using a generic machine learning model to perform the steps of:
receiving an inspection report comprising original text from a user;
recognizing, in the original text, a first text entity representing damage a second text entity representing and cause;
testing the first and second text entities for consistency with each other using a training dataset provided with inspection reports containing text entities comprising consistent damage and cause pairs annotated by an equipment inspection specialist;
in accordance with the first and second text entities being inconsistent with each other, notifying the user that the first and second text entities are inconsistent with each other and presenting the user with one or more suggestions for editing the original text to make the first and second text entities consistent with each other;
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms according to API 571 standard for the first and second text entities, a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism according to API 571 standard and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms according to API 571 standard for the first and second text entities in the original text, notifying the user that the original text lacks the estimated deterioration mechanisms and presenting the user with one or more suggestions for editing the original text to include at least one of the estimated deterioration mechanisms according to API 571 standard based on output;
in accordance with the original text comprising at least one of the estimated deterioration mechanisms, estimating, based at least in part on output and output, recommended actions to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism according to API 571 standard;
in the absence of the estimated recommended actions to be taken to ensure integrity of the equipment in the original text, notifying the user that the original text lacks the estimated recommended actions to be taken to ensure integrity of the equipment and presenting a user with one or more suggestions for editing the original text with at least one of the estimated recommended action to be taken to ensure integrity of the equipment; and
in instances when at least one of the recommended actions to be taken to ensure integrity of the equipment is in the original text, concluding the method,
amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept.
Additionally:
Claims 2 is directed to human activities and extra-solution activities, i.e. providing some type of output to inform a human about some information, in this case, information is missing, wrong, or the like
Claims 4, 5 are directed towards generic technology recited at a high level of generality and applying it to the abstract idea, as weas discussed above with respect to machine learning.
Claim 8 has already been discussed above.
In summary, the dependent claims are simply directed towards providing additional descriptive factors that are considered for reviewing a document for missing, incorrect, inaccurate, correct, and etc. information. Accordingly, the claims are not patent eligible.
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 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 – 5, 8 are rejected under 35 U.S.C. 103 as being unpatentable over Knuffman et al. (US Patent 11,080,841 B1) in view of Ghosh et al. (US PGPub 2020/0097921 A1) in further view of Inspectioneering (New From the American Petroleum Institute (API) - API RP 571).
In regards to claim 1, Knuffman discloses a method for detecting inaccuracies and gaps in wording of inspection reports of equipment in industrial facilities, characterized by comprising the following steps:
receiving, via a user interface, an inspection report comprising original text from a user;
recognizing, in the original text, a first text entity representing damage a second text entity representing and cause;
testing the first and second text entities for consistency with each other using a first model trained with a training dataset provided with inspection reports containing text entities comprising consistent damage and cause pairs annotated by an equipment inspection specialist;
in accordance with the first and second text entities being inconsistent with each other, notifying the user that the first and second text entities are inconsistent with each other […];
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms […] for the first and second text entities, by a second model trained with a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism […] and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms […] for the first and second text entities in the original text, [suggesting] […] the estimated deterioration mechanisms […] based on output of the second model via the user interface;
in accordance with the original text comprising at least one of the estimated deterioration mechanisms, estimating, based at least in part on output of the first model and output of the second mode, recommended action to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism […];
in the absence of the estimated recommended actions to be taken to ensure integrity of the equipment in the original text, notifying the user […] [of] the estimated recommended actions to be taken to ensure integrity of the equipment and presenting a user with one or more suggestions for […] at least one of the estimated recommended action to be taken to ensure integrity of the equipment via the user interface; and
in instances when at least one of the recommended actions to be taken to ensure integrity of the equipment is in the original text, concluding the method.
Knuffman discloses a system and method for resolving inaccurate information being provided in a report by an individual inspecting an asset (Col. 1 Lines 54 – 60). Knuffman discloses that relying on only manual inspection processes have notable disadvantages and limitations, such as, but not limited to, difficulty of inspecting an asset, thereby resulting in an incomplete or inaccurate inspection of the asset, significant costs, and delays (Col. 4 Lines 16 – 56).
Consequently, Knuffman discloses a system and method of resolving these deficiencies by having a system that receives a report of damages or defects that correspond to the asset and verifying whether the reported damages/defects are correct (Col. 15 Lines 45 – 60). Knuffman discloses that the system collects additional information of the asset, e.g., images, to determine if its findings align with the reported findings (Col. 15 Lines 45 – 60). Knuffman discloses that the system receives a textual explanation of the damage/defect, e.g., report, insurance claim, or the like (Col. 16 Lines 47 – 65; Col. 21 Lines 9 – 40), and utilizes machine learning (ML) with optical character recognition (OCR), natural language processing (NLP), semantic analysis, and/or automatic reasoning, which has been trained to identify damages/defects (Col. 17 Lines 1 – 16, 34 – 46) so that the report/claim, which is received from an entity, can be analyzed to determine if the reported damages/defects align with the system’s findings (Col. 9 – 10 Lines 44 – 5).
As a non-limiting example, Knuffman discloses that an entity can provide a report that the front bumper is damaged because it has a dent, i.e. the first text is “damaged front bumper” and the second text is “dent”; the vehicle drivability/steering is damaged because it is not driving straight, i.e. the first text is “damaged steering/drivability” and the second text is “not driving straight”, “wheel alignment is improper”, or “damaged or defective undercarriage”; the undercarriage is damaged because it has been pierced, i.e. the first text is “damaged undercarriage” and the cause is “pierced undercarriage”; or the lights are damaged because they are not turning on because the headlights are broken, i.e. the first text is “the lights are not on” or “the headlights are damaged” and the second text is “broken or defective headlight” (Col. 3 Lines 45 – 65; Col. 4 Lines 6 – 15; Col. 7 Lines 21 – 32; Col. 15 Lines 13 – 28; Col. 26 Lines 18 – 26), which the system receives and analyzes to determine if what is reported coincides with what the system has determined based on the information that it has been provided with, i.e. whether the provided image substantiates the entity’s claim (Col. 9 – 10 Lines 44 – 5).
As a result, Knuffman discloses (limitations 1, 2, 3, 4, 5) a system and method for recognizing the text in a report (Col. 9 Lines 21 – 23; Col. 13 Lines 49 – 52; Col. 16 Lines 47 – 65; Col. 9 Lines 51 – 53) and utilizes machine learning to determine (test) if the reported damages, repair, servicing, and/or cause align or consistent with its own findings (Col. 15 Lines 46 – 61; Col. 17 – 18 Lines 1 – 4) (e.g., the light does not work because the headlight/taillight is broken or defective (Col. 3 Lines 45 – 65; Col. 26 Lines 18 – 26).
Knuffman discloses (limitations 3, 4, 5) that the analysis involves comparing the reported information with known information to verify if a reported damage, cause, and deterioration mechanism are supported by known information (Col. 8 Lines 6 – 25) and the system will transmit its assessment to an entity, wherein the assessment includes the indication of the identified damage/defect (Col. 9 Lines 11 – 21; Col. 10 Lines 1 – 5; Col. 22 Lines 21 – 26). The system will analyze the claim and provided information to identify damage (e.g., poor drivability/steering), cause (e.g., damaged or defective undercarriage part), and deterioration mechanism (e.g., improper wheel alignment) to verify whether the entity submitted claim is correct or not and, if not, provide the results of its analysis. Further still, Knuffman discloses that the machine learning model utilizes not only image analysis, but optical character recognition (OCR), natural language processing (NLP), semantic analysis, and/or automatic reasoning (Col. 17 Lines 34 – 46). In other words, in order for the system to perform an image analysis to determine if the textually reported damage and cause is consistent with the image analysis results performed by the system, the system utilizes machine learning that has been trained using, for example, image analysis and OCR and NLP to extract (i.e. recognize) the text within the textual report provided by user(s) who have inspected and reported damage/cause and comparing it against the image of the vehicle and images that it has been trained on to determine whether the reported damage, e.g., damage caused be vandalism, cracked windshield due to inclement weather (e.g., Col. 3 Lines 45 – 57), matches with the description associated with the image the system has taken and images it has been trained on. That is to say, in order for the models to determine that an image of a cracked windshield is an image of a cracked windshield, the model would have been required to be trained on a plurality of images of cracked windshields with textual information that the model has been provided to indicate that these are images of cracked windshields and not images of flat tires. Knuffman discloses that the model can be trained using, at least, supervised machine learning (Col. 17 Lines 17 – 33), thereby resulting in an entity, user or other machine learning model, instructing the system that “yes, this is a correct assessment of this image representing damage due to vandalism” or “no, this is an incorrect assessment of this image representing damage due to vandalism”. The Examiner asserts that the trained machine learning model categorizes images along with textual descriptors in order to properly associate the reported damage and cause with the correct image that corresponds to a corresponding damage and cause, otherwise it would not know what image or assessment to provide when a user reports damage due to vandalism and the image the system took and is analyzing to arrive at a conclusion.
However, although the Examiner asserts that Knuffman teaches the entirety of limitations 3, 5, in the interest of compact prosecution, the Examiner has provided Ghost to more explicitly teach this aspect of the invention, which will be discussed below.
With regards to (limitations 6, 8, 9), in the absence of certain information due to the information not matching or if the reported information is correct, e.g., actual damage, cause, or the like, the system will include and provide its findings (which would include any missing information) for use by an entity, e.g., owner or insurance representative. As a non-limiting example, Knuffman discloses that a disadvantage/limitation to manual inspections is failing to identify an issue due to there being insufficient space for a mechanic. Accordingly, in addition to what has been discussed above, Knuffman resolves this by utilizing an inspection device that is capable of inspecting difficult to inspect areas. In other words, the system is able to determine when an asset’s damage/defect is missing (Col. 4 Lines 6 – 56), thereby allowing the system to identify that if certain information is missing, the system will notify the entity of the missing information, provide the correct information, and provide a recommendation to resolve the claim and, if the information is correct, the system will proceed with providing a recommendation to resolve the claim.
Furthermore, Knuffman discloses (limitation 6, 7, 8, 9) that the system, based on its findings, will provide a recommended action for the damage (e.g., poor drivability/steering), cause (e.g., damaged or defective undercarriage part), and deterioration mechanism (e.g., improper wheel alignment) so that it can be resolved, e.g., schedule maintenance, repair, and/or replacement, identify a replacement part, initiate an order for the replacement part, and provide an estimate of the monetary cost of the maintenance, service, and/or repair for correcting the identified damage/defect (Col. 21 Lines 56 – 66).
In summary, Knuffman discloses the evaluation of an inspection report performed by machine learning (ML) that employs cognitive computing and/or predictive model techniques, including machine learning techniques or algorithms, which has been trained on text and images to identify damages/defects so that the system can identify/schedule maintenance, repairs, part replacement and/or services for correcting damages/defects; identify replacement parts; generate or modify insurance policies and terms; estimate monetary costs; and transmit indications of vehicle damages/defects (Col. 17 Lines 1 – 16). Knuffman discloses that the ML can be trained using a wide range of techniques and, more specifically, can be retrained so that more accurate and correct assessments can be performed (Col. 17 – 18 Lines 34 – 4). Further still, with respect to retraining, Knuffman discloses that the ML can be improved upon by saving useful information by way of supervised, unsupervised, deep learning, and other techniques that involve an iterative process of improving the ML’s processes (Col. 17 Lines 34 – 60). Moreover, Knuffman discloses that the ML can be provided with newly acquired information in order to improve its analysis, e.g., images captured by an inspection device (Col. 18 Lines 13 – 40).
Despite this, and as was discussed above, Knuffman does not explicitly disclose that an ML can be trained using text in order to review a written report (or the like) for the purposes of assessing damage, accuracy, content, recommendations, or etc. and providing recommendations to correct the report.
To be more specific, Knuffman fails to explicitly disclose:
testing the first and second text entities for consistency with a first model trained with a training dataset provided with inspection reports containing text entities comprising consistent damage and cause pairs annotated by an equipment inspection specialist;
in accordance with the first and second text entities being inconsistent with each other, notifying the user that the first and second text entities are inconsistent with each other and presenting the user with one or more suggestions for editing the original text to make the first and second text entities consistent with each other via the user interface;
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms according to American Institute (API) 571 standard for the first and second text entities, by a second model trained with a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism according to API 571 standard and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms according to API 571 standard for the first and second text entities in the original text, notifying the user that the original text lacks the estimated deterioration mechanisms and presenting the user with one or more suggestions for editing the original text to include at least one of the estimated deterioration mechanisms according to API 571 standard based on output of the second model via the user interface
in the absence of the estimated recommended actions to be taken to ensure integrity of the equipment in the original text, notifying the user that the original text lacks the estimated recommended actions to be taken to ensure integrity of the equipment and presenting a user with one or more suggestions for editing the original text with at least one of the estimated recommended actions to be taken to ensure integrity of the equipment via the user interface
(emphasis added)
However, Ghosh, which is also directed towards utilizing machine learning models to analyze reports containing textual descriptions of equipment issues, further teaches that it is well-known and obvious to train the machine learning models on textual content. Similar to Knuffman, Ghosh teaches that historical information is used to train the models so that the system can compare the information provided in a report against historical information so that the system can determine a recommendation concerning the reported issue. However, Ghosh explicitly teaches that it is not only well-known in the art to receive a report provided in a textual format and to utilize OCR and NLP to extract and process the text provided in a textual report, but for the models to be trained on textual information by way of training it on past reports, thereby allowing the system to perform a comparison and matching of reported issue against historical reported issues and their corresponding recommendations to then, in turn, determine and provide a recommendation for the reported issue. Ghosh teaches that the models can be trained to identify keywords within reports so that it can remove text that is not useful to the problem description, correct typographical errors, and etc. and that it is trained on historical textual reports so that it can assess a provided report and, ultimately provide an appropriate recommendation, e.g., typographic corrections, repair/replacement solutions, identifying similarities so that the best solution is provided, and etc. Finally, Ghosh teaches that this process results in saving time to diagnose and resolve an issue and assist technicians who have a knowledge gap due to being new and having sufficient experience.
(For support see: ¶ 1, 25, 35, 44, 61, 67, 69, 70, 71, 96)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate into the machine learning diagnosis and resolution system for reported issues of Knuffman with the ability to not only train machine learning models on images, but text, as well, as taught by Ghosh, as this results in saving time to diagnose and resolve an issue and assist technicians who have a knowledge gap due to being new and having sufficient experience, e.g., correcting typographical errors provided by an inexperienced technician based on training data of historical reports.
The combination of Knuffman and Ghosh discloses a system and method that utilizes machine learning to assist a user with drafting and submitting inspection reports by referring to writing rules, requirements pertaining to information that should be included in a report, and training data that allows the machine learning model to review a report, identify issues found within the report, and providing recommendations to resolve identified issues. Although the combination of Knuffman and Ghosh teaches various rule types that the machine learning model has been trained on and that machine learning is a useful technology that can be applied to different industries to assist a user with drafting an inspection report, the combination of Knuffman and Ghosh fails to explicitly disclose all types of rules or standards.
To be more specific, the combination of Knuffman and Ghosh fails to disclose:
in accordance with the first and second text entities being consistent with each other, estimating one or more deterioration mechanisms according to American Institute (API) 571 standard for the first and second text entities, by a second model trained with a second dataset comprising data representing inspection reports containing consistent damage, cause, and deterioration mechanism according to API 571 standard and pre-defined labels of recommended actions annotated by inspection experts;
in the absence of the estimated deterioration mechanisms according to API 571 standard for the first and second text entities in the original text, notifying the user that the original text lacks the estimated deterioration mechanisms and presenting the user with one or more suggestions for editing the original text to include at least one of the estimated deterioration mechanisms according to API 571 standard based on output of the second model via the user interface;
in accordance with the original text comprising at least one of the estimated deterioration mechanisms, estimating, based at least in part on output of the first model and output of the second mode, recommended action to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism according to API 571 standard.
However, Inspectioneering teaches that API 571 is a well-known and established standard that has existed long before the applicant’s effective filing date. API 571 lays out the required standards that should be followed in the petroleum industry. The sole difference between the claimed invention and the combination of Knuffman and Ghosh is that the combination of Knuffman and Ghosh fails to disclose all types of industries that require inspections and all types of rules, standards, regulations, or the like that correspond to the particular inspection that is being performed in the particular industry, in this case, applying API 571 when performing inspections in the petroleum field. However, Inspectioneering teaches that it would have been obvious to one of ordinary skill in the art of the petroleum industry and performing inspections in the petroleum industry to refer to and apply API 571 as this allows for the proper management of refining equipment integrity and is an excellent reference for inspection, operations, and maintenance personnel.
One of ordinary skill in the art looking upon the combination of Knuffman and Ghosh would have found that training and applying machine learning to review text-based inspection reports to identify issues within the contents of the report and provide recommendations to resolve the identified issues is not limited to any particular field and that the examples provided by the combination of Knuffman and Ghosh are simply examples of how certain fields of endeavor can benefit from use of this technology. As a result, one of ordinary skill in the art of performing inspections in the petroleum industry would have found it motivated to look upon the teachings of Inspectioneering to determine that API 571 is well-known standard that is used in the petroleum industry when performing inspections and, consequently, found it obvious to substitute or include API 571 for the rules of the combination of Knuffman and Ghosh as it would have been obvious that the rules to be utilized by the machine learning model when reviewing an inspection report would correspond to what the contents of the report are directed to, e.g., vehicle based rules for vehicle inspections and API 571 (or the like) for inspections performed in the petroleum industry. One of ordinary skill in the art would have found it obvious that the same predictable result would be achieved regardless of the rules, standards, regulations, or the like that the machine learning model is referring to, i.e. the machine learning model would still review a document’s contents, identify issues, and recommend solutions to the identified issues in accordance to the rules that correspond to the environment that the technology is being utilized in.
(For support see Page 1)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention that since each individual element and its function are shown in the prior art, albeit shown in separate references, the difference between the claimed subject matter and the prior art rests not on any individual element or function but in the very combination itself-that is in the substitution of API 571 as the rules, standards, regulations, or the like when performing an inspection in the petroleum field, as taught by Inspectioneering, for the other rules, standards, regulations, or the like, as disclosed by the combination of Knuffman and Ghosh. That is to say, if the inspection is being performed in the petroleum industry, then it would have been obvious to refer to the well-known standards set forth by API 571, if the inspection is being performed in the vehicle field, then standards pertaining to vehicles would be used, if the review is based on grammar, then grammar rules would be used, and so forth. In the end, one of ordinary skill in the art would have found it obvious to use the appropriate standards that correspond to the environment of use that the technology will be used is as the technology itself and functions remain the same, i.e. view a document’s contents, identify issues, and recommend solutions to the identified issues in accordance to the rules that correspond to the environment that the technology is being utilized in.
Thus, the simply substitution of one known element for another producing a predictable result renders the claim obvious.
In regards to claim 2, the combination of Knuffman, Ghosh, and Inspectioneering discloses he method according to claim 1, further comprising informing a user about the lack of reporting the cause, damage, or both of the first and second text, when at least one or both entities representing damage and cause, respectively, are absent (Knuffman discloses a system and method for recognizing the text in a report and utilizes machine learning to determine (test) if the reported damages, repair, servicing, and/or cause align or consistent with its own findings (e.g., the light does not work because the headlight/taillight is broken or defective (Col. 3 Lines 45 – 65; Col. 26 Lines 18 – 26)). Knuffman discloses that the analysis involves comparing the reported information with known information to verify if a reported cause and damage are supported by known information (Col. 8 Lines 6 – 25) and the system will transmit its assessment to an entity, wherein the assessment includes the indication of the identified damage/defect (Col. 22 Lines 21 – 26). In other words, in the absence of certain information due to the information not matching or if the reported information is correct, e.g., actual damage, cause, or the like, the system will include and provide its findings (which would include any missing information) for use by an entity, e.g., owner or insurance representative. As a non-limiting example, Knuffman discloses that a disadvantage/limitation to manual inspections is failing to identify an issue due to there being insufficient space for a mechanic. Accordingly, Knuffman resolves this by utilizing an inspection device that is capable of inspecting difficult to inspect areas. In other words, the system is able to determine when an asset’s damage/defect is missing (Col. 4 Lines 6 – 56)
In summary, Knuffman discloses that a user is notified when information regarding damage/defect and cause are missing or if there is an inconsistency between the report the system receives and the findings of the system (Col. 13 – 14 Lines 61 – 19).).
4. In regards to claim 4, the combination of Knuffman, Ghosh, and Inspectioneering discloses the method according to claim 1, further comprising when the is consistent, estimating one or more deterioration mechanisms according to API 571 standard for the first and second text entities through an intent detector algorithm (Knuffman discloses a system and method for recognizing the text in a report and utilizes machine learning to determine (test) if the reported damages, repair, servicing, and/or cause align or consistent with its own findings (e.g., the light does not work because the headlight/taillight is broken or defective (Col. 3 Lines 45 – 65; Col. 26 Lines 18 – 26)). Knuffman discloses that the analysis involves comparing the reported information with known information to verify if a reported cause and damage are supported by known information (Col. 8 Lines 6 – 25) and the system will transmit its assessment to an entity, wherein the assessment includes the indication of the identified damage/defect (Col. 22 Lines 21 – 26). In other words, in the absence of certain information due to the information not matching or if the reported information is correct, e.g., actual damage, cause, or the like, the system will include and provide its findings (which would include any missing information) for use by an entity, e.g., owner or insurance representative. As a non-limiting example, Knuffman discloses that a disadvantage/limitation to manual inspections is failing to identify an issue due to there being insufficient space for a mechanic. Accordingly, Knuffman resolves this by utilizing an inspection device that is capable of inspecting difficult to inspect areas. In other words, the system is able to determine when an asset’s damage/defect is missing (Col. 4 Lines 6 – 56).
In regards to claim 5, the combination of Knuffman, Ghosh, and Inspectioneering discloses the method according to claim 1, further comprising when at least one of the estimated deterioration mechanisms according to API 571 standard is in the original text, estimating recommended action to be taken to ensure integrity of the equipment for the damage, cause, and deterioration mechanism according to API 571 standard through an intent detector algorithm (Knuffman discloses a system and method for recognizing the text in a report and utilizes machine learning to determine (test) if the reported damages, repair, servicing, and/or cause align or consistent with its own findings (e.g., the light does not work because the headlight/taillight is broken or defective (Col. 3 Lines 45 – 65; Col. 26 Lines 18 – 26)). Knuffman discloses that the analysis involves comparing the reported information with known information to verify if a reported cause and damage are supported by known information (Col. 8 Lines 6 – 25) and the system will transmit its assessment to an entity, wherein the assessment includes the indication of the identified damage/defect (Col. 22 Lines 21 – 26). In other words, in the absence of certain information due to the information not matching or if the reported information is correct, e.g., actual damage, cause, or the like, the system will include and provide its findings (which would include any missing information) for use by an entity, e.g., owner or insurance representative. As a non-limiting example, Knuffman discloses that a disadvantage/limitation to manual inspections is failing to identify an issue due to there being insufficient space for a mechanic. Accordingly, Knuffman resolves this by utilizing an inspection device that is capable of inspecting difficult to inspect areas. In other words, the system is able to determine when an asset’s damage/defect is missing (Col. 4 Lines 6 – 56)
In summary, Knuffman discloses that a user is notified when information regarding damage/defect and cause are missing or if there is an inconsistency between the report the system receives and the findings of the system (Col. 13 – 14 Lines 61 – 19). Further still, the regardless of the outcome of its assessment, i.e. matching, missing, and/or misaligned information, the system estimates a deterioration mechanism for the damage and cause using machine learning that has been trained on, at least, images and text, e.g., headlight/taillight not working as intended because it is broken or defective, poor/worsening driving/handling of a vehicle because of poor wheel alignment or an axle having a break. Given the results of the analysis, the system will provide a solution to resolve the issue, e.g., part replacement.).
In regards to claim 8, the combination of Knuffman, Ghosh, and Inspectioneering discloses the method according to claim 1, further comprising:
in instances when the suggestions for the deterioration mechanism according to API 571 standard and recommended actions to be taken to ensure integrity of the equipment p