DETAILED ACTION
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 .
Status of Claims
Claims 1, 16 and 34 have been amended and are hereby entered.
Claims 4 – 6, 15 and 17 – 23 were cancelled.
Claims 1 - 3, 7 - 14, 16 and 24 - 34 are pending and have been examined.
This action is made a FINAL.
Response to Arguments
Applicant's arguments filed on September 26, 2025 have been fully considered but they are not persuasive.
Regarding to applicant's arguments against the 101 rejection of the pending claims on pages 10 – 14: Applicant’s arguments directed to 101 analysis were considered. However, these arguments are not persuasive and the examiner respectfully disagrees for the following reasons:
For Step 2A-Prong 1 starting in p. 10: The Applicant argues that the pending claims are not directed to any of the abstract ideas identified because “claims recite a specific technological system that improves how computers process fragmented and incomplete fraud investigation data” and the claimed features “represent a particular machine-based architecture and a concrete improvement in fraud detection technology, not a judicial exception”. However, the Examiner find these arguments unpersuasive and respectfully disagrees. Because in response to Applicant’s arguments in pp. 10 – 11 from Remarks, for each of the identified abstract ideas:
At least claims 1 and 34 are reciting mathematical concepts because the steps related with “identifying discoverable gaps” using “progressive and regressive algorithms” to sort classified data and to “show data missing or expected” (from claim 1 and 34 refer to ¶0036 from Applicant disclosure) and the steps of “adapt the behavioral model and classifier using adaptive pattern recognition algorithms…” and “…and the classifier learns from the identification of abnormal data points using adaptive pattern recognition algorithms” (from claims 16 and 34) are requiring mathematical calculations.
All claims are directed to methods of organizing human activity under “fundamental economic principles or practices”, “managing personal behavior or interactions between people”, and “engaging in commercial or legal interactions”. Because the claims and their steps recite the use and tracking of user’s social activities and behaviors via user inputs to predict and mitigate the risk of having fraudulent transactions and understanding the transaction behavior to comply with legal obligations such as federal and state regulations (see pages 16 – 17 in the Applicant’s disclosure for the claimed invention final outputs wherein state rules are applied for the “analytic roadmap” generation as claimed in at least claims 1 and 34).
All claims are directed to mental processes. Because the steps are reciting functions for classifying and sorting data inputs by categories, verify or modify the classifications by user to generate an analytic roadmap to investigate additional data and identify discoverable gaps by recognizing data patterns which requires observation, evaluation, judgment, and opinion.
As for the “specific technological system” being improved as recited in the claims, allegedly, are not reflecting the improvement of particularly reciting “how computers process fragmented and incomplete fraud investigation data” to achieve the intended result of generating an “analytic roadmap” to direct “investigators toward additional categories of data”. Rather, as the Applicant asserts, different features “a snowflake schema data warehouse with fact and dimension tables, progressive and regressive algorithms for iterative sufficiency testing”, among other features such as the general ML technology (e.g. classifier) recited, are used (e.g. applied) by the computer in addition to the high level of generality that the claim limitations are recited which is improving the identified abstract idea itself. Thus, the claim steps further describe the end result without providing details on how this alleged “improvement” to the computer functioning and/or to the existing technology of “fraud detection” systems is achieved.
For Step 2A-Prong 2 and Step 2B starting in p. 12: The Applicant alleges that the claims integrate, the judicial exception identified, into a practical application and further alleges that the several feature elements integrate the “abstract concept into a concrete technological solution” as they “are particular implementations that improve the functioning of computer-based fraud detection systems” for integrating the abstract idea to a “specific and technological implementation that solves the computer-based problem of analyzing fragmented and incomplete investigative data”. However, the Examiner finds these arguments unpersuasive and respectfully disagrees. Because such particular implementations are not reflecting the alleged “improvement” of the “functioning of computer-based fraud detection systems”. Rather, these elements do not amount to “more than generally linking the use of a judicial exception to a particular technological environment or field of use” such as technology related to “fraud detection systems”, as alleged by the Applicant which in turn failed to “reflect the disclosed improvement for the functioning of a computer or any other technology/technical field” (see MPEP 2106.04 (d)(1) and MPEP 2106.05(h)).
Similarly, these claim limitations and element features are invoking “computers merely as a tool” to perform the business process in which is merely adding the words “apply it” to the judicial exception (MPEP 2106.05(f)). These additional elements are not adding significantly more to the abstract idea. Because they were simply applying the abstract idea on computer components and broad ML technology that do not include or recites details of how to carry out the abstract idea. Instead, the claims are directed to “an improvement in the abstract idea itself” which is not an “improvement to technology” (see MPEP 2106.05(a)(II)).
For Step 2B starting in p. 12: The Applicant alleges that the claims and its additional elements were concluded to be “well-understood, routine, and conventional computer functions”. However, the Examiner finds this argument unpersuasive as well as it is moot. Because the Examiner did not consider this analysis for the claims in the previous OA and thus, the Examiner did not need to provide a factual determination was provided to support a conclusion that an additional element (or combination of additional elements) is well-understood, routine, conventional activity (see Berkheimer memo and MPEP 2106.05(d)(I)(2)). Rather, the Examiner considered the claims to be invoking the use of a generic computer (e.g. “apply it”) or reciting mere instructions to implement an abstract idea on a computer used as a tool (see MPEP 2106.05 (f)), as already explained above.
Regarding to applicant's arguments under 35 USC § 103 rejection for the pending claims on pages 14 – 21: Applicant’s arguments with respect to limitations in claim(s) 1, 16 and 34 have been considered but are not persuasive for the following reasons:
Arguments against the prior art references in pp.14 – 18: Applicant argues that “references can not be combined as alleged and if they were, they still would not have taught or disclosed the features of at least independent claims 1, 16 and 34”. However, the Examiner disagrees and find these arguments unpersuasive as well for each of the references later argued. Because the Applicant is alleging element features that are not reflected positively and actively recited while being actually supported by the Applicant’s disclosure in the claim limitations recitation. For instance, (1) the “fact and dimension table schema” is descriptive matter in the claims and are not functionally tied to a respective function, (2) the “iterative sufficiency testing” and “adaptive classifier integration with schema and roadmap” is not claimed or supported in this exact manner in the claim limitations recitation (3) the generation of the “roadmap” is still satisfied under the broadest reasonable interpretation (BRI) since it is not positively and actively reciting how it is directing users or investigators in the claim language. Rather, the recitation is broad enough for this element features to still be satisfied by the corresponding prior art of Busch, Adjoute and Bakalash that are already stated in the 35 USC § 103 rejection.
“Ordered Combination and Its Technical Effect” and “Lack of Motivation to Combine” in pp. 18 – 19: Applicant argues that these references lack the claimed “ordered combination” and “do not recognize or attempt to solve the problem of fragmented or incomplete fraud investigation datasets, which is at the heart of the claimed invention”. However, the Examiner disagrees and find these arguments unpersuasive. Firstly, because Applicant’s general allegations do not reflect what the Applicant actually claimed, as explained above. Also, these prior art references are in the field of endeavor for detecting fraud cases while Bakalash provides “a novel MDB-based system that enables fast knowledge discovery and accurate predictive business modeling for applications such as “financial/risk analysis, fraud management” and “business intelligence applications (e.g. Balanced Scorecard, Activity-Based Costing)” (C9; L25 – 33; Bakalash). Thus, this is reasonable since rationale to modify or combine the prior art does not have to be expressly stated in the prior art; the rationale may be expressly or impliedly contained in the prior art or it may be reasoned from knowledge generally available to one of ordinary skill in the art, established scientific principles, or legal precedent established by prior case law (see MPEP 2144(I)). Also, the Examiner considered, only the knowledge which was within the level of ordinary skill in the art, and does not include knowledge gleaned only from the applicant’s application. It is noticed that there is no requirement that an express, written motivation to combine must appear in prior art references before a finding of obviousness. (see MPEP 2145(X)(A)).
“Non-Conventional Ordered Combination” in p. 20: The related Applicant arguments regarding this section are moot since these assertions are misapplied and are not related or directed to the 35 USC § 103 rejection for obviousness in the claimed invention when in view of the prior art.
Thus, due to all the reasons stated above the Examiner maintains 35 USC § 103 rejection for these pending claims.
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 - 3, 7 - 14, 16 and 24 - 34 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The analysis of this claimed invention recited in the claims begins in view of independent claims 1, 16 and 34 which are separately evaluated, as follows:
At Step 1: Claims 1, 16 and 34 fall under statutory category of a machine.
At Step 2A Prong 1: Claim 1, 16 and 34 recite an abstract idea, which is defined by the following underlined elements (e.g. functional steps) while omitting any hardware components (e.g. represented as “…”) per claim:
For claim 1:
…configured to receive data inputs for an investigation case of fraud, waste or abuse;
…classifying the data inputs into a plurality of data categories of a framework of a fraud, waste, abuse or anomaly model…
the classifying the data inputs into the plurality of data categories is verified by the user such that the user accepts the classification or modifies the classification and any classification that is not accepted or modified is identified as an abnormal data point, wherein under the condition that there is an identification of an abnormal data point, …will automatically generate an analytic roadmap to investigate additional data inputs based on the abnormal data point and data inputs from more than one case of fraud, waste or abuse, and
…learns from the modification of the classification and identification of an abnormal data point;
…corresponding to the data categories of the framework, …programmed to sort the classified data inputs… by data category…
…
…to identify discoverable gaps…wherein …identifies the discoverable gaps by recognizing patterns of data…that show data is missing or expected using…that iteratively prompt for user-supplied data to resolve the gaps
For claim 16:
…corresponding to a plurality of data categories of a framework of a fraud, waste, abuse or anomaly model, …containing data inputs from prior investigated cases;
…programmed to identify discoverable gaps… identifies the discoverable gaps by recognizing patterns of data… that show data is missing or expected; and
…programmed on the data inputs from prior investigated cases in which totality of data was achieved to identify abnormal data points… will automatically generate an analytic roadmap to investigate additional data inputs based on the abnormal data points and data inputs from all of the prior investigated cases, and …learns from the identification of abnormal data points…and… learns from the identification of abnormal data points using …integrated with the snowflake schema fact table and dimension tables
For claim 34:
…configured to receive data inputs associated with one or more investigation cases, the data inputs comprising transactional, behavioral, and contextual information;
…configured to classify the data inputs into a plurality of data categories, the data categories comprising at least a player category, a rules-based category, a transparency category, and a consequence category; and
a behavioral model for players, wherein…is programmed to compare the classified data inputs to the behavioral model to identify deviations indicative of fraud, waste, abuse, or anomalies;
…storing: …corresponding to the data categories, each database comprising a plurality of data elements, and each data element being associated with a player component key pointing to a player in the player database; and
a fact table and dimension tables organized according to a snowflake schema, the fact table containing reference keys pointing to dimension tables…
…configured to: identify discoverable gaps in the data inputs by recognizing patterns indicative of missing or inconsistent data using…that iteratively prompt for user-supplied data to resolve the gaps; and
generate an analytic roadmap providing a structured sequence of investigative steps based on the identified discoverable gaps and deviations, the analytic roadmap guiding the user to input additional data for resolving the gaps or investigating anomalies; and
…adapt the behavioral model and classifier using…with the snowflake schema fact and dimension tables on modifications made by the user during classification and analysis; and
store and incorporate new data inputs from resolved investigation cases to improve the accuracy of fraud detection in subsequent cases.
These limitations generally, describe a system for identifying, understanding and categorizing user information and their behavior to investigate and analyze cases of fraud waste or abuse and obtain and understand the gaps or fragmentations of transaction information. As disclosed in the specification in ¶0030, this invention “establishes an analytical roadmap, a mechanism to mitigate inadequate, disconnected, un-unified, fragmented information while addressing personal bias and professional, political, psychosocial and socioeconomic conditions, allowing for a holistic "head-to-toe" approach to combating fraud, waste, abuse, and anomaly”. Thus, these limitations are directed to the abstract idea of a certain method of organizing human activity in the forms of “fundamental economic principles or practices” (by mitigating risk and insurance), “managing personal behavior or interactions between people” (through social activities) and “engaging in commercial or legal interactions” (through legal obligations). Because these claims recite the steps of “receiving” user data input related to investigation cases about fraudulent activity to “classify” the data into at least two categories of a “framework of a fraud, waste, abuse or anomaly model” in which the user “verifies” and “accepts” the classification or “modifies” the classification. If not, the data inputs are considered “abnormal data points” and the system will automatically generate a “roadmap” or an “analytic roadmap” to investigate additional data inputs and finally sort the classified data inputs into different databases and identify “discoverable gaps” based on patterns of data recognized as “missing or expected”. Thus, these claims recite the use and tracking of user’s social activities and behaviors via user inputs to predict and mitigate the risk of having fraudulent transactions and understanding the transaction behavior to comply with legal obligations such as federal and state regulations (e.g. the USA PATRIOT Act, the Bank Secrecy Act, and the Financial Crimes Enforcement Network, etc.; see pages 16 – 17 for the claimed invention final outputs in the Applicant’s disclosure).
The claims recite a judicial exception because the judicial exception is “described” in the claim (see MPEP 2106.04). Specifically, steps b – c, e and g in claim 1; steps b – c in claim 16 and steps b – c and f - h from claim 34, have steps reciting functions for classifying and sorting data inputs by categories, verify or modify the classifications by user to generate an analytic roadmap to investigate additional data and identify discoverable gaps by recognizing data patterns, which fall within the mental process grouping. Because these steps cover concepts performed in the human mind or with a pen and paper, including observation, evaluation, judgment, and opinion (See MPEP 2106.04(a)(2), subsection III). Also, in claims 1 and 34, the steps related with “identifying discoverable gaps” using “progressive and regressive algorithms” to sort classified data and to “show data missing or expected” (from claim 1 and 34 refer to ¶0036 from Applicant disclosure) and the steps of “adapt the behavioral model and classifier using adaptive pattern recognition algorithms…” and “…and the classifier learns from the identification of abnormal data points using adaptive pattern recognition algorithms” (from claims 16 and 34) are encompassing the performance of mathematical calculations. Thus, the above steps do not negate and further still reads in the mental nature of the limitation(s), when obtaining such information, as well as the concept is merely claimed to be performed on a generic computer and is merely using a computer as a tool to perform the concept recited in the steps (see MPEP 2106.04(a)(2)(III)(B & C)). Finally, all of these limitation steps from all independent claims are considered together as a single abstract idea for further analysis.
Step 2A Prong 2: For independent claims 1, 16 and 34, This judicial exception is not integrated into a practical application. Because the claims steps and their additional feature element(s) of a server, a plurality of databases; an artificial intelligence data inputs classifier, a data warehouse; (from claims 1, 16 and 34); a pooled database; adaptive recognition algorithms (from claim 16); a machine (from claim 1); a fraud, waste, abuse or anomaly model (from claims 1 and 16); using progressive and regressive algorithms (from claims 1 and 34); using adaptive pattern recognition algorithms (from claims 16 and 34), a data classification engine, an artificial intelligence-based classifier, and a behavioral model (from claim 34). These additional elements, individually and in combination, and while considering the claims as a whole, are is used as a tool to perform the abstract idea (refer to MPEP 2106.05(f)). These element features including the computer, data classification engine and the AI data inputs classifier with “non-generic” artificial intelligence techniques, including neural networks, decision trees, and expert systems that learns data modifications (e.g. iterative learning mechanism in claim 34), are recited at a high level of generality and are performed to generally apply the abstract idea without placing any limits on how the steps for at least “learning” this data modifications by the AI classifier are different from nominal use of other generic ML/AI models functions. The same reason subsequently applies to the “non-generic artificial intelligence techniques, including neural networks, decision trees, and expert systems” (from claim 1) which are not further detailed/described on how these elements would be different from generic ML/AI models. Rather, these limitations are recited in a high level of generality that does not limits how these steps are distinct from the types of models used in the classifier and how these are trained (e.g. “iterative learning mechanism” claimed in claim 34) from other ML/AI models. As for the recited “adaptive recognition algorithms” and “progressive and regressive algorithms” being used to treat the data, these are merely mentioned as instructions to apply in a computer and to general ML/AI technology. See MPEP 2106.05(f).
As for the “receive” steps in claims 1 and 34 and “store” data steps in claim 34 are really nothing more than links to computer implementing the use of ordinary capacity for implementing the use of 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 (refer to MPEP 2106.05 f (2)).
Thus, these limitations are also “merely indicating a field of use or technological environment in which to apply a judicial exception do not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application” (MPEP 2106.05(h)). In this case, the AI/ML technology and algorithms used for the “classifier” and to treat the received data inputs by the computer are broadly recited and lacks details on how this “identification” is particularly and specifically performed in contrast with general technology that achieve this same function. Rather, these limitations are simply limited to identifying “discoverable gaps” in missing/expected data, adapting the “behavioral model and classifier” and “learning” of the “classifier” for data classification that attempts to limit the use of the abstract idea of identifying fraud data to computer environments and/or the general AI/ML technology (see MPEP 2106.05(h) for examples (vi), (ix) and (x)). Therefore, this is indicative of the fact that the claim set has not integrated the abstract idea into a practical application and therefore, the claims are found to be directed to the abstract idea identified by the examiner.
Step 2B: For independent claims 1, 16 and 34, these claims do not provide an inventive concept. The recited additional elements of the claim(s) are the following: a server, a plurality of databases; an artificial intelligence data inputs classifier, a data warehouse; (from claims 1, 16 and 34); a pooled database; adaptive recognition algorithms (from claim 16); a machine (from claim 1); a fraud, waste, abuse or anomaly model (from claims 1 and 16); using progressive and regressive algorithms (from claims 1 and 34); using adaptive pattern recognition algorithms (from claims 16 and 34), a data classification engine, an artificial intelligence-based classifier, and a behavioral model (from claim 34). These additional elements are not sufficient to amount significantly more than the judicial exception. Meaning, that there are no additional element(s) claimed in the dependent claims that could be significantly more than the judicial exception, but rather, further recites the abstract idea. As indicated in Step 2A Prong 2, the additional element(s) in the claims are merely, using a generic computer device or computing technologies and/or other machinery merely as a tool to perform an abstract idea that does not constitute a practical application and only amounts to a mere instruction to practice the invention. Thus, these elements do not render the claims as being eligible (refer to MPEP 2106.05(f)) and 2106.05(h). This is because the claimed invention must improve “upon conventional functioning of a computer, or upon conventional technology or technological processes (e.g. such as the generically recited Machine Learning (ML) Models and the “learning” step and the steps directed to the “iterative learning mechanism” for the “data inputs classifier machine” disclosed in this case), a technical explanation as to how to implement the invention should be present in the specification. That is, the disclosure must provide sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement. The specification need not explicitly set forth the improvement, but it must describe the invention such that the improvement would be apparent to one of ordinary skill in the art” (see MPEP 2106.05(a)). The rationale set forth for the 2nd prong of the eligibility test above is also applicable and re-evaluated in step 2B. Therefore, this rationale is sufficient for its rejection basis as it is not patent eligible and no comments are necessary as it is also consistent with the MPEP 2106.
For dependent claims 2 - 3, 7 - 14 and 24 - 33, these claims cover or fall under the same abstract idea of a method of organizing human activity and mental process. They describe additional limitations steps of:
Claims 2 – 3, 10 – 12 and 24 – 33: further describes the abstract idea of the “artificial intelligence” system that iteratively investigates “cases of fraud, waste, abuse and anomaly” and its data inputs classifier (and its second classifier) and its categories as it further describing a classifier that might comprised of a group of “an expert system, machine learning, a decision tree, a neural network or combinations thereof” and its “data categories” which comprises to at least two of the groups categories (e.g. “players, benchmarks, functional information, rules-based, transparency, consequence and combinations thereof”). Thus, being directed to the abstract idea group of “managing personal behavior or interactions between people” as it is categorizing and understanding the user’s behaviors and its transactions.
Claims 7 and 13: further describes the abstract idea of the “artificial intelligence” system for “investigating cases of fraud, waste, abuse and anomaly” and its programming to identify discoverable gaps or “missing player component key” and alert a user about these “discoverable gap responsive to its identification” that clearly directs to the “managing personal behavior or relationships or interactions between people” and “engaging in commercial or legal interactions” when alerting user behaviors and their transaction activities due to law enforcement and compliance.
Claims 8 – 9 and 14: further describes the abstract idea of the “artificial intelligence” system for “investigating cases of fraud, waste, abuse and anomaly” and its second classifier in which is programmed to generate an “analytic roadmap of the investigation case” to aid in the investigation and compare and identify abnormal data points (including player component key values) in the “pooled database” which also attributes to the abstract idea group of “managing personal behavior or relationships or interactions between people” and “fundamental economic principles or practices” as well.
Step 2A Prong 2 and Step 2B: For dependent claims 8 - 12, these claims recite the additional elements: a behavioral model (from claim 8), a second classifier (from claims 8 – 9 and 12) an expert system, machine learning, a decision tree, a neural network or combinations thereof (from claims 10 - 12) which are also recited to be merely used as a tool to perform the abstract idea to identify “abnormal data points”. Thus, amounting to no more than mere instructions to “apply” the exception using a generic computer component (MPEP 2106 .05(f) and (f)(2)). Accordingly, for the same reasons stated above, these additional element(s) claimed cannot provide an inventive concept at Step 2B.
Therefore, the additional elements previously mentioned above, are nothing more than descriptive language about the elements that define the abstract idea, and these claims remain rejected under 101 as well.
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 - 3, 7 - 14, 16 and 24 - 34 are rejected under 35 U.S.C. 103 as being unpatentable over Busch (U.S. Pub No. 20160019479 A1) in view of Adjaoute (U.S. Pub No. 20150046181 A1) in further view of Bakalash (U.S. Patent No. 6408292 B1).
Regarding claim 1:
Busch teaches:
a server configured to receive data inputs for an investigation case of fraud, waste or abuse; (In ¶0023 – 24; Figs 1 – 2 and 5 – 6: teaches “system includes a model run by a server that will start to analyze and process known components” (e.g. “known behavioral components related to a case”) (see ¶0006). The system’s “roadmap component allows the user to provide data input” wherein the user “will be prompted to enter available and known components” to “identify the remaining inputs required within the identified six behavioral components.” (see ¶0024 – 25). Also, refer to claims 1 – 6 for more details.)
…the data categories comprising a players category and at least two categories selected from the group consisting of a benchmarks category, a functional information category, a rules-based category, a transparency category, and a consequence category; (In ¶0008; Fig 1 – 2 and 5: teaches in Fig. 2 “an illustration of a skeletal structure designed to support the methodological process includes data components required for a comprehensive outcome determination driven by six critical behavioral components: “Players Component,” “Benchmarks Component,” “Functional Information Component,” “Rules-Based Component,” “Transparency Component” and “Consequence Component”. These categories are then selected and validated by the user to create “relevant industry revenue cycle component(s) (see FIG. 5)” (see ¶0033). Also, refer to ¶0006 and claims 1 and 7 for general details of what each component (or “category”) consists of.)
wherein the classifying the data inputs into the plurality of data categories is verified by the user such that the user accepts the classification or modifies the classification and any classification that is not accepted or modified is identified as an abnormal data point, (In ¶0033; Fig. 5: teaches an example that relates to “a fraud victim, a female spouse in the process of getting divorced, who attempts to open up a checking account to deposit cash” (see ¶0032). Thus, the “Computer system provides and user validates the selection and/or creates relevant industry revenue cycle component(s) (see FIG. 5)” and ultimately, the system “provides and the user validates current and updated components of revenue cycle-system compares to data base and or new data inputs and progresses in the accumulation of data for interim outputs leading towards final out” which results in “discoverable gaps” that are considered from “isolating “normal” data driver components” from “exceptions” that are considered “abnormal” data points (see ¶0031, ¶0087 – 89 and claim 1 of this prior art). Also, see Fig. 7A – 7J in which the user is verifying their claims and requesting the “review verification information” including the review or approval of an auditor (see Fig. 7I).)
wherein under the condition that there is an identification of an abnormal data point, the classifier will automatically generate an analytic roadmap to investigate additional data inputs based on the abnormal data point and data inputs from more than one case of fraud, waste or abuse, and (In ¶0031; Figs. 1 – 5: teaches this conditional limitation for abnormal data points being identified, an analytic roadmap is generated to investigate additional data inputs which is directed to the systems’ “model and framework incorporates data drivers” that isolate ““normal” data driver components within the FWA-IIRB Model, Framework, and Analytic Roadmap” so the “exceptions” that are “abnormal,” result in “discoverable gaps that facilitate the FWA-IIRB Model, Framework, and Analytic Roadmap” by being derived to “channel prevention, detection and mitigation work flows within the FWA-IIRB Model, Framework, and Analytic Roadmap (FIGS. 1-5)” while providing “interactive, iterative, and/or reiterative behavioral methodology to complete the required data point, comprehensively identifying, assessing, and analyzing gaps leading towards an outcome determination”. For example, in ¶0097, the system’s “FWA-IIRB is comprehensive in data collection and effective in handling a wide variety of situations, players and industries. The details and data captured will determine a path or paths for the algorithm or data flow to follow. Simultaneously, the system builds data volume by creating additional data points and discovering gaps as the model/framework proceeds to final output/results”.)
a plurality of databases corresponding to the data categories of the framework… (In ¶0037; Fig. 6: teaches that the “Computer system conducts a human and entity capital analysis and feeds aggregated data pool of case... —computer system ties data back to original continuum components and feeds aggregated data pool of case” (see ¶0038 – 44 for more examples). Also, refer to claims 1 and 8 of this prior art for more details regarding the databases.)
the server programmed to sort the classified data inputs into the databases by data category and into a pooled database of the case; (In ¶0006; Figs. 1 – 3 and 6: teaches that the system has the function of sorting the classified data inputs by data category and into the pooled case which is directed to the function executed by the server when running the “model” (see ¶0006 and ¶0023) in which the “computer system” “analyzes and processes” the “known components” corresponding to the data categories which are also “tied” and “fed” in the “pool of case”. Also, in ¶0027 the system, in “conjunction with the model, an analytical roadmap is also provided, as a guide to arrange data points, create a plan, and/or accomplish or determine an outcome within the skeletal structure of the model. Data that drive the model/framework and roadmap may include data generated by each identified player and/or within or among the other behavioral continuum components”. Finally, in claim 1 from this prior art, discloses that the system’s server does “sort the data inputs”.)
server programming to identify discoverable gaps in the pooled database of the case, wherein the server programming identifies the discoverable gaps by recognizing patterns of data in the pooled database that show data is missing or expected… (In ¶0006; Figs 2 – 3: teaches that the system “includes a model run by a server that will start to analyze and process known components…When prompted by the system, the user may realize that she is missing data relating to one or more of the remaining inputs—thus learning of a “discoverable gap” that leads toward a conclusion of the investigation of the case… If new data becomes expected, the process is repeated in which the user either provides the missing data or is alerted to the discoverable gap, which may lead toward a conclusion of the investigation of the case”. Also, in ¶0026 the model can be a “pattern, mode of structure or formation within or among the behavioral continuum components” which implies that patterns of data are recognized. Finally, refer to ¶0031, ¶0045 and ¶0056 for general details and an example and claims 1 and 4 for “discoverable gaps” identification executed by the “server”.)
Busch teaches that its system’s server can learn through the “relational rule” when implementing previous cases (see claim 6; Busch) and learns “discoverable gaps” that leads toward a conclusion of the investigation of the case” (see ¶0006; Busch). However, Busch does not explicitly teach an artificial intelligence data inputs classifier as a machine using non-generic artificial intelligence techniques while learning from modified classification and identification of abnormal data points, the data elements tables having a player component key pointing to a player in the players database and the server using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps. However, Adjaoute teaches:
an artificial intelligence data inputs classifier, the classifier further configured as a machine, for classifying the data inputs into a plurality of data categories of a framework of a fraud, waste, abuse or anomaly model… (In ¶0058; Fig 1; Fig 2 (226 and 221 – 225); Fig 3A – 3B, Fig. 5, Fig. 7, Fig. 9 (908 and 916) and Fig. 10: teaches that the system’s “smart agents” (directed to the classifier) are “assisted and signaled by case-by-case analyses provided in real time by several classification models or technologies operating in parallel that independently “crunch” the claim data 200 according to their own styles and methods”. Moreover, the “smart agents” also use “profiling” (e.g. “Real-time profiling, long-term profiling, recursive profiling”) to “assign a final classification of good/maybe/bad to the claim.” (see ¶0062 – 63). Therefore, and in conjunction with the “Analytical engine 226” (directed to the classifier as a machine) which “includes real-time profilers that analyze the data as they come in, and use it to update aspects of each profile” (see ¶0067), it can also define “business rules” and can “map” them “for relevant claim data inputs” (see ¶0070). Also, refer to ¶0049 – 55 for “Data classification model building” details and to ¶0134 for the multiple “classification techniques” implemented in this prior art such as “smart-agents, real-time profiling, long-term profiling, recursive profiling, business rules, fuzzy technology, neural network algorithms, case-based reasoning, genetic algorithms, data mining, and adaptive learning”.).
the artificial intelligence data inputs classifier uses non-generic artificial intelligence techniques, including neural networks, decision trees, and expert systems, and learns from the modification of the classification and identification of an abnormal data point; (In ¶0174; Fig. 2 (226 and 221 – 225) ); Fig. 4; Figs. 7 – 8 and Fig. 10: teaches that the system can combine the implementation of “smart agents” (¶0102 and ¶0106 for more details) with other technologies such as “Neural Networks” that can “interpret historical data and help identify trends and patterns against which to compare subject cases” related to abnormal data (see ¶0072 and ¶0175 for more details) or can combine “adaptive learning” of “three learning techniques” such as “the automatic creation of profiles (smart-agents) from historical data (long-term profiling), the enrichment of smart-agents based on real-time activities, and adaptive learning carried by incremental learning algorithms” (see ¶0080) which is directed to the utilization of non-generic AI techniques such as adaptive pattern recognition algorithms, and thus this system can provide “major advantages over neural networks and pattern-recognition techniques” (see ¶0182), in accordance to ¶0036 – 37 applicant specs. Refer to ¶0049 wherein “decision trees” are used for the “data classification model” while this model learns to map the claim data. Also, refer to ¶0195 for an example of “Smart Agent for Billings-for-a-Non-Covered-Service-as-a-Covered-Service” and to ¶0240 in which a “real-time smart agent profiling engine 916 receives behavioral digests of the latest transactions 918 and uses them to update three populations of profiles 920-922” including the use of “trained models 912” to “understand what is “normal” for a particular card, merchant, and user device”.)
…each of the databases comprising a plurality of data elements tables and each of the data elements tables having a player component key pointing to a player in the players database… (In ¶0224; Fig 7 (700, 712, 714 and 716): teaches the plurality of data elements table with a player component key pointing to a player from the corresponding database was directed to the tables shown in ¶0224 for the different “profiles” (e.g. providers) that are followed by the “smart agents” (see ¶0222 – 223) and their corresponding behaviors that are based on each of their “claims data 704”.)
…wherein the server programming identifies the discoverable gaps by recognizing patterns of data in the pooled database that show data is missing or expected using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps (In ¶0050 – 51: teaches the use of a regressive algorithm that iteratively prompt for user-supplied data to resolve the gaps which is interpreted as the “Data classification model building” which begins “with a learning step and is followed by a test of the performance”. For the learning step, a “mapping function y=f(x)” is learned that can predict the associated class label y of a given tuple x” (e.g. regressive algorithm) and “once trained and tested, the variables reported in the claim data in each reporting category are fit to a tuple” (e.g. “tuples” are referred to user-supplied data; see ¶0114 for more tuple details). This “mapping function can be in the form of classification rules, decision trees, or mathematical formulae” directed to the progressive algorithm that via a “decision tree” employed it can “classify the claim data inputs” with tuples that are associated with “unknown” class labels to then be run through the “decision tree” to predict a “tuple’s class” (see ¶0050 – 51), in accordance to the example given in ¶0036 – 37 from Applicant disclosure. Further, the system can use genetic algorithms to “solve for unrecognized problems” and create “a high quality solution” to such problems which is directed to resolve the discoverable gaps (see ¶0054).)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the abilities of providing an artificial intelligence data inputs classifier as a machine using non-generic artificial intelligence techniques while learning from modified classification and identification of abnormal data points, the data elements tables having a player component key pointing to a player in the players database and the server using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps, as taught by Adjaoute in order to provide “Real-time profiling, long-term profiling, recursive profiling [which] are used in new situations the other classification models would miss, or misinterpret by getting out-of-date. Conventional classification models are limited to using what is immediately before them in the claim data and applying to that rules and structures that are inflexible. Profiling recognizes that normal behavior does not always appear normal in every sample. So in instances where scores from the classification models 226 cannot agree, the smart agents will use the profiling they maintain internally on the corresponding healthcare provider to assign a final classification of good/maybe/bad to the claim” (¶0062 – 63; Adjaoute).
Busch teaches the “reference key” values that correspond to the six behavioral components in accordance to the applicant specifications in ¶0032, which was directed to the “six behavioral components” disclosed (see ¶0006 and ¶0023; Busch). Neither, Busch or Adjaoute explicitly teach a data warehouse database comprising of a fact table containing this reference keys that points to dimension tables in each database. However, Bakalash teaches:
a data warehouse comprising the plurality of databases, the data warehouse comprising a fact table and dimension tables organized according to a snowflake schema, the fact table containing reference keys pointing to dimension tables in each of the databases: and (In C3; L19 – 31; Figs. 2A, 3A, 8C, Fig. 10A – 10B and Fig. 17: teaches the data warehouse with a fact table containing reference keys pointing to dimension tables, which was directed to the “Data Warehouse’s Metadata directory” of “every candidate element” or “data element” of the “MDB” (directed to the six behavioral components in accordance to ¶0032 from applicant specifications) which can be found “in the set of lists of the RDBMS-based Data Warehouse” and the “detailed structure data of the relational data (e.g. snowflakes), dimensions and hierarchy relations associated with the RDBMS-Data Warehouse” (see C17; L15 – 22) which is directed to having dimension tables organized according to a snowflake schema.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch as modified by Adjaoute with the ability of providing a data warehouse database comprising of a fact table containing this reference keys that points to dimension tables in each database, as taught by Bakalash in order to provide “a novel MDB-based system that enables fast knowledge discovery and accurate predictive business modeling for applications such as database marketing, financial/risk analysis, fraud management, bioinformatics, return-on- investment (ROI) justification, business intelligence applications (e.g. Balanced Scorecard, Activity-Based Costing), customer relations management (CRM), enterprise information portals and the like” (C9; L25 – 33; Bakalash).
Regarding claim 16:
Busch further teaches:
…corresponding to a plurality of data categories of a framework of a fraud, waste, abuse or anomaly model, (In ¶0008; Fig 1 – 2 and 5: teaches in Fig. 2 “an illustration of a skeletal structure designed to support the methodological process includes data components required for a comprehensive outcome determination driven by six critical behavioral components: “Players Component,” “Benchmarks Component,” “Functional Information Component,” “Rules-Based Component,” “Transparency Component” and “Consequence Component”. These categories are then selected and validated by the user to create “relevant industry revenue cycle component(s) (see FIG. 5)” (see ¶0033). Also, refer to ¶0006 for general details of what each component (or “category”) consists of which is related to a fraud, waste, abuse or anomaly model that analyzes and process known components.)
the databases containing data inputs from prior investigated cases, (In ¶0023 – 24; Fig. 6: teaches that the system’s “roadmap component allows the user to provide data input” wherein the user “will be prompted to enter available and known components” to “identify the remaining inputs required within the identified six behavioral components.” (see ¶0024 – 25) See claim 6 wherein the system learns “relational rules” to implement the system in previous cases which implies that historical data from investigated cases are stored in different databases as shown in Fig. 6.)
…the data categories comprising a players category and at least two categories selected from the group consisting of a benchmarks category, a functional information category, a rules-based category, a transparency category, and a consequence category, (In ¶0008; Fig 1 – 2 and 5: teaches in Fig. 2 “an illustration of a skeletal structure designed to support the methodological process includes data components required for a comprehensive outcome determination driven by six critical behavioral components: “Players Component,” “Benchmarks Component,” “Functional Information Component,” “Rules-Based Component,” “Transparency Component” and “Consequence Component”. These categories are then selected and validated by the user to create “relevant industry revenue cycle component(s) (see FIG. 5)” (see ¶0033). Also, refer to ¶0006 and claims 1 and 7 for general details of what each component (or “category”) consists of.)
a pooled database of the case; (In ¶0006; Figs. 1 – 3 and 6: teaches that the system has the function of sorting the classified data inputs by data category and into the pooled case which is directed to the function executed by the server when running the “model” (see ¶0006 and ¶0023) in which the “computer system” “analyzes and processes” the “known components” corresponding to the data categories which are also “tied” and “fed” in the “pool of case” directed to a pooled database of the case as shown in Fig. 6.)
a server programmed to identify discoverable gaps in the pooled database of the case, wherein the server programming identifies the discoverable gaps by recognizing patterns of data in the pooled database that show data is missing or expected; and (In ¶0006; Figs 2 – 3: teaches that the system “includes a model run by a server that will start to analyze and process known components…When prompted by the system, the user may realize that she is missing data relating to one or more of the remaining inputs—thus learning of a “discoverable gap” that leads toward a conclusion of the investigation of the case… If new data becomes expected, the process is repeated in which the user either provides the missing data or is alerted to the discoverable gap, which may lead toward a conclusion of the investigation of the case”. Also, in ¶0026 the model can be a “pattern, mode of structure or formation within or among the behavioral continuum components” which implies that patterns of data are recognized. Finally, refer to ¶0031, ¶0045 and ¶0056 for general details and an example and claims 1 and 4 for “discoverable gaps” identification executed by the “server”.)
a classifier programmed on the data inputs from prior investigated cases in which totality of data was achieved to identify abnormal data points in the pooled database wherein under the condition that abnormal data points are identified, the classifier will automatically generate an analytic roadmap to investigate additional data inputs based on the abnormal data points and data inputs from all of the prior investigated cases, (In ¶0031; Figs. 1 – 5: teaches this conditional limitation for abnormal data points being identified, an analytic roadmap is generated to investigate additional data inputs which is directed to the systems’ “model and framework incorporates data drivers” that isolate ““normal” data driver components within the FWA-IIRB Model, Framework, and Analytic Roadmap” so the “exceptions” that are “abnormal,” result in “discoverable gaps that facilitate the FWA-IIRB Model, Framework, and Analytic Roadmap” by being derived to “channel prevention, detection and mitigation work flows within the FWA-IIRB Model, Framework, and Analytic Roadmap (FIGS. 1-5)” while providing “interactive, iterative, and/or reiterative behavioral methodology to complete the required data point, comprehensively identifying, assessing, and analyzing gaps leading towards an outcome determination”. For example, in ¶0097, the system’s “FWA-IIRB is comprehensive in data collection and effective in handling a wide variety of situations, players and industries. The details and data captured will determine a path or paths for the algorithm or data flow to follow. Simultaneously, the system builds data volume by creating additional data points and discovering gaps as the model/framework proceeds to final output/results”.)
Busch teaches that its system’s server can learn through the “relational rule” when implementing previous cases (see claim 6; Busch) and learns “discoverable gaps” that leads toward a conclusion of the investigation of the case” (see ¶0006; Busch). However, Busch does not explicitly teach the abilities of having data elements tables having a player component key pointing to a player in the players database, the classifier learning from the abnormal points identified using adaptive pattern recognition algorithms and a data warehouse database comprising of a plurality of databases and a fact table containing this reference keys that points to dimension tables in each database. However, Adjaoute teaches the data elements tables having a player component key pointing to a player in the players database, the classifier learning from the abnormal points identified using adaptive pattern recognition algorithms:
each of the databases comprising a plurality of data elements tables and each of the data elements tables having a player component key pointing to a player in the players database; (In ¶0224; Fig 7 (700, 712, 714 and 716): teaches the plurality of data elements table with a player component key pointing to a player from the corresponding database was directed to the tables shown in ¶0224 for the different “profiles” (e.g. providers) that are followed by the “smart agents” (see ¶0222 – 223) and their corresponding behaviors that are based on each of their “claims data 704”.)
and the classifier learns from the identification of abnormal data points using adaptive pattern recognition algorithms integrated with the snowflake schema fact table and dimension tables (In ¶0174; Fig. 2 (226 and 221 – 225): teaches that the system can combine the implementation of “smart agents” with other technologies such as “Neural Networks” that can “interpret historical data and help identify trends and patterns against which to compare subject cases” (see ¶0072 and ¶0175 for more details) or can combine “adaptive learning” of “three learning techniques” such as “the automatic creation of profiles (smart-agents) from historical data (long-term profiling), the enrichment of smart-agents based on real-time activities, and adaptive learning carried by incremental learning algorithms” (see ¶0080) which is directed to the utilization of adaptive pattern recognition algorithms, and thus this system can provide “major advantages over neural networks and pattern-recognition techniques” (see ¶0182), in accordance to ¶0036 – 37 applicant specs .)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the abilities of having data elements tables having a player component key pointing to a player in the players database and the classifier learning from the abnormal points identified using adaptive pattern recognition algorithms, as taught by Adjaoute in order to provide “Real-time profiling, long-term profiling, recursive profiling [which] are used in new situations the other classification models would miss, or misinterpret by getting out-of-date. Conventional classification models are limited to using what is immediately before them in the claim data and applying to that rules and structures that are inflexible. Profiling recognizes that normal behavior does not always appear normal in every sample. So in instances where scores from the classification models 226 cannot agree, the smart agents will use the profiling they maintain internally on the corresponding healthcare provider to assign a final classification of good/maybe/bad to the claim” (¶0062 – 63; Adjaoute). Finally, the use of these technologies in the system can provide “several major advantages over neural networks and pattern-recognition techniques” and provide the “ability to use richer feature sets for describing examples makes them at least as accurate and many time more precise” (¶0182; Adjaoute).
Busch teaches the “reference key” values that correspond to the six behavioral components in accordance to the applicant specifications in ¶0032, which was directed to the “six behavioral components” disclosed (see ¶0006 and ¶0023; Busch). Neither, Busch or Adjaoute explicitly teach a data warehouse database comprising of a plurality of databases and a fact table containing this reference keys that points to dimension tables with a snowflake schema fact table in each database. However, Bakalash teaches:
a data warehouse comprising a plurality of databases…;(In C21; L23 – 32; Fig. 17: teaches that “a RDBMS-based consumer shopping profile Data Warehouse 31 for storing consumer shopping profile information (e.g. representative of buying patterns, interests, hobbies as a function of time, personal information, credit history, income, home and auto ownership, marital status, etc.) collected from electronic commerce based transactions, compiled databases, publicly-traded response databases and the like” which is directed to a data warehouse with multiple databases.)
the data warehouse comprising a fact table containing reference keys pointing to dimension tables in each of the databases, (In C3; L19 – 31; Figs. 2A, 3A, 8C, Fig. 10A – 10B and Fig. 17: teaches the data warehouse with a fact table containing reference keys pointing to dimension tables, which was directed to the “Data Warehouse’s Metadata directory” of “every candidate element” or “data element” of the “MDB” (directed to the six behavioral components in accordance to ¶0032 from applicant specifications) which can be found “in the set of lists of the RDBMS-based Data Warehouse” and the “detailed structure data of the relational data (e.g. snowflakes), dimensions and hierarchy relations associated with the RDBMS-Data Warehouse” (see C17; L15 – 22) which is directed to having dimension tables organized according to a snowflake schema.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch as modified by Adjaoute with the ability of data warehouse database comprising of a plurality of databases and a fact table containing this reference keys that points to dimension tables with a snowflake schema fact table in each database, as taught by Bakalash in order to provide “a novel MDB-based system that enables fast knowledge discovery and accurate predictive business modeling for applications such as database marketing, financial/risk analysis, fraud management, bioinformatics, return-on- investment (ROI) justification, business intelligence applications (e.g. Balanced Scorecard, Activity-Based Costing), customer relations management (CRM), enterprise information portals and the like” (C9; L25 – 33; Bakalash).
Regarding claim 34:
Busch further teaches:
a server configured to receive data inputs associated with one or more investigation cases, the data inputs comprising transactional, behavioral, and contextual information; (In ¶0023 – 24; Figs 1 – 2 and 5 – 6: teaches “system includes a model run by a server that will start to analyze and process known components” (e.g. “known behavioral components related to a case”) (see ¶0006) wherein the system “roadmap component allows the user to provide data input” wherein the user “will be prompted to enter available and known components” to “identify the remaining inputs required within the identified six behavioral components” (see ¶0024 – 25) which is directed to transactional, behavioral and contextual information under the broadest reasonable interpretation (BRI). Also, refer to claims 1 – 6 for more details.)
…the data categories comprising at least a player category, a rules-based category, a transparency category, and a consequence category; and (In ¶0008; Fig 1 – 2 and 5: teaches in Fig. 2 “an illustration of a skeletal structure designed to support the methodological process includes data components required for a comprehensive outcome determination driven by six critical behavioral components: “Players Component,” “Benchmarks Component,” “Functional Information Component,” “Rules-Based Component,” “Transparency Component” and “Consequence Component”. These categories are then selected and validated by the user to create “relevant industry revenue cycle component(s) (see FIG. 5)” (see ¶0033). Also, refer to ¶0006 and claims 1 and 7 for general details of what each component (or “category”) consists of.)
server programming configured to: identify discoverable gaps in the data inputs by recognizing patterns indicative of missing or inconsistent data…; and (In ¶0006; Figs 2 – 3: teaches that the system “includes a model run by a server that will start to analyze and process known components…When prompted by the system, the user may realize that she is missing data relating to one or more of the remaining inputs—thus learning of a “discoverable gap” that leads toward a conclusion of the investigation of the case… If new data becomes expected, the process is repeated in which the user either provides the missing data or is alerted to the discoverable gap, which may lead toward a conclusion of the investigation of the case”. Also, in ¶0026 the model can be a “pattern, mode of structure or formation within or among the behavioral continuum components” which implies that patterns of data are recognized. Finally, refer to ¶0031, ¶0045 and ¶0056 for general details and an example and claims 1 and 4 for “discoverable gaps” identification executed by the “server”.)
generate an analytic roadmap providing a structured sequence of investigative steps based on the identified discoverable gaps and deviations, the analytic roadmap guiding the user to input additional data for resolving the gaps or investigating anomalies; and (In ¶0031; Figs. 1 – 5: teaches this conditional limitation for abnormal data points being identified, an analytic roadmap is generated to investigate additional data inputs which is directed to the systems’ “model and framework incorporates data drivers” that isolate ““normal” data driver components within the FWA-IIRB Model, Framework, and Analytic Roadmap” so the “exceptions” that are “abnormal,” result in “discoverable gaps that facilitate the FWA-IIRB Model, Framework, and Analytic Roadmap” by being derived to “channel prevention, detection and mitigation work flows within the FWA-IIRB Model, Framework, and Analytic Roadmap (FIGS. 1-5)” while providing “interactive, iterative, and/or reiterative behavioral methodology to complete the required data point, comprehensively identifying, assessing, and analyzing gaps leading towards an outcome determination”. For example, in ¶0097, the system’s “FWA-IIRB is comprehensive in data collection and effective in handling a wide variety of situations, players and industries. The details and data captured will determine a path or paths for the algorithm or data flow to follow. Simultaneously, the system builds data volume by creating additional data points and discovering gaps as the model/framework proceeds to final output/results”.)
store and incorporate new data inputs from resolved investigation cases to improve the accuracy of fraud detection in subsequent cases. (In ¶0033; Fig. 5: teaches that the system “provides and the user validates current and updated components of revenue cycle-system compares to data base and or new data inputs and progresses in the accumulation of data for interim outputs leading towards final out” (directed to storing and incorporating new data inputs) which results in “discoverable gaps” that are considered from “isolating “normal” data driver components” from “exceptions” that are considered “abnormal” data points (see ¶0031, ¶0087 – 89 and claim 1 of this prior art).)
Busch does not explicitly teach the abilities of providing an artificial intelligence-based classifier that classifies the data inputs and compares the classified data inputs to the behavioral model to identify deviations, having data elements tables associated with a player component key pointing to a player in the players database, having an iterative learning mechanism for adapting the behavioral model and classifier using modifications made by the user and the server using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps. However, Adjaoute further teaches:
a data classification engine, having: an artificial intelligence-based classifier configured to classify the data inputs into a plurality of data categories…(In ¶0058; Fig 1; Fig 2 (226 and 221 – 225); Fig 3A – 3B, Fig. 5, Fig. 7, Fig. 9 (908 and 916) and Fig. 10: teaches that the system’s “smart agents” are “assisted and signaled by case-by-case analyses provided in real time by several classification models or technologies operating in parallel that independently “crunch” the claim data 200 according to their own styles and methods”. Moreover, the “smart agents” also use “profiling” (e.g. “Real-time profiling, long-term profiling, recursive profiling”) to “assign a final classification of good/maybe/bad to the claim.” (see ¶0062 – 63). Therefore, and in conjunction with the “Analytical engine 226” which “includes real-time profilers that analyze the data as they come in, and use it to update aspects of each profile” (see ¶0067), it can also define “business rules” and can “map” them “for relevant claim data inputs” (see ¶0070). Also, refer to ¶0049 – 55 for “Data classification model building” details and to ¶0134 for the multiple “classification techniques” implemented in this prior art such as “smart-agents, real-time profiling, long-term profiling, recursive profiling, business rules, fuzzy technology, neural network algorithms, case-based reasoning, genetic algorithms, data mining, and adaptive learning”.).
a behavioral model for players, wherein the classifier is programmed to compare the classified data inputs to the behavioral model to identify deviations indicative of fraud, waste, abuse, or anomalies; (In ¶0174; Fig. 2 (226 and 221 – 225): teaches that the system can combine the implementation of “smart agents” with other technologies such as “Neural Networks” that can “interpret historical data and help identify trends and patterns against which to compare subject cases” (see ¶0072 and ¶0175 for more details) which is directed to comparing classified data inputs to identify deviations, in accordance to ¶0036 – 37 applicant specs.)
…each database comprising a plurality of data elements, and each data element being associated with a player component key pointing to a player in the player database; and (In ¶0224; Fig 7 (700, 712, 714 and 716): teaches the plurality of data elements table with a player component key pointing to a player from the corresponding database was directed to the tables shown in ¶0224 for the different “profiles” (e.g. providers) that are followed by the “smart agents” (see ¶0222 – 223) and their corresponding behaviors that are based on each of their “claims data 704”.)
…using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps (In ¶0050 – 51: teaches the use of a regressive algorithm that iteratively prompt for user-supplied data to resolve the gaps which is interpreted as the “Data classification model building” which begins “with a learning step and is followed by a test of the performance”. For the learning step, a “mapping function y=f(x)” is learned that can predict the associated class label y of a given tuple x” (e.g. regressive algorithm) and “once trained and tested, the variables reported in the claim data in each reporting category are fit to a tuple” (e.g. “tuples” are referred to user-supplied data; see ¶0114 for more tuple details). This “mapping function can be in the form of classification rules, decision trees, or mathematical formulae” directed to the progressive algorithm that via a “decision tree” employed it can “classify the claim data inputs” with tuples that are associated with “unknown” class labels to then be run through the “decision tree” to predict a “tuple’s class” (see ¶0050 – 51), in accordance to the example given in ¶0036 – 37 from Applicant disclosure. Further, the system can use genetic algorithms to “solve for unrecognized problems” and create “a high quality solution” to such problems which is directed to resolve the discoverable gaps (see ¶0054).)
an iterative learning mechanism configured to adapt the behavioral model and classifier using adaptive pattern recognition algorithms with the snowflake schema fact and dimension tables on modifications made by the user during classification and analysis; and (In ¶0076: teaches that “Case-based reasoning (CBR) processes included in analytical engine 226” from the system, “use past experiences to help solve current problems” by searching “their case memories for a pre-existing case that matches the current input specifications” and when “new cases are solved they are added and continue to increase the database of cases solved” which is directed to an iterative learning mechanism. Also, the “analytical engine 226” utilizes a neural network which is “incremental and adaptive, allowing the size of the output classes to change dynamically” (see ¶0072). Also, healthcare providers (e.g. user) can be prompted to submit or supply additional claim information (see ¶0219 – 220))
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the abilities of providing an artificial intelligence-based classifier that classifies the data inputs and compares the classified data inputs to the behavioral model to identify deviations, having data elements tables associated with a player component key pointing to a player in the players database, having an iterative learning mechanism for adapting the behavioral model and classifier using modifications made by the user and the server using progressive and regressive algorithms that iteratively prompt for user-supplied data to resolve the gaps, as taught by Adjaoute in order to provide “Real-time profiling, long-term profiling, recursive profiling [which] are used in new situations the other classification models would miss, or misinterpret by getting out-of-date. Conventional classification models are limited to using what is immediately before them in the claim data and applying to that rules and structures that are inflexible. Profiling recognizes that normal behavior does not always appear normal in every sample. So in instances where scores from the classification models 226 cannot agree, the smart agents will use the profiling they maintain internally on the corresponding healthcare provider to assign a final classification of good/maybe/bad to the claim” (¶0062 – 63; Adjaoute).
Neither, Busch or Adjaoute explicitly teach a data warehouse comprising of a plurality of databases and a fact table containing this reference keys that points to dimension tables with a snowflake schema fact table in each database. However, Bakalash further teaches:
a data warehouse storing: a plurality of structured databases corresponding to the data categories… (In C21; L23 – 32; Fig. 17: teaches that “a RDBMS-based consumer shopping profile Data Warehouse 31 for storing consumer shopping profile information (e.g. representative of buying patterns, interests, hobbies as a function of time, personal information, credit history, income, home and auto ownership, marital status, etc.) collected from electronic commerce based transactions, compiled databases, publicly-traded response databases and the like” which is directed to a data warehouse with multiple databases.)
a fact table and dimension tables organized according to a snowflake schema, the fact table containing reference keys pointing to dimension tables in each of the structured databases; (In C3; L19 – 31; Figs. 2A, 3A, 8C, Fig. 10A – 10B and Fig. 17: teaches the data warehouse with a fact table containing reference keys pointing to dimension tables, which was directed to the “Data Warehouse’s Metadata directory” of “every candidate element” or “data element” of the “MDB” (directed to the six behavioral components in accordance to ¶0032 from applicant specifications) which can be found “in the set of lists of the RDBMS-based Data Warehouse” and the “detailed structure data of the relational data (e.g. snowflakes), dimensions and hierarchy relations associated with the RDBMS-Data Warehouse” (see C17; L15 – 22) which is directed to having dimension tables organized according to a snowflake schema.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch as modified by Adjaoute with the ability of providing a data warehouse comprising of a plurality of databases and a fact table containing this reference keys that points to dimension tables with a snowflake schema fact table in each database, as taught by Bakalash in order to provide “a novel MDB-based system that enables fast knowledge discovery and accurate predictive business modeling for applications such as database marketing, financial/risk analysis, fraud management, bioinformatics, return-on- investment (ROI) justification, business intelligence applications (e.g. Balanced Scorecard, Activity-Based Costing), customer relations management (CRM), enterprise information portals and the like” (C9; L25 – 33; Bakalash).
Regarding claim 2:
The combination of Busch, Adjaoute and Bakalash, shown in the rejection above, discloses the limitations of claim 1.
Busch does not explicitly teach the incorporation of artificial intelligence for the “data inputs classifier” in the system. However, Adjaoute further teaches:
wherein the data inputs classifier comprises artificial intelligence selected from the group consisting of a decision tree, a neural network, an expert system and combinations thereof. (In ¶0072; Fig 2 (226): teaches a data inputs classifier that is directed to the “Analytical engine 226” and its “smart-agents” which “includes the use of neural networks that interpret historical data to identify trends and patterns against which to compare subject cases” (e.g. machine learning and neural networks). Also, in ¶0079 – 80, the “Analytical engine” uses “data mining methods to extract implicit, previously unknown and potentially useful information from claims database 208. E.g., to distill a Decision Tree”, uses “particular classifications, association rules and analyzing sequences are used to extract data” and uses “adaptive learning” to combine “three learning techniques, the automatic creation of profiles (smart-agents) from historical data (long-term profiling), the enrichment of smart-agents based on real-time activities, and adaptive learning carried by incremental learning algorithms”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the ability of incorporating artificial intelligence such as neural nets and/or decision trees in the “data inputs classifier”, as taught by Adjaoute in order to provide significant acceleration of the “convergence over conventional back propagation, and other neural network algorithms” and “neural net is incremental and adaptive, allowing the size of the output classes to change dynamically” as well as having a “a well-trained neural network will be more successful” when having “more samples (experiences) to which a neural network has a direct correlation, the greater the likelihood of its success.” (¶0072 and ¶0074; Adjaoute).
Regarding claims 3 and 28 – 29:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 1 and 16, respectively.
This dependent claim set is represented by claim 3
Busch further teaches:
wherein the data categories comprises at least one category selected from the group consisting of a players category, a benchmarks category, a functional information category, a rules-based category, a transparency category, a consequence category and combinations thereof. (In ¶0008; Figs. 1 – 2 and 6: teaches in Fig 2 “data components” that are “driven by six critical behavioral components: “Players Component,” “Benchmarks Component,” “Functional Information Component,” “Rules-Based Component,” “Transparency Component” and “Consequence Component.” Also, refer to ¶0006 and claims 1 and 7 from this prior art for general details of what each component (or “category”) consists of.)
Regarding claim 7:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claim 1.
Busch further teaches:
further comprising programming to identify a missing player component key from a data input. (In ¶0006; Fig 1 – 2: teaches that the system identifies missing player component key from data input from the user by prompting the user to provide “supplemental data or any previously provided data” that is determined if its “normal or abnormal in view of the new totality of data” and “whether any other data not yet obtained becomes expected in view of the new totality of data provided to the system. If new data becomes expected, the process is repeated in which the user either provides the missing data or is alerted to the discoverable gap, which may lead toward a conclusion of the investigation of the case”. Also, refer to claim 4 and to ¶0004 and ¶0023 for general details of “data points” and “key metric components or key elements”.)
Regarding claim 8:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claim 1.
Busch does not explicitly teach the comparing data with a “second classifier” to identify abnormal data. However, Adjaoute teaches:
further comprising a second classifier, the second classifier comprising a behavioral model for players, the second classifier programmed to compare data having the same value for player component key to the behavioral model for players to identify abnormal data (In ¶0076; Fig 2 (226): teaches the second classifier which is directed to the combination of “neural networks” (see ¶0072) by identifying “trends and patterns against which to compare subject cases” and Case-based reasoning (CBR) processes” from the multiple classification models of the “Analytical engine 226” to detect abnormal datapoints based on the comparison of the providers’ historical behavior by searching “case memories for a pre-existing case that matches the current input specifications”. Also, refer to ¶0067 for more details regarding the “Analytical engine’s” “real-time profilers”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the ability of having a “second classifier” to identify abnormal data, as taught by Adjaoute in order to provide the “neural networks and Case-based reasoning (CBR) processes” of the “Analytical engine 226”, which are also used by the “smart agents” as the “Analytical engine’s” “Neural networks here translate a database into neurons without user intervention, and will significantly accelerate the speed of convergence over conventional back propagation, and other neural network algorithms. The present invention's neural net is incremental and adaptive, allowing the size of the output classes to change dynamically…So, clearly, a well-trained neural network will be more successful than a poorly trained neural network (the training referring to its environment and the experiences and samples that help shape it). The more samples (experiences) to which a neural network has a direct correlation, the greater the likelihood of its success” and similarly, the “Case-based reasoning (CBR) processes” can increase the likelihood of success” as well (¶0072 and ¶0076; Adjaoute).
Regarding claim 9 and 14:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 8 and 1, respectively.
This claim set is represented by claim 14
Busch does not explicitly teach the function of identifying the abnormal data via the generation of an analytic roadmap of the investigation case for the “second classifier” to perform such function. However, Adjaoute further teaches:
further comprising a second classifier programmed to generate an analytic roadmap of the investigation case to aid in the investigation and programmed to identify abnormal data points in the pooled database. (In ¶0049; Fig 2 (226): teaches that the system has a “Data classification model” directed to the generation of an “analytic roadmap” that includes a “a mapping function y=f(x) is learned that can predict the associated class label y of a given tuple x” and can “be in the form of classification rules, decision trees, or mathematical formulae”. Also, in ¶0070, the “Analytical engine 226 includes business rule checkers” that were directed to second classifiers to “test for activities that fall outside business norms or that are in violation of the business' policies, or that experts have concluded are telltale of problems. This then implies the business rules are defined in the programming, and a mapping to them is provided for relevant claim data inputs”. In other words, the “business rule checkers” (e.g. second classifier) from the “Analytical engine 226” detect abnormal datapoints and “map” them for “relevant claim data inputs” in combination with the “data classification model” (e.g. the analytic roadmap) used by “smart agents”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the ability of having a “second classifier” to specifically perform the function of identifying the abnormal data via the generation of an “analytic roadmap of the investigation case”, as taught by Adjaoute in order to provide a “computer [that] can be programmed to collect and profile various facets and aspects of the behavior of healthcare providers. The resulting behavioral profiles can be usefully employed to anticipate, adapt, and predict even never before seen frauds.” (¶0106; Adjaoute) as well as the combination of multiple “different classification techniques” to align them into “one cohesive whole to cooperatively model, personalize, individualize, group, compare, and profile the behaviors of individuals and institutions according to their respective locations, specialties, seasons, time, month, year, place, neighborhood, and other normalizing measures” (¶0134; Adjaoute).
Regarding claim 10:
The combination of Busch, Adjaoute and Bakalash, shown in the rejection above, discloses the limitations of claim 8.
Busch does not explicitly teach the incorporation of artificial intelligence for the “second classifier” in the system. However, Adjaoute further teaches:
wherein the second classifier comprises an expert system, machine learning, a decision tree, a neural network or combinations thereof. (In ¶0072; Fig 2 (226): teaches a second classifier that is directed to the “Analytical engine 226” and its “smart-agents” which “includes the use of neural networks that interpret historical data to identify trends and patterns against which to compare subject cases” (e.g. machine learning and neural networks). Also, in ¶0079 – 80, the “Analytical engine” uses “data mining methods to extract implicit, previously unknown and potentially useful information from claims database 208. E.g., to distill a Decision Tree”, uses “particular classifications, association rules and analyzing sequences are used to extract data” and uses “adaptive learning” to combine “three learning techniques, the automatic creation of profiles (smart-agents) from historical data (long-term profiling), the enrichment of smart-agents based on real-time activities, and adaptive learning carried by incremental learning algorithms”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the ability of incorporating artificial intelligence such as neural nets and/or decision trees in the “second classifier”, as taught by Adjaoute in order to provide significant acceleration of the “convergence over conventional back propagation, and other neural network algorithms” and “neural net is incremental and adaptive, allowing the size of the output classes to change dynamically” as well as having a “a well-trained neural network will be more successful” when having “more samples (experiences) to which a neural network has a direct correlation, the greater the likelihood of its success.” (¶0072 and ¶0074; Adjaoute).
Regarding claims 11 and 12:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 1 and 11, respectively.
This dependent claim set is represented by claim 12
Busch does not explicitly teach the identification of discoverable gaps in which a server programming and its second classifier comprises of “machine learning” or an “expert system”. However, Adjaoute further teaches:
further comprising a second classifier wherein the second classifier comprises an expert system or machine learning. (In ¶0072 – 73; Fig 2 (200 and 226); Figs 3A – 3C: teaches that the system and its server (see Fig. 1 (102) and ¶0043) includes the “Analytical engine 226” which uses “neural networks that interpret historical data to identify trends and patterns against which to compare subject cases”. Also, in ¶0096, the system’s “smart agents 221 - 225” which is directed to the second classifiers can judge “claim data 220 to be suspicious in the context of its corresponding healthcare provider, then a score 241-245 will issue identifying the healthcare provider and details on why their claim is suspicious.” Thus, the identification of discoverable gaps has been interpreted as the combination of the “Analytical engine 226” and the “smart agents” that are within the “health care fraud prevention software-as-a-service (SaaS) 200 that receives medical payment claims by thousands, or even millions of healthcare industry healthcare providers and healthcare providers” which is in “a network of specialized servers” (see ¶0056). Thus, the “Analytical engine 226” and the “smart agents” use the NNs and “scores” to identify discoverable gaps (e.g. missing, unmatched and suspicious data) in the player’s or provider’s data based on the historical behavior data. Refer to ¶0153 – 0164 which mentions other prior art that use “various analytical fraud detection” such as “gap testing” to identify “missing values in sequential data” and to ¶0076 – 77 which discusses the “Case-based reasoning (CBR) processes included in analytical engine 226” which “CBR systems can learn, whenever a non-perfect match is found but the problem is nevertheless solved, the case parameters are added to the case memory for future reference” and its corresponding “CBR score” will be issued if “previous instances have discovered fraud with the present input variables” which was directed to identifying discoverable gaps.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to have provided Busch with the ability of having machine learning such as neural networks in the “second classifier”, as taught by Adjaoute in order to provide significant acceleration of the “convergence over conventional back propagation, and other neural network algorithms” and “neural net is incremental and adaptive, allowing the size of the output classes to change dynamically” as well as having a “a well-trained neural network will be more successful” when having “more samples (experiences) to which a neural network has a direct correlation, the greater the likelihood of its success.” (¶0072 and ¶0074; Adjaoute).
Regarding claim 13:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claim 11.
Busch further teaches:
wherein the system alerts a user to a discoverable gap responsive to its identification. (In ¶0006; Fig 1 – 5: teaches that the system alerts the user regarding a “discoverable gap, which may lead toward a conclusion of the investigation of the case. Likewise, if the system alerts the user to an abnormal data point, the discovery of the abnormal data point may lead toward a conclusion of the investigation of the case”. Also, refer to claim 1 for this prior art.)
Regarding claims 24 – 27:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 1 and 16 for every pair of claims, respectively.
This claim set is represented by claim 24
Busch further teaches:
wherein the data categories comprise at least three categories selected from the group consisting of the benchmarks category, the functional information category, the rules-based category, the transparency category, and the consequence category. (In ¶0006; Figs. 1 – 2 and 6: teaches the data categories that can comprise of three categories (claims 24 - 25) and four categories (claims 26 - 27) which are directed to the “six critical behavioral components, consisting of the “Player Component,” “Benchmark Component,” “Functional Informational Component,” “Rules-Based Component,” “Transparency Component,” and “Consequence Component” in which the user “may realize that she is missing data relating to one or more of the remaining inputs—thus learning of a “discoverable gap” that leads toward a conclusion of the investigation of the case” and “prompts the user to identify the remaining inputs” required from these categories as “behavioral components”. See ¶0008, ¶0025 and Fig. 2 for more categories details)
Regarding claims 30 and 32:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 1 and 16, respectively.
This dependent claim set is represented by claim 30
Busch further teaches:
wherein under the condition that a data input is classified as an abnormal data point, the user provides additional data inputs relative to the abnormal data point. (In ¶0006: this conditional limitation is satisfied as it is directed to the system’s condition that when “the user provides the missing data in response to the prompt, the system will determine whether the supplemental data or any previously provided data is normal or abnormal in view of the new totality of data provided to the system, and whether any other data not yet obtained becomes expected in view of the new totality of data provided to the system”. See ¶0033 and ¶0045 for various examples.)
Regarding claims 31 and 33:
The combination of Busch, Adjaoute and Bakalash, as shown in the rejection above, discloses the limitations of claims 1 and 16, respectively.
This dependent claim set is represented by claim 31
Busch further teaches:
wherein the system is iterative with the user or other users. (In ¶0006; Figs. 1 – 5: teaches that the system determines “whether the supplemental data or any previously provided data is normal or abnormal in view of the new totality of data provided to the system, and whether any other data not yet obtained becomes expected in view of the new totality of data provided to the system. If new data becomes expected, the process is repeated in which the user either provides the missing data or is alerted to the discoverable gap, which may lead toward a conclusion of the investigation of the case”. Thus, implying that the system is iterative with the user or other users (see Figs 1 – 5, ¶0031 and ¶0073 for more details).)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Kalyan (U.S. Pub No. 20160063501 A1) is pertinent because it “relates to a system for detecting banking frauds. More particularly, the present invention relates to a system for analyzing banking transaction data and finding similar fraud examples given one or several user defined specimen frauds.”
Chu (U.S. Pub No. 20200175518 A1) is pertinent because it “relates to an apparatus and method for real-time detection of fraudulent digital transactions.”
Pati (U.S. Pub No. 20210383407 A1) is pertinent because it “relates to the field of anomaly detection and risk analysis. More specifically, the invention belongs to advanced probabilistic feature engineering for generating a dataset of features for a Machine Learning model for a more accurate detection of anomaly such as risk or financial crime.”
Barbour (U.S. Pub No. 20120109821 A1) is pertinent because it is “online entity identity validation and transaction authorization for self-service channels provided to end users by financial institutions. Even more particularly, embodiments disclosed herein related to a system, method, and computer program product for adversarial masquerade detection and detection of potentially fraudulent or unauthorized transactions.”
Kramme (U.S. Patent No. 10832248 B1) is pertinent because it is “generally relates to financial fraud and, more specifically, to processing techniques that use customer data and/or machine learning to reduce false positive fraud alerts.”
Sampath (U.S. Pub No. 20190228419 A1) is pertinent because it is “directed to a self-learning system and method for detecting fraudulent transactions by analyzing data from disparate sources and autonomously learning and improving the detection ability and results quality of the system.”
Arrabothu (U.S. Pub No. 20190385170 A1) is pertinent because it is a “fraud detection for transactions, and more specifically, to an automatically-updating fraud detection system configured to aid in fraud detection”.
Klindworth (U.S. Pub No. 20140081652 A1) is pertinent because it is “directed to an invention defined as an “Automated Healthcare Risk Management System for Preventing And Detecting Healthcare Fraud, Abuse, Waste And Errors [which] is a software application and interface that assists law enforcement, investigators and risk management analysts by focusing their research, analysis, strategy, reporting, prevention and detection efforts on the highest risk and highest value claims, providers, healthcare merchants or beneficiaries for fraud, abuse, over-servicing, over-utilization, waste or errors”.
Hunt (U.S. Pub No. 20080288889 A1) is pertinent because it “relates to methods and systems for analyzing data, and more particularly to methods and systems for analyzing data associated with the sales and marketing efforts of enterprises.” Specifically, discloses visuals graphics regarding “Data Warehousing” in OLAP applications.
Dupont (U.S. Pub No. 20120137367 A1) is pertinent because it “falls into the general area of anomaly detection, and more particularly, anomaly in human behavior analyzed from electronic data. It can be applied to a number of domain-specific scenarios such as compliance monitoring or risk management in high-risk domains such as investment banking or intelligence. It is also applicable to malicious insider detection in areas ranging from corporate theft to counter-intelligence or the broader intelligence community.”
Lee (U.S. Pub No. 20180357643 A1) is pertinent because it “relates to a method of detecting abnormal financial transactions and an electronic apparatus thereof.”
Binns (U.S. Pub No. 20180293582 A1) is pertinent because it “relates generally to transaction security, specifically fraud detection and remediation.”
Zhang et. al, Medical Fraud and Abuse Detection System Based on Machine Learning is pertinent because it proposes “a neural network with fully connected layers and sparse convolution” to “quantify the disease–drug relationship into relationship score and do anomaly detection based on this relationship score and other features” and “adapt to the data imbalance and a relative probability score to measure the model’s performance.”
Bakker, Data Warehousing & On-Line Analytical Processing (2018) is pertinent because it discusses the technologies of a “Data Warehouse and On-line Analytical Processing.”
THIS ACTION IS MADE FINAL. 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 Ivonnemary Rivera Gonzalez whose telephone number is (571)272-6158. The examiner can normally be reached Mon - Fri 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, Nathan Uber can be reached at (571) 270-3923. 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.
/IVONNEMARY RIVERA GONZALEZ/Examiner, Art Unit 3626
/NATHAN C UBER/Supervisory Patent Examiner, Art Unit 3626