hartleyDETAILED 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 .
The following is a Non-Final Office Action in response to communications received August 11, 2025. Claim(s) 11-12, 14-15 and 19-54 have been canceled. Claim(s) 1 and 13 have been amended. No new claims have been added. Therefore, claims 1-10, 13 and 16-18 are pending and addressed below.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17 (e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant’s submission has been entered.
Priority
Application No. 18/318,514 filed 05/17/2023.
Assignee/Applicant Name: Avalara, Inc
Inventor(s): Chan, Andrew; Maselli, Michael; Vilis, Jurgis; Goldschmidt, Thomas; Nicolov, Nicolas
Response to Amendments/Arguments
Claim Rejections - 35 USC § 112(a)
The cancellation of claim 14 in response to the previous Office action rejection for failing to comply with written description requirement is sufficient to overcome the 112(a) rejection. The examiner withdraws the 112(a) rejection of claim 14.
Claim Rejections - 35 USC § 101
Applicant's arguments filed 08/11/2025 have been fully considered but they are not persuasive.
In the remarks applicant recites the claim limitations arguing that the confidence score of each of the resource codes output by the ML model is identified and based on a determination that a confidence score does not exceed a threshold the system transmits a prompt to a user device associated with the primary entity, receives a response to the prompts, determines a resultant resource code based on response to prompts, the resource codes, attributes and indications of the first/second domains and the model retrains based on the resultant resource codes, prompts and response to the prompts. Applicant points to MPEP 2106.04(a)(1), specifically, example vii where the neural network is trained on “digital facial images” and later trained on “digital facial images” that were incorrectly identified as facial images in the first training stage. Applicant interprets the MPEP teaching to be that the refining of training the ML model as patent eligible subject matter in the facial image digital example. Applicant argues that the amended claim 1, uses a trained ML model to determine resource codes and based on the confidence score of each resource code output by the ML model the system determines using threshold values exceeded that the resource code is correct. Where if the threshold value is not exceeded a prompt is transmitted to a user and the system receives user responses to the prompt, determines resultant resource code (refine resource code that classifies item) and based on response to prompts the resource codes, digital rules, attributes of the item and first/second domain is used as training data for retraining the classification ML model. Applicant argues that similar to the facial digital image example, the claim limitations similar refining the training of the ML model is patent eligible. The examiner respectfully disagrees with the premise of applicant’s argument. Example vii of MPEP 2106.04(a)(1), did not find patent eligibility because the ML model was refined. A more complete explanation for this example is found in example 39. The example found patent eligibility in providing a technical solution to a problem rooted in technology, not in the refining of a ML model. The example improved upon existing applications of ML models which were trained on a set of facial and non-facial images, which suffered from the inability to detect human faces in images with shifts, distortions and variations in scale or rotation of facial patterns in the image. To address this issue the technical solution was to first use a combination of first expanding training sets of facial images and second applying mathematical transformation functions on acquired facial images where the transformations included affine transformations such as rotating, shifting, mirroring and filtering transformation for example smoothing/contrast reduction. The ML model was trained on the expanded training set using stochastic learning using gradient loss function to adjust weights which increased false positives by using the expanded training data which was resolved by the minimization of the false positives through iterative algorithms where the model was retrained with the training data with false positives produced after face detection and the combination of the training set with the detect faces in distorted imaging while limiting false positive. Thereby providing a specific technical solution to a plurality of technical problems in using machine learning models training data sets. This is not the case of the current application, the refining of the model as claimed is not directed toward a particular technical solution to a problem in using machine learning models or using data sets with issues. Instead the refining of the model claimed is directed toward calculating more accurate outputs related to a business process using technology as a tool, where the results of the analysis by the ML model is outputted, and then manipulated by the user and business data and reinputted into the model in order to refine the output of the business data. The rejection is maintained.
In the remarks applicant argues that unlike Recentive v Fox, the amended claim limitations do more than apply the ML model in a new field. When considered as a whole the claimed system uses the ML as part of the process for classification of items with a specific process of refining the classification of items based on other data. Applicant argues that the classification of items based on digital rules, first/second domains (geographic regions), attributes of the items and responses to prompts from users, when the system detects that the model has incorrectly output potential classifications of an item the additional steps taken to refine the classification of the item provides a technical solution to a technical problem by machine classification. Applicant argues that typical machine classifiers assume outputs are correct and are unable to receive additional training based on incorrect classifications and therefore are unable to receive additional data for training to arrive at correct classification. The examiner respectfully disagrees that typical classification models are unable to received additional data for training for more accurate outputs. As evidence to rebut applicant’s statement the examiner provides: NPL article “Retraining an existing machine learning model with new data” by Stack OverFlow (2018); Optimal Strategy: Classification Models and Thresholds by Fanous (2021); “Turning the Crank: A simulation of Optimizing Retraining” by Duling (2020); Determining the best Classification threshold value for deep learning model” by Stack Overflow (2020); Finding the Best Classification Threshold in Imbalanced Classification” by Zou (2015). Accordingly the examiner is not persuaded that typical classification models cannot be retrained with additional data and that applying threshold values to prompt retraining is a new way of trigging the retraining of a model. The rejection is maintained.
Claim Rejections - 35 USC § 103
Applicant's arguments filed 08/11/2025 have been fully considered but they are no longer applicable.
In the remarks applicant argues that the prior art references fail to teach that the prior art references fail to teach looking up rules separate from applying a classification learning model. Applicant’s argument is moot as a new reference has been applied to address the amendment limitation that the rules are separate from applying ML model.
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 1-10, 13 and 16-18 are rejected under 35 U.S.C. § 101 because the instant application is directed to non-patentable subject matter. Specifically, the claims are directed toward at least one judicial exception without reciting additional elements that amount to significantly more than the judicial exception. The rationale for this determination is in accordance with the guidelines of USPTO, applies to all statutory categories, and is explained in detail below.
In reference to Claims 1-10, 13 and 16-18:
STEP 1. Per Step 1 of the two-step analysis, the claims are determined to include a system, as in independent Claim 1 and the dependent claims. Such systems fall under the statutory category of "machine." Therefore, the claims are directed to a statutory eligibility category.
STEP 2A Prong 1. The claimed invention is directed to an abstract idea without significantly more. System claim 1 recites a functional process 1) receiving a dataset 2) identifying first domain 3) identifying second domain 4) determining one or more attributes 5) applying ML model to the attributes, indication of first and second domain 6) output resource codes based on attributes 7) determining resultant resource code 8) identifying score for each outputted resource code 9) determining score for each code exceeds threshold level 10) transmitting prompts for each score not exceeding threshold 11) receiving a response 12) determining resource code based on prompt response, resource codes, digital rules, attributes and indication of first/second domain 13) retraining model based on resource code, prompts, and prompts response 14)
generating a response based on resource code.
When considered as a whole the claimed subject matter is directed toward a commercial process where data related to relationships between entities and identified domains (geographic regions) is analyzed the data according to rules to output resource codes representing attributes of an item and determining the one or more attributes domain indications and generating a response based on the result. The specification discloses operational buy/sell transaction is a use case of the relationship instance between entities where a code is generated for an item classification of a transaction. (see FIG. 14-16). Accordingly in light of the specification, the focus of the invention is to analyze transaction data based on attributes and domains in order to output resource codes representing one or more attributes of an item. Such concepts can be found in the abstract category of commercial interactions and sales activity.
The claimed limitations which under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer system and applying a ML model. That is, other than reciting system one or more processors, nothing in the claim element precludes the step from practically being performed in the mind.
The steps recite steps that can easily be performed in the human mind as mental processes because the steps of receiving datasets, identifying first domain (geographic region), identifying second domain (geographic region), determining one or more attributes related to an item, analyzing data to output resource codes, looking up rules, determining resultant code based on output, identifying scores, determining scores exceed thresholds, generating a response, mimic human thought processes of observation, evaluation and opinion, and communication of result which, where the data interpretation is perceptible only in the human mind. For example the receiving step mimics mental processes of observation. The limitations “identifying first domain (geographic region), identifying second domain (geographic region), determining one or more attributes related to an item, analyzing data to output resource codes, looking up rules, determining resultant code based on output, identifying scores, determining scores exceed threshold, receiving response, determining resource code based on information, generating a response”, mimic mental processes of evaluation and opinion, and communication of result. See In re TLI Commc'ns LLC Patent Litig., 823 F.3d 607, 611 (Fed. Cir. 2016); FairWarning IP, LLC v. Iatric Sys., Inc., 839 F.3d 1089, 1093-94 (Fed. Cir. 2016). That is except for the limitations “transmitting prompts” and “re-training …ML model”, nothing in the claim limitations cannot reasonably be performed using mental processes
These concepts are enumerated in Section I of the 2019 revised patent subject matter eligibility guidance published in the federal register (84 FR 50) on January 7, 2019) is directed toward abstract category of mental processes and methods of organizing human activity.
STEP 2A Prong 2: The identified judicial exception is not integrated into a practical application because the claims fail to provide indications of patent eligible subject matter that integrate the alleged abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a system comprising an one or more processors, one or more non-transitory computer readable media coupled to one or more processors having instructions stored and executed by the one or more processors and a classification machine learning model .
The claimed processor applied to perform the operations of “identifying a first domain…” and “identifying a second domain…” where in light of the specification the term “domain” refers to geographic region. The processor is merely applied to perform the “identifying” processes as part of analyzing business data for classification.
The additional element “classification machine learning model” that has been trained using data in a classification database that is applied to an “item” attributes”, an “indication of the first domain” and “second domain” to “obtain resource codes classifying the item” which amounts to no more than instructions to analyze an item for classification in order to determine resource codes in a business process. The model is further applied to output the resource codes based on the attributes and domains analyzed. The outputting of the codes by the model according to MPEP 2106.05(d) II (see also MPEP 2106.05(g)) is directed toward extra solution activity. The courts have recognized the following computer functions are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014)
The claim limitations “receiving a dataset…”, “transmitting one or more prompts” and “receiving a response to the one or more prompts” are not tied to any particular technology for performing the operation thereby lacking any technical details and similar to the outputting operation according to MPEP 2106.05(d) II, is insignificant extra solution activity.
The additional limitations recited in the claim that have not been tied to any particular technology include “looking up one or more digital rules…”, “determining a resultant resource code classifying the item” that has been outputted by the model, the “digital rules”, “one or more attributes”, and indication of “first” and “second” domain where the determination is determined by “identifying a confidence score for each resource code outputted…determining whether …one confidence score for each of the resource codes exceeds a threshold level…” are is recited at a high-level of generality such that it amounts to no more than analyzing the outputted data in a generic system environment components for the purpose of analyzing business related data for classification that is represented by resource codes. The further limitation “determining the resultant resource code …based on response to the …prompts,…resource codes…digital rules, …attributes…indications of first domain…and second domain” that is applied as data to be acted upon in the “retraining the classification machine learning model…” where the additional element “retraining” is not to improve technology but rather toward applying the retraining process for accurate reason code outputted results. When considered individually the limitations are not directed toward improving the model or solving a problem in the model analysis but rather to refine the data analyzed by the model in order to generate a response based on the determined resultant resource code. Instead the limitations when considered individual merely apply a system process and a classification model for performing insignificant extra solution activity or for analyzing item data for classification in a system environment for field of use. Taking the claim elements separately, the operation performed by the system at each step of the process is purely in terms of results desired and devoid of implementation of details.
The functions are is recited at a high-level of generality such that it amounts to no more than applying the exception using generic computer components.
when the claims are taken as a whole, as an ordered combination, the combination of limitations 1, 2-4 and 5 are directed toward analyzing received data related to a transaction to identify domains and determine attributes related to items, where a machine learning model is applied to output one or more resource codes based on limitations 1-4. The combination of limitations 1-5 and 6-9 is directed toward looking up rules based on domains applied in the analysis and determining a resultant resource code based on outputted codes by the applied model, the digital rules (looked up), attributes, indication of first and second domain where the determining is by identifying confidence scores or each code outputted by the model according to threshold level exceeded which is directed toward analyzing the outputted resource code for accuracy using a confidence score threshold a business practice for analyzing data for accuracy in an outputted result. The combination of limitations 1-8 and 10-14 is directed toward if a confidence score of the analysis 1-8 does not exceed a threshold transmitting a prompt to user device, where a response is received an applied for determining a resultant resource code which along with the one or more resource codes outputted, digital rules, attributes and indications of first and second domains are applied for retraining the classification model for a business practice.
Accordingly, when the claims are taken as a whole, as an ordered combination, the combination of steps not integrate the judicial exception into a practical application as the claim process fails to impose meaningful limits upon the abstract idea. Instead the claim limitations as a whole are directed toward receiving and analyzing item data information in order to output resource codes. The ML model is used to generally apply the abstract idea without limiting how the ML model functions. The model functions are described at a high level such that amounts to using a computer with a generic ML model to apply the abstract idea. This is true with respect to the additional element “retraining…model” as the claim limitations and specification lack technical description and merely apply the “retaining” process at a high level used to output more accurate resultant reason code results. These limitations recited analyzing data for an expected outcome without any details about how the analysis and result is accomplished. The claim provides no technical details regarding how the “applying” operation of the ML model is performed. The claimed subject matter is directed toward applying technology to receive, analyze transaction related data and output a result that is applied to generate a result. The combinations of parts is not directed toward any technical process or technological technique or technological solution to a problem rooted in technology.
Technology is not integral to the process as the claimed subject matter is so high level that any generic programming could be applied and the functions could be performed by any known means. Instead, similar to the claims at issue in Intellectual Ventures I LLC v. Capital One Financial Corp., 850 F.3d 1332 (Fed. Cir. 2017), “the claim language . . . provides only a result-oriented solution with insufficient detail for how a computer accomplishes it. Our law demands more.” Intellectual Ventures, 850 F.3d at 1342 (citing Elec. Power Grp. LLC v. Alstom, S.A., 830 F.3d 1350, 1356 (Fed. Cir. 2016)).
Accordingly, the claimed limitations fail to provide additional elements or combination or elements to apply or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The functions recited in the claims recite the concept of applying technology to analyze transaction data in order to generate a response which is a process directed toward a business practice. The specification discloses that the focus of the invention is to analyze transaction data based on attributes and domains in order to output resource codes representing one or more attributes of an item.
The integration of elements do not improve upon technology or improve upon computer functionality or capability in how computers carry out one of their basic functions. The integration of elements do not provide a process that allows computers to perform functions that previously could not be performed. The integration of elements do not provide a process which applies a relationship to apply a new way of using an application. The instant application, therefore, still appears only to implement the abstract idea to the particular technological environments apply what generic computer functionality in the related arts. The steps are still a combination made to use technology to apply an abstract idea and does not provide any of the determined indications of patent eligibility set forth in the 2019 USPTO 101 guidance. The additional steps only add to those abstract ideas using generic functions, and the claims do not show improved ways of, for example, an particular technical function for performing the abstract idea that imposes meaningful limits upon the abstract idea. Moreover, Examiner was not able to identify any specific technological processes that goes beyond merely confining the abstract idea in a particular technological environment, which, when considered in the ordered combination with the other steps, could have transformed the nature of the abstract idea previously identified. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
STEP 2B; The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because as discussed above with respect to concepts of the abstract idea into a practical application. The additional elements recited in the claim include a system comprising one or more processors, one or more non-transitory computer-readable storage media coupled to the one or more processors, the media having stored instructions executed by the one or more processors to perform the operations “receiving”, “identifying”, “determining”, “applying ML model”, “looking up rules”, “determining resultant resource code” and “generating a response” are recited in a generic way being employed in a customary manner. With respect to the system configuration, nearly every computer will include a “one or more processors, one or more non-transitory computer-readable storage media coupled to the one or more processors, the media having stored instructions executed by the one or more processors functions required by the system claims. Taking the claim elements separately, the function performed by the computer at each step of the process is purely conventional. Using a computer for “receiving”, “identifying”, “determining”, “applying ML model”, “looking up rules”, “determining resultant resource code”, “transmitting”, “retraining the …model” and “generating a response”----are some of the most basic functions of a computer. When the claims are taken as a whole, as an ordered combination, the combination of steps does not add “significantly more” by virtue of considering the steps as a whole, as an ordered combination. All of these computer functions are generic, routine, conventional computer activities that are performed only for their conventional uses. See Elec. Power Grp. v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016). Also see In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1316 (Fed. Cir. 2011) Absent a possible narrower construction of the terms “receiving”, “identifying”, “determining”, “applying ML model”, “looking up rules”, “determining resultant resource code” and “generating a response”... are functions can be achieved by any general purpose computer without special programming. None of these activities are used in some unconventional manner nor do any produce some unexpected result.
As to the data operated upon, "even if a process of collecting and analyzing information is 'limited to particular content' or a particular 'source,' that limitation does not make the collection and analysis other than abstract." SAP America, Inc. v. Invest Pic LLC, 898 F.3d 1161, 1168 (Fed. Cir. 2018). Considered as an ordered combination, the computer components of Applicant’s claimed functions add nothing that is not already present when the steps are considered separately. The sequence of data reception-analysis modification-transmission is equally generic and conventional. See Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 715 (Fed. Cir. 2014) (sequence of receiving, selecting, offering for exchange, display, allowing access, and receiving payment recited as an abstraction), Inventor Holdings, LLC v. Bed Bath & Beyond, Inc., 876 F.3d 1372, 1378 (Fed. Cir. 2017) (sequence of data retrieval, analysis, modification, generation, display, and transmission), Two-Way Media Ltd. v. Comcast Cable Communications, LLC, 874 F.3d 1329, 1339 (Fed. Cir. 2017) (sequence of processing, routing, controlling, and monitoring). The ordering of the steps is therefore ordinary and conventional. The analysis concludes that the claims do not provide an inventive concept because the additional elements recited in the claims do not provide significantly more than the recited judicial exception.
According to 2106.05 well-understood and routine processes to perform the abstract idea is not sufficient to transform the claim into patent eligibility. As evidence the examiner provides:
With respect to the “applying the ML model”, the claim limitations and specification lacks technical disclosure and states that any other ML modes can be applied (para 0269). The specification lacks technical disclosure, instead describes high level functions which focuses on the data the ML model analyzes and the outputting of resource codes based on the analysis rather than conventional technical process.
The specification discloses that the system in para 0039-0040 without any indication of an non-generic system.
[0105] In the present example, the operations and methods described with reference to the flowcharts illustrated in Figures 6 and 15 are described as being performed by the computer system 190 or computer system 1490. The operations and methods described with reference to the flowcharts illustrated in Figures 7-12B and 16-17 are described as being performed by the OSP 198 or OSP 1498. Although the operations and methods described with the flowcharts illustrated in Figures 6-12B and 15-17 are described as being performed by the computer system 190, computer system 1490, OSP 198 or OSP 1498, embodiments are not so limited, and any of the operations or
methods may be performed by any of the computer system 190, computer system 1490, OSP 198 or
OSP 1498.
Furthermore, similar to Electric Power Group, the claim limitations collect data, analyze data and output the result which is a conventional application of technology.
The model is further applied to output the resource codes based on the attributes and domains analyzed. The outputting of the codes by the model according to MPEP 2106.05(d) II (see also MPEP 2106.05(g)) is directed toward extra solution activity. The courts have recognized the following computer functions are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014)
The claim limitations “receiving a dataset…”, “transmitting one or more prompts” and “receiving a response to the one or more prompts” are not tied to any particular technology for performing the operation thereby lacking any technical details and similar to the outputting operation according to MPEP 2106.05(d) II, is insignificant extra solution activity.
With respect to the “retraining” of the model based on the determination that the retraining is required based on threshold applied using additional data evidence includes:
“Retraining an existing machine learning model with new data” by Stack OverFlow (2018); Optimal Strategy: Classification Models and Thresholds by Fanous (2021); “Turning the Crank: A simulation of Optimizing Retraining” by Duling (2020); Determining the best Classification threshold value for deep learning model” by Stack Overflow (2020); Finding the Best Classification Threshold in Imbalanced Classification” by Zou (2015).
The instant application, therefore, still appears to only implement the abstract ideas to the particular technological environments using what is generic components and functions in the related arts. The claim is not patent eligible.
The remaining dependent claims—which impose additional limitations—also fail to claim patent-eligible subject matter because the limitations cannot be considered statutory. In reference to claims 2-18 these dependent claim have also been reviewed with the same analysis as independent claim 1. Dependent claim 2 is directed toward accessing item sensed data in dataset, determining attributes related to item- collecting and analyzing data – directed toward a business process. Dependent claim 3 is directed toward generating data, generating item identity data, determining attributes based on item identify, text data and item-sensed data- analyzing transaction related data- a business practice. Dependent claim 4 is directed toward transmitting prompts to a user, receiving response to prompts, generating text data- insignificant extra solution activity of transmitting/receiving data and generating text related to a transaction process – a business process. Dependent claim 5 is directed toward additional data of a dataset, determining attributes -analyzing transaction related data- a business process. Dependent claim 6 and 7 are directed toward applying ML model to output text data- analyzing data and outputting result- well known and understood application of technology for a business process. Dependent claim 8 is directed toward data content- non-functional descriptive subject matter. Dependent claims 9 and 10 are directed toward transmitting results- insignificant extra solution activity. Dependent claim 13 is directed toward generating prompts – directed toward a business practice. Dependent claims 16 is directed toward determining first classification of [transaction] item, generating prompts based on classification, transmitting prompts to user, receiving a response to prompts, determining second classification and applying ML model to second classification of item. – applying technology in a conventional manner to classify and analyze transaction data, generate prompts to request and receive data from a user which is applied in order to determine and analyze a second classification-classifying transaction data, gathering additional data and performing a second analysis and classification using generic technology in a conventional manner- a business process. Dependent claim 17 is directed toward accessing data, determining whether identify of the item is recognized based on repository of data for classified items and based on item not recognized receiving user input and designating item data and identity of the item as training data- analyzing accessed transaction data for identity and if data is not identified receiving additional data and using the data as training data- analyzing and collecting data to apply as training data for business process. Dependent claim 18 is directed toward determining whether item data can be used to obtain resultant resource code, receiving user input if data cannot be used, and designating resultant resource code, item sensed data and identity of the item as training data- analyzing and collecting data to apply as training data for business process.
The dependent claim(s) have been examined individually and in combination with the preceding claims, however they do not cure the deficiencies of claim 1. Where all claims are directed to the same abstract idea, “addressing each claim of the asserted patents [is] unnecessary.” Content Extraction & Transmission LLC v. Wells Fargo Bank, Nat 7 Ass ’n, 776 F.3d 1343, 1348 (Fed. Cir. 2014). If applicant believes the dependent claims 2-10, 13 and 16-18 are directed towards patent eligible subject matter, they are invited to point out the specific limitations in the claim that are directed towards patent eligible subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-3, 5-10, 13 and 17-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over US Pub No. 2022/0058227 A1 by Balakrishnan et al (Balakrishnan) in view of US Pub. No. 2009/0006156 A1 by Hunt et al. (Hunt) in view of US Pub No. 2022/0237683 A1 by Castinado et al. (Castinado) and further in view of WO 2021/186176 A1 by Hartley et al (Hartley)
In reference to Claim 1:
Balakrishnan teaches:
(Currently Amended) A system ((Balakrishnan) in at least Abstract), comprising:
one or more processors ((Balakrishnan) in at least para 0042);
one or more non-transitory computer-readable storage media coupled to the one or more processors, the media having stored thereon instructions which, when executed by the one or more processors ((Balakrishnan) in at least para 0042), result in operations including at least:
receiving a dataset indicative of a relationship instance between a primary entity and a secondary entity, the dataset including data indicating an item that is to be classified; ((Balakrishnan) in at least FIG. 6B; para 0056-0057, para 0061-0062, para 0067, para 0069, para 0087, para 0097-0098, para 0106, para 0108, para 0117 wherein the prior art teaches input data based on user generated text, para 0120, para 0126, para 0145, para 0153 wherein the prior art teaches user input applied to determine relevant product data)…
identifying, by the one or more processors, a second domain based on the secondary entity ((Balakrishnan) in at least para 0106 wherein the prior art teaches identifying accurate product information including country of origin sent to ML modes for analysis; para 0113 wherein the prior art teaches product information include country of origin, para 0120);
determining, based on contents of the dataset, one or more attributes related to the item ((Balakrishnan) in at least Abstract; FIG. 6C-E, FIG. 19; para 0043, para 0053-0054, para 0058, para 0066-0067, para 0106, para 0152);
in response to the determining, applying a classification machine learning model to the one or more attributes, …and an indication of the second domain, the machine learning model being configured to output one or more resource codes based on one or more attributes, … and an indication of a second domain as inputs, and the classification machine learning model having been trained using data included in a classification database ((Balakrishnan) in at least FIG. 19; para 0043, para 0055, para 0062, para 0066-0069, para 0087-0088, para 0094, para 0098, para 0105, para 0108, para 0113, para 0117-0118, para 0120, para 0133, para 0152-0153, para 0156-0157); …
determining a resultant resource code classifying the item, based on the one or more resource codes outputted by the classification machine learning model, the one or more digital rules, the one or more attributes, …and the indication of the second domain ((Balakrishnan) in at least FIG. 11A; para 0092, para 0094, para 0096, para 0108-0114, para 0152, para 0156); by
identifying a confidence score for each of the one or more resource codes outputted by the classification machine learning model ((Balakrishnan) in at least para 0105 wherein the prior art teaches selecting from output of model product code with highest confidence score, para 0110, para 0112);
determining whether at least one confidence score for each of the one or more resource codes exceeds a threshold level ((Balakrishnan) in at least para 0110 wherein the prior art teaches selecting data records that meet a threshold with respect to similarity score and ranked accordingly where higher similarity scores can be vectorized, weighted and applied to the model for deep learning, para 0112 , para 0118 wherein the prior art teaches ranking product records and attributes that align with ML parameters, where a similarity score is calculated for each first and secondary attributes for generating a product identifier, para 0128); …
determining the resultant resource code based on the response to the one or more prompts [feed back/interaction options for selection], the one or more resource codes, the one or more digital rules, the one or more attributes, the indication of the first domain, and the indication of the second domain ((Balakrishnan) in at least Fig. 7B-C; para 0057 wherein the prior art teaches an interaction system to access user input on products para 0062, para 0067, para 0075-0079 wherein the prior art teaching automated interactions product page generating product pages to distinguish product variations for options available, where options prompt users to select attributes of products (size, color) where keywords, patterns are selected along with coordinates, para 0088, para 0134); and
re-training the classification machine learning model based on the resultant resource code, the one or more prompts, and the response to the one or more prompts ((Balakrishnan) in at least Fig. 11B ref # A224; para 0067, para 0088, para 0110-0111 wherein the prior art teaches selecting data records that meet a threshold with respect to similarity score and ranked according to attributes, accordingly where higher similarity scores can be vectorized, weighted and applied to the model for deep learning, para 0113 wherein the prior art teaches feeding data into ML models to estimate additional data in the product data which feeds a merged record into one or more ML models, para 0119, para 0153, para 0155, para 0157);
generating a response based on the determined resultant resource code ((Balakrishnan) in at least FIG. 7F, FIG. 9, FIG. 10D, FIG. 11B, FIG. 19; para 0081, para 0085-0090, para 0100, para 0114, para 0151-0152).
Although Balakrishnan does not explicitly teach the language “prompt”, the prior art does teach a continuation for extracting product attributes where such product attributes are collected from user interactions in product attribute options and other pertinent data. Interactive options and menu options are equitable with the term prompt for a response from the user. Accordingly the prior art provides some teaching or motivation that would have led one of ordinary skill in the art to arrive at the claimed invention.
Balakrishnan does not explicitly teach:
identifying, by the one or more processors, a first domain based on the primary entity;
in response to the determining, applying a machine learning model to … an indication of the first domain,…
looking up one or more digital rules based on the indication of the first domain, of the second domain, and the relationship instance, in which the one or more digital rules are identified by one or more entities associated with one or more domains and in which looking up the one or more digital rules is separate from applying the classification machine learning model;
based on a determination that at least one confidence score does not exceed a threshold level;
transmitting one or more prompts to a user device associated with the primary entity;
receiving a response to the one or more prompts;
Hunt teaches:
looking up one or more digital rules based on the indication of the first domain, of the second domain, and the relationship instance, , in which the one or more digital rules are identified by one or more entities associated with one or more domains and in which looking up the one or more digital rules is separate from applying the classification machine learning model((Hunt) in at least para 0117, para 0134, para 0293-0294, para 0296, para 0349, para 0387-0390)
Both Balakrishnan and Hunt teach that in analyzing market data that it is needed to applying parameters related to customer/product attributes in order to focus the analysis. Hunt teaches the motivation in data analysis when large amounts of data is available for analysis to project data related to products in order to predict likely outcomes of decisions related to enterprise activities using sample data sets from a larger universe of data and that one means of determining sample data sets related to market analysis is to perform a query process which queries for example, querying rules associated with products, geographies and so on from a dimension management facility and query processing facility so that the rules may be implements from a hierarchy of rules chosen used to provide governance to the query. It would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to expand the details of attributes and other analysis requirements of analyzing market to include lookup/discovering rules to apply as taught by Hunt since Hunt teaches the motivation in data analysis when large amounts of data is available for analysis to project data related to products in order to predict likely outcomes of decisions related to enterprise activities using sample data sets from a larger universe of data and that one means of determining sample data sets related to market analysis is to perform a query process which queries for example, querying rules associated with products, geographies and so on from a dimension management facility and query processing facility so that the rules may be implements from a hierarchy of rules chosen used to provide governance to the query.
Castinado teaches:
identifying, by the one or more processors, a first domain based on the primary entity ((Castinado) in at least Abstract; para 0004, para 0035, para 0052, ;
identifying, by the one or more processors, a second domain based on the secondary entity ((Castinado) in at least Abstract; para 0004, para 0033, para 0047, para 0052)
in response to the determining, applying a machine learning model to one or more attributes, an indication of the first domain, and an indication of a second domain…((Castinado) in at least para 0033, para 0035, para 0047-0048, para 0052)
According to KSR, known work in one field of endeavor may prompt variations of it for use in either the same field or a different one based on design incentives, if the variations are predictable to one of ordinary skill in the art. The scope and content of the prior art includes a similar analogous element when analyzing and classifying products using ML modeling. The prior art Castinado provides design incentives such as different regions have different market demands which would have prompted one of ordinary skill in the to adapt the attributes analyzed by the ML model to include geographic domains. The prior art Balakrishnan determines locations of second entity and first entity by using URL’s whereas Castinado determines locations of second and first entity by using geographic locations. Therefore, the prior art references provide evidences that the differences between the claimed product and the prior art where known variations or in a principle known in the art. Therefore, in view of the identified design incentives or other market forces, could have implemented the claimed variation of the prior art, and the claimed variation would have been predictable to one of ordinary skill in the art
Both Balakrishnan and Castinado are directed toward applying machine learning to analyze product data attributes which include data related to product feature and geographic information to output a result of the analysis. Castinado teaches the motivation of identifying and analyzing customer and vender geographic data in order to identify in which specific states, regions, towns, zip codes that customers purchase specific goods offered by a merchant in order to recommend to the merchant to start marketing or discounting goods in specific regions. It would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to expand the data collected, identified and analyzed by the ML model of Balakrishnan to include the merchant and customer geographic data as taught by Castindo since Castinado teaches the motivation of identifying and analyzing customer and vender geographic data in order to identify in which specific states, regions, towns, zip codes that customers purchase specific goods offered by a merchant in order to recommend to the merchant to start marketing or discounting goods in specific regions.
Hartley teaches:
based on a determination that at least one confidence score does not exceed a threshold level. ((Hartley) in at least page 3 lines 17-page 4 lines 1-6; page 11 lines 14-25 wherein the prior art teaches verification logic fails on accuracy in both test and training, sent back for a new training set ; page 29 lines 27-29, )
transmitting one or more prompts to a user device associated with the primary entity ((Hartley) in at least page 17 lines 20-page 19 lines 1-4, page 22 lines 24-30, page 29 lines 30-page 30 lines 1-5, lines 10-page 31 lines 1-29 wherein the prior art teaches in test phase deploying model onto local devices);
receiving a response to the one or more promp