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 .
This Office Action is responsive to Applicant's amendment filed on 30 September 2026. Applicant’s amendment on 30 September 2026 amended Claims 1, and 28. Claims 31 and 32 were previously rejected. Currently Claims 1-30 are pending and have been examined. The Examiner notes that the 101 rejection has been maintained.
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 filed on 3 March 2025 has been entered.
Claim Objection
The Examiner finds the Applicant’s arguments persuasive with respect to the prior art rejections, therefore claims 1-30 are objected to and would be viewed as allowable if rewritten or amended to overcome the 101 rejection(s).
Response to Arguments
Applicant's arguments filed 30 September 2026 have been fully considered but they are not persuasive.
The Applicant argues on page 3 that “Claim 1 is not directed merely to "providing a set of price estimates for multiple forms of transportation", as alleged by the Office on pages 18, 19, and 22 of the Office Action. That characterization abstracts the claim at too high a level and omits the ordered combination actually recited. Claim 1 requires, among other things: receiving observable live bookable prices that do not include complete fare class information; storing those prices as an incomplete historical travel-related price dataset that uses less storage than a complete historical dataset; training and storing classifiers comprising structured data including derived statistics; calculating estimates from stored classifiers; querying a Distribution System; comparing the calculated estimates with Distribution System fare prices; inferring fare class availability; outputting and storing the inferred fare class availability; and periodically repeating the process, including periodic retraining and periodic storage”.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the Applicant's argument that the Office improperly characterized claim 1 as directed merely to "providing a set of price estimates for multiple forms of transportation" is partially well-taken, and the Examiner acknowledges that characterization. The Examiner agrees that the December 5, 2025 Memorandum from the Deputy Commissioner for Patents, incorporating Ex Parte Desjardins, Appeal No. 2024-000567 into the MPEP, instructs that examiners and panels "should not evaluate claims at such a high level of generality that potentially meaningful technical limitations are dismissed without adequate explanation," and that examiners should "be careful to avoid oversimplifying the claims by looking at them generally and failing to account for the specific requirements of the claims." See also McRO, Inc. v. Bandai Namco Games Am. Inc. The Examiner therefore agrees that the phrase "providing a set of price estimates for multiple forms of transportation" alone was an overly compressed summary of the claim and does not by itself constitute an adequate articulation of the identified abstract idea for purposes of the 101 analysis.
However, this concession does not overcome the rejection. The problem with Applicant's argument is that it conflates the level of generality used to label the abstract idea with the actual analytical work of the 101 framework. Acknowledging that a shorthand label was overly compressed does not mean that the claim as a whole fails to recite a judicial exception, nor does it mean that the additional elements beyond that exception integrate it into a practical application or provide significantly more. The USPTO's subject matter eligibility framework does not require that the abstract idea be reducible to a single pithy phrase; it requires that the examiner identify the specific limitations in the claim that recite a judicial exception and explain why those limitations fall within at least one of the enumerated groupings. See MPEP 2106.04(a); 2019 PEG. That specific limitation-by-limitation analysis has been performed in the current Office Action and is maintained here.
Specifically, and consistent with the guidance against oversimplification, the identified abstract idea is more precisely the concept of: collecting and storing an incomplete historical dataset of observable travel-related prices; grouping those historical prices by category; deriving statistics from those groups; training probabilistic classifiers (Naïve Bayes) using those statistics to predict unobserved prices; using the trained classifiers to calculate estimated fare prices for a requested journey; comparing those estimated prices against prices received from a Distribution System; and inferring fare class availability from that comparison with all steps periodically repeated to update the stored inferred availability. This is not a vague or overbroad characterization; it tracks the specific claim language step by step. Evaluated at this level of specificity, the identified abstract idea still falls squarely within the mental processes grouping encompassing the observations, evaluations, comparisons, and inferences that make up the core analytical process as well as the mathematical concepts grouping to the extent the Naïve Bayes posterior probability calculations are expressly recited. See MPEP 2106.04(a)(2), subsections I and III; CyberSource Corp. v. Retail Decisions, Inc.; Electric Power Group, LLC v. Alstom S.A.
Turning to Applicant's enumeration of the specific ordered steps receiving observable live bookable prices without complete fare class information; storing those prices as an incomplete historical dataset using less storage than a complete dataset; training and storing classifiers comprising structured data including derived statistics; calculating estimates from stored classifiers; querying a Distribution System; comparing calculated estimates with Distribution System fare prices; inferring fare class availability; outputting and storing the inferred availability; and periodically repeating the process including periodic retraining and periodic storage the Examiner has carefully considered each of these steps and their combination and finds that they do not alter the eligibility analysis for the following reasons.
The fact that these steps form a more detailed and intricate ordered combination than the shorthand label suggested does not transform the abstract analytical process into a practical application of that process. An abstract idea does not become patent eligible simply because it is recited with greater operational specificity or in a longer ordered sequence. See Alice Corp. v. CLS Bank Int'l, ("Steps that do no more than require a generic computer to perform generic computer functions" do not supply the necessary inventive concept). Each of the steps Applicant enumerates receiving, storing, grouping, deriving statistics, training classifiers, calculating estimates, querying, comparing, inferring, outputting, and periodically repeating is, at the level of generality recited in the claim, an instance of the abstract analytical and organizational process the Examiner has identified: data gathering, mathematical/statistical calculation, mental comparison, inference, and output. The addition of specific sequential organization to a set of abstract steps does not integrate those steps into a practical application if the sequence itself does not reflect a technological improvement to a computer or other technology, impose a meaningful technological constraint on the abstract idea, or apply the abstract idea by use of a particular machine, transformation, or other qualifying consideration. See MPEP 2106.04(d), 2106.05(a)–(h).
The Examiner acknowledges that the ordered combination as articulated by Applicant reflects a more detailed characterization of the claim than the prior shorthand label, and the rejection has been framed accordingly in the present Office Action. However, that more detailed characterization, carefully considered, confirms rather than undermines the rejection: the ordered combination describes, in procedural detail, a workflow for performing an abstract analytical process price estimation and fare class inference through statistical modeling using generic computer components. The specification itself at paragraph [0247] confirms: "The novelty of the above-described embodiment lies not in the specific programming techniques but includes the use of the steps described to achieve the described results." This concession in the specification is instructive: the claim's novelty, as disclosed, lies in the ordered steps of the abstract analytical process, not in any improvement to computer technology, classifier technology, storage architecture, or any other technology. That is precisely the kind of claim the Alice/Mayo framework identifies as directed to an abstract idea without significantly more. Accordingly, Applicant's Argument 1 does not overcome the rejection, and the rejection is maintained.
The Applicant argues on pages 3-4 that “Claim 1 does not recite a mental process. The Office's analysis on pages 19 to 20 of the Office Action depends on disregarding the claim's concrete data-processing architecture and treating the claim as if it merely covered a person "manually determining price estimates," to quote the Office on page 19 of the Office Action. But the claim is not so broad. A human mind cannot practically receive and store live bookable price data at the claimed scale, maintain an incomplete historical travel-related price dataset lacking complete fare class information, train and store multiple structured classifiers from that dataset, communicate with a Distribution System, compare calculated estimates with received Distribution System fare prices, infer fare class availability, and periodically repeat the process to update stored inferred availability.
The application explains why the claimed environment is not a simple human pricing exercise. In scheduled airfare pricing, building a complete set of route/date fare data would require about 66,795 return queries for one route, about 185 times more queries than the budget model, and about 10^12 possible queries across commercial airports for a year, with correspondingly very large data storage, as described in the application as filed paras. [0086] to [088]. Claim 1 addresses that computer-scale data-processing problem, not a human mental calculation”.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the Applicant's Argument has been carefully considered but is not persuasive, for the reasons set forth below. While the Examiner acknowledges the governing principle confirmed by both MPEP 2106.04(a)(2)(III) and the August 4, 2025 Memorandum from the Deputy Commissioner for Patents that the mental process grouping is not without limits and that claim limitations that cannot practically be performed in the human mind do not fall within this grouping, Applicant's argument misapplies that principle to the specific limitations of claim 1.
Applicant's central contention is that the human mind cannot practically perform the claimed steps at the scale required specifically, processing data across approximately 10¹² possible route/date/airport combinations and maintaining an incomplete historical dataset of the corresponding volume. The Examiner agrees, as a factual matter, that no human could manually process data at that computational scale within a practical timeframe. However, the controlling legal question under the mental processes analysis is not whether a human could perform all the claimed steps simultaneously or at machine speed with real-world datasets; it is whether the claimed limitations, taken under their broadest reasonable interpretation, recite concepts that can practically be performed in the human mind i.e., whether they encompass operations of the kind performed through observation, evaluation, judgment, and opinion. Data volume and practical impracticability of manual execution due to scale are legally distinct from the question of whether the human mind is, as a categorical matter, equipped or unequipped to perform the type of operation recited. See Electric Power Group, LLC v. Alstom S.A., (claims directed to monitoring and analyzing power grid data an environment no human could monitor manually in real time were still found directed to a mental process abstract idea because the claimed analytical steps were recited at a level of generality encompassing mental observation and evaluation); CyberSource Corp. v. Retail Decisions, Inc., (claims ineligible even though the method involved processing large amounts of transaction data that would be impractical for a human to process manually). The Federal Circuit in CyberSource specifically recognized that the practical impracticability of manually performing steps with large datasets does not determine whether those steps fall within the mental processes grouping; what matters is whether the type of operation observation, comparison, inference is the kind of thing the human mind performs.
The examples that the USPTO has identified as falling outside the mental processes grouping because they cannot practically be performed in the human mind share a critical feature: those operations are categorically beyond what a human mind is equipped to do, not merely time-consuming or voluminous. GPS signal processing requires a human to simultaneously receive and decode satellite timing signals and compute pseudorange calculations across multiple satellites the human mind is not equipped to perform those signal-level operations at all, irrespective of quantity. See SiRF Tech., Inc. v. Int'l Trade Comm'n,. Network packet analysis requires a human to intercept and process live data packets traversing a network interface operations the human mind is not equipped to perform in any quantity. See SRI Int'l, Inc. v. Cisco Systems, Inc. A specific hardware-based RFID serial number data structure requires physically encoding data on a transponder a physical hardware operation. See ADASA Inc. v. Avery Dennison Corp. None of these examples is simply a human-performable operation applied to a very large dataset.
The operations recited in claim 1 are qualitatively different. Grouping historical price quotes by category (step (iv)(b)) is a categorization judgment that a human performs routinely, whether the dataset has ten entries or ten thousand. Deriving statistics for each group (step (iv)(c)) is a mathematical calculation that a human performs with pen, paper, and a calculator. Identifying groups corresponding to a requested price (step (iv)(e)) is a lookup and matching evaluation. Calculating estimated prices using statistics from stored classifiers (step (v)) is a mathematical computation specifically, computing Bayes posterior probabilities, as further specified in dependent claim 11 that is precisely the type of calculation courts have repeatedly found to be a mathematical concept/mental process even when performed computationally. See Parker v. Flook; Gottschalk v. Benson. Comparing calculated estimates with Distribution System fare prices (step (ix)) is a numerical comparison and inference precisely the kind of evaluation and judgment that constitutes a mental process. See In re Killian, (collecting information and "determining" or "making determinations" from that information constitutes a mental process because such steps "involve making determinations and identifications, which are mental tasks humans routinely do"). Storing and outputting results (steps (x)–(xi)) are data management functions. And periodically repeating these steps (step (xii)) is simply reapplication of the same abstract process, which does not change its character.
The fact that claim 1 recites these operations at the scale of 365 × 5,000 different requests (step (iii)) and with reference to a dataset that would require approximately 10¹² possible queries to complete does not move the core operations outside the mental processes grouping. The claim recites no limitation that describes a categorically human-mind-incapable operation of the kind identified in SiRF Tech., SRI, or ADASA. What it recites is that these human-mind-type operations grouping, deriving statistics, comparing, inferring are performed by a computer server and processors on a large dataset. A computer performing at scale, at speed, an operation that a human could perform on a smaller dataset is still performing that type of mental/mathematical operation. See MPEP 2106.04(a)(2)(III)(C) ("Claims can recite a mental process even if they are claimed as being performed on a computer"). The Supreme Court recognized this as long ago as Benson, finding that an algorithm for converting binary-coded decimal to binary was a mental process even though "the procedures can be carried out in existing computers long in use." Gottschalk v. Benson. The fact that the algorithm becomes practically useful only when implemented in a computer does not convert it into something other than a mental/mathematical process.
Applicant's reliance on the specification's description of the scale of the problem approximately 10¹² possible queries, 66,795 return queries per route, and 185 times more data than the budget model (paragraphs [0086]–[0088]) is therefore misplaced for Step 2A, Prong One purposes. That description establishes why a computer is useful and why a human could not practically complete the task in a reasonable timeframe; it does not establish that the type of operation recited statistical grouping, inference, and comparison is one the human mind is categorically unequipped to perform. The mental processes grouping turns on the nature of the operation, not the volume of data to which it is applied.
The Examiner additionally notes that, even if Applicant's argument were partially persuasive at Prong One with respect to some of the claim steps, the rejection is independently supportable on the mathematical concepts grouping. The Naïve Bayes classifier training and classification process recited in claim 1, steps (iv)(c)–(v), expressly encompasses mathematical calculations specifically, the derivation of statistics and the calculation of Bayes posterior probabilities, which are mathematical formulas or equations within the meaning of MPEP 2106.04(a)(2)(I). See also claim 11 (calculating Bayes posterior probabilities for each price range). Mathematical calculations do not become patent-eligible simply because they are applied to a large dataset or because performing them manually would be impractical. See Parker v. Flook; SAP Am., Inc. v. InvestPic, LLC, ("No matter how much of an advance in the finance field the claims represent, the advance lies entirely in the realm of abstract ideas."). The Examiner thus maintains the Step 2A, Prong One finding that claim 1 recites a judicial exception, and Applicant's Argument 2 does not overcome the rejection. The rejection is therefore maintained.
The Applicant argues on pages 4-6 that “Even assuming arguendo that claim 1 recites an abstract idea, claim 1 integrates it into a practical application. On pages 3, 4, 13 and 14 of the Office Action, the Office characterizes the reduced dataset as merely "storing less data." Applicant respectfully disagrees. Claim 1 does not merely decide to omit data. The claim recites a particular computer-implemented architecture that preserves technical utility despite incomplete stored information: the system stores an incomplete live-bookable-price dataset lacking complete fare class information, trains structured classifiers from that incomplete dataset, uses those classifiers to estimate prices, compares the estimates with Distribution System fare prices, and infers and stores fare class availability. This is a specific technological solution to the specification's identified technical problem, identified in paras. [0009] and [0089] of the application as filed, of excessive GDS querying and excessive storage for complete fare availability/price data.
The Office's analogy to "serving fewer users", on pages 4 and 14 of the Office Action, is therefore inapt. Claim 1 does not reduce storage by reducing functionality. It recovers missing fare-class availability information through a recited classifier-based data structure and comparison workflow. The stored classifiers are not generic labels; in paras. [0104], [0110], [0111], [0116], and [0119] of the application as filed, the specification describes classifiers trained by category, using historical quotes, weighted features, price ranges, and periodic retraining.
This is materially closer to Enfish (Enfish, LLC V. Microsoft Corp.) and McRO (McRO, Inc. v. Bandai Namco Games America Inc., Court of Appeals, Federal Circuit 2016) than to the cases relied on by the Office. Like Enfish, claim 1 improves a computer data architecture for storing and retrieving useful information from a specifically structured dataset. Like McRO, claim 1 recites a particular rules/classifier-based process that achieves a result not by claiming the result itself, but by requiring a specific ordered technique.
The claim is unlike CyberSource, cited on page 8 of the Office Action, which involved fraud-detection logic that could be performed mentally. Claim 1 requires a server-based workflow involving incomplete live-bookable-price datasets, stored classifiers, Distribution System communications, and periodic retraining/storage. Claim 1 is unlike Electric Power Group, cited on page 10 of the Office Action, where claims merely collected, analyzed, and displayed information without a specific technological means. Claim 1 recites the specific means by which incomplete stored travel-pricing data is transformed into stored inferred fare-class availability. Claim 1 is unlike SAP V. InvestPic, cited on pages 10 and 16 of the Office Action, where the alleged advance resided only in statistical analysis of financial information; here, the ordered combination uses structured classifier storage and Distribution System comparison to reduce computer storage/querying burden while producing stored fare-class availability.
The 2024 USPTO Artificial Intelligence Subject Matter Eligibility Update supports our analysis. On slides 5, 42, 66 and 99, the Update states that claims need not mirror the examples to be eligible, and that eligibility turns on whether the claim integrates any abstract idea into a practical application. In Example 48, regarding its Claims 2 and 3, the USPTO found eligibility where otherwise abstract DNN operations were integrated into a practical speech-processing application by reciting how the processing enabled separated signals and transcription. Likewise, claim 1 integrates any statistical/classifier operations into a practical travel-pricing data- processing application by reciting how the classifiers trained on incomplete data are used with Distribution System fare prices to infer and store fare-class availability”.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the argument has been carefully considered but remains unpersuasive, for the following reasons. The Examiner addresses each of Applicant's contentions in turn:
Regarding Applicant's Characterization of the "Storing Less Data" Argument - The Examiner acknowledges Applicant's point that the prior characterization of the claimed storage reduction as "merely storing less data" was imprecise, and the Examiner does not rely on that formulation standing alone as the basis for the rejection. The Examiner agrees that claim 1 recites more than a bare decision to collect fewer records. The claim does describe a workflow in which an incomplete dataset is used as the input for classifier training, and those trained classifiers are subsequently used in conjunction with a Distribution System comparison to infer fare class availability. The Examiner has considered this full workflow in evaluating Step 2A, Prong Two. The rejection is accordingly maintained on the grounds stated below, not on the ground that the claim merely stores less data.
Regarding Whether the Claim Reflects a Technological Improvement Sufficient to Integrate the Abstract Idea Into a Practical Application - The controlling framework under MPEP 2106.04(d)(1) and 2106.05(a), as updated by the December 5, 2025 Memorandum from the Deputy Commissioner for Patents incorporating Ex Parte Desjardins (precedential), requires a two-part inquiry: first, whether the specification provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement in the functioning of a computer or improvement to another technology or technical field; and second, whether the claim itself reflects that disclosed improvement. See also Intellectual Ventures I LLC v. Symantec Corp., (claims must have limitations that address the asserted improvement; it is insufficient that the specification describes an improvement if the claims themselves do not reflect it).
Critically, it is established that an improvement in the abstract idea itself is not an improvement in the technology. See MPEP 2106.05(a)(II); In re Board of Trustees of Leland Stanford Junior University (Stanford I), (a claim yielding "a greater number of haplotype phase predictions" was not an improved technological process, but an improved mathematical process, and thus remained directed to an abstract idea); In re Board of Trustees of Leland Stanford Junior University (Stanford II), (Fed. Cir. 2021) (improvement in "the accuracy of a mathematically calculated statistical prediction" is an improvement to the abstract idea, not to another technology). The Examiner finds that the alleged improvement in claim 1 inferring fare class availability with reduced GDS querying and reduced data storage is precisely an improvement to the accuracy and efficiency of the abstract analytical process itself, not an improvement to computer technology, classifier technology, storage architecture, or any other technology as such.
The specification at paragraphs [0009] and [0089], which Applicant cites as identifying the technical problem, states: "an advantage is that the method does not require the time or energy required to access one or more remote servers many times to actually establish fare availability and to calculate fare prices" and "the massive increase in the number of queries needed to form a 'complete picture' of a scheduled model route combined with the costs of GDSs means that it is not practical (i.e. it is disadvantageous) to build such a cache." Critically, the specification frames the problem and its solution in commercial and operational terms reducing GDS query costs and avoiding the expense and time of maintaining a complete dataset not in terms of improving how computers store data, how classifiers function as technology, or how any computer system operates. The specification at paragraph [0247] further confirms: "The novelty of the above-described embodiment lies not in the specific programming techniques but includes the use of the steps described to achieve the described results." This statement in the specification is a direct concession that the claimed improvement is in the steps of the analytical process i.e., the abstract idea not in any underlying computer or technological improvement. When the specification itself disclaims that the novelty lies in specific programming techniques, it forecloses the argument that the claim reflects an improvement to computer technology.
Regarding Applicant's Enfish Analogy - Applicant argues that claim 1 is analogous to Enfish because it improves a computer data architecture for storing and retrieving useful information from a specifically structured dataset. The Examiner disagrees. In Enfish, the claims were directed to a specific self-referential table data structure that concretely changed how a computer organized and accessed data in memory the self-referential structure was itself the technological innovation, enabling faster searching and more efficient data organization as compared to conventional relational database structures. The improvement was to the mechanics of how the computer stored and retrieved data; the self-referential table was not a conventional data structure applied to a new problem, but a new data structure architecture. The specification in Enfish described in technical detail how the self-referential table operated differently from conventional relational databases, and the claims reflected that specific structural innovation.
Claim 1 here does not recite any specific data structure architecture that changes how a computer stores or retrieves data. The "incomplete historical travel-related price dataset" is a conventional dataset that happens to contain fewer records because certain data (complete fare class information) has not been collected. The classifiers are Naïve Bayes classifiers a well-established statistical method trained in a conventional manner on the available data. There is no new data structure, no new storage architecture, and no new retrieval mechanism described in the specification or recited in the claim. The claim differs from Enfish because in Enfish, the data structure itself was the improvement to computer functionality; here, the data management decisions are inputs to and outputs from an abstract analytical process, not innovations in computer data architecture. The Enfish analogy is therefore inapt.
Regarding Applicant's McRO Analogy - Applicant argues that claim 1 is analogous to McRO because it recites a particular rules/classifier-based process that achieves a result through a specific ordered technique rather than merely claiming the result. The Examiner disagrees. In McRO, the claims recited a specific set of rules specifically defined morph weight functions and transition rules that automated a previously manual animation process in a technically superior way that prior automated methods had not achieved. The specific rule structure was the technical innovation, and the Federal Circuit found that the claims were limited to that particular rule-based technique for achieving the animation result, rather than broadly claiming all possible ways of achieving lip synchronization. The specification in McRO explained in technical detail how the particular rules enabled automation of tasks that previously could only be performed subjectively by human animators.
Claim 1 does not recite a specific rule structure or specific classifier architecture that constitutes a technical innovation in classifier design or machine learning methodology. The claim recites training Naïve Bayes classifiers using historical price data grouped by category, deriving statistics, and periodically retraining all of which are general descriptions of a well-established statistical methodology applied to travel pricing data. Unlike the specific morph weight transition rules in McRO, the Naïve Bayes approach recited in claim 1 is not a novel rule structure defined in the claims; it is a known statistical method invoked generically and applied to a new domain. The claim does not specify, at the level of technical particularity present in McRO, a specific technical technique that was not previously available or that constituted an innovation in classifier technology or machine learning methodology. The McRO analogy is therefore inapplicable.
Regarding Applicant's Distinctions of CyberSource, Electric Power Group, and SAP v. InvestPic - Applicant argues that claim 1 is distinguishable from these cases because it recites specific means by which incomplete stored travel-pricing data is transformed into stored inferred fare-class availability, whereas those cases lacked specific technological means. The Examiner acknowledges that claim 1 recites more operational detail than the claims in some of those cases. However, the presence of additional operational steps does not, by itself, distinguish the claim from those authorities; what matters is whether the additional steps represent a technological solution to a technological problem or merely implement an abstract analytical process using generic computer components. As established above, the specification itself frames the problem and solution in commercial/operational terms, not technological improvement terms. Claim 1 recites steps that perform an abstract analytical and inferential process grouping, statistical derivation, Naïve Bayes classification, comparison, and inference using generic computer components. The specific means recited are specific steps in the abstract process, not specific technological innovations in computer or classifier technology. The distinctions Applicant draws therefore do not overcome the rejection.
Regarding Applicant's Reliance on Example 48 of the 2024 AI Subject Matter Eligibility Update - Applicant argues that, like Example 48 Claims 2 and 3, claim 1 integrates statistical/classifier operations into a practical travel-pricing data-processing application. The Examiner finds this analogy unpersuasive. In Example 48, Claims 2 and 3 were found eligible because the claims reflected the specific technological improvement described in the disclosure particularly, the steps of synthesizing speech waveforms from masked clusters and combining the speech waveforms to generate a mixed speech signal excluding audio from an undesired source (Claim 2), and extracting spectral features from separated speech signals and generating a transcript (Claim 3). These additional steps were not merely instructions to apply the DNN to speech signals; they were concrete technical operations signal synthesis, waveform combination, spectral feature extraction, and transcription generation that reflected the disclosed technical improvement to speech separation and speech-to-text technology and that are operations the human mind is categorically unequipped to perform in the same manner. The claims in Example 48 integrated the abstract idea into a practical application by reciting steps that went beyond applying the DNN generically, to include specific signal processing and output generation operations that embodied the technical advance. For all of the foregoing reasons, Applicant's Argument 3 is not persuasive, and the rejection is therefore maintained.
The Applicant argues on pages 6-7 that “Claim 1 also recites significantly more than any alleged abstract idea. The ordered combination is not routine "apply it" computerization. It requires a specific sequence: incomplete live-bookable-price storage; classifier training and storage; estimate calculation from identified classifier groups; Distribution System querying; comparison of calculated estimates with Distribution System prices; inference and storage of fare-class availability; and periodic retraining and re-storage.
The Office's Step 2B analysis on pages 23 to 25 of the Office Action treats each limitation separately and dismisses receiving, storing, training, comparing, outputting, and periodic repetition as generic computer functions. That approach fails to consider the claim as an ordered combination. The claimed combination solves the concrete technical problem identified in the specification: avoiding the impractical storage and query burden of a complete fare availability/pricing cache while still inferring useful fare-class availability.
For at least these reasons, claim 1 is not directed to a judicial exception, or alternatively integrates any alleged exception into a practical application and recites significantly more than the alleged exception. Applicant respectfully requests withdrawal of the 101 rejection of claim 1.
Independent Claim 28 is patent eligible, for reasons similar to the reasons that Claim 1 is patent eligible. The dependent Claims are patent eligible, at least by virtue of their dependence on a respective independent Claim”. The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the argument has been carefully considered but remains unpersuasive for the reasons set forth below. The Examiner addresses both the procedural argument regarding the ordered combination and the substantive argument regarding significantly more.
Regarding the Ordered Combination Analysis
The Examiner acknowledges the governing requirement that additional elements must be evaluated both individually and in combination at Step 2B. See MPEP 2106.05; Alice Corp. v. CLS Bank Int'l, ("it is consistent with the general rule that patent claims 'must be considered as a whole'"); BASCOM Global Internet Servs. v. AT&T Mobility LLC, (inventive concept may be found in the non-conventional and non-generic arrangement of elements that are individually well-known). The Examiner confirms that the ordered combination of claim 1 has been evaluated as such, and the rejection is maintained on that basis. The question at Step 2B is not merely whether each individual element is routine or conventional, but whether the combination of additional elements considered together provides an inventive concept that is more than the judicial exception. See Diamond v. Diehr. The Examiner has performed that combined evaluation and finds, for the reasons detailed below, that the ordered combination does not provide significantly more.
Regarding Whether the Ordered Combination Provides Significantly More
The specific sequence Applicant identifies incomplete live-bookable-price storage; classifier training and storage; estimate calculation from identified classifier groups; Distribution System querying; comparison of calculated estimates with Distribution System prices; inference and storage of fare-class availability; and periodic retraining and re-storage has been evaluated as an ordered combination. The Examiner finds that this combination, while operationally coherent and commercially useful, does not constitute a non-conventional and non-generic arrangement of the additional elements that amounts to significantly more than the identified abstract idea.
The key distinction that BASCOM established for when a combination of individually conventional elements amounts to an inventive concept is that the combination must represent a non-conventional and non-generic arrangement that provides a technical improvement in the art. In BASCOM, the eligibility-conferring feature was the specific and unconventional placement of a filtering tool at a particular network location the remote ISP server with individually customizable features specific to each end user. That specific architectural arrangement was not the conventional way to implement Internet content filtering; it was a technically unconventional configuration that produced a concrete technical improvement over prior filtering systems. See BASCOM. Similarly, in DDR Holdings, the claims were eligible because they specified how interactions with the Internet were manipulated to yield a result that overrode the routine and conventional sequence of events ordinarily triggered by the click of a hyperlink a result that could not be achieved using the Internet in its conventional manner. DDR Holdings, LLC v. Hotels.com, L.P.
The ordered combination of claim 1 does not present a comparable non-conventional and non-generic arrangement. The claim sequences the following operations: a computer server receives price data and stores it in a non-transitory storage medium; processors group the data, derive statistics, and train Naïve Bayes classifiers; other processors calculate estimates using those classifiers; the server queries a Distribution System and receives a response; the server compares the calculated estimates to the received prices and infers fare class availability; the server outputs the result to a computing device, which stores it; and the entire sequence is periodically repeated. Each step in this sequence is performed in the conventional manner one would expect for that type of operation: data is received by a server and stored conventionally; classifiers are trained using conventional Naïve Bayes methodology as described in the specification at paragraphs [0121]–[0143] and well-established in the art; a Distribution System is queried in the conventional manner for travel pricing data (as the specification at paragraphs [0003] and [0181]–[0188] confirms is the established industry practice); a comparison is made; and results are stored. The sequencing of these steps in this order does not represent an unconventional or technically innovative arrangement; it represents the natural logical sequence one would follow to implement the described business workflow on generic computing hardware. Unlike BASCOM's unconventional placement of the filtering tool, there is nothing in the arrangement of claim 1's steps that departs from or overrides the conventional sequence of operations one would ordinarily use to implement such a workflow.
Applicant contends that the combination "solves the concrete technical problem identified in the specification: avoiding the impractical storage and query burden of a complete fare availability/pricing cache while still inferring useful fare-class availability." The Examiner acknowledges that the combination produces this operationally useful result. However, as established in the analysis of Argument 3 above and confirmed by the guidance that "an improvement in the abstract idea itself is not an improvement in the technology" (MPEP 2106.05(a)(II); Stanford I), the problem solved by the combination is a commercial and operational problem the cost and volume of GDS queries and data storage not a technical problem with how computer technology functions. The specification at paragraph [0089] frames this explicitly in economic terms: the incomplete cache exists because "it is not practical (i.e., it is disadvantageous) to build such a cache" due to the cost of GDS queries, not because of any technical limitation in computer storage or processing capability. A combination of conventional steps that, taken together, produces a commercially useful result by applying an abstract analytical process more efficiently does not constitute an inventive concept under Step 2B when the result is an improvement to the analytical process itself rather than to any underlying technology. See In re Stanford I; SAP Am., Inc. v. InvestPic, LLC.
The Examiner further notes that the periodic retraining step (step (xii)) which Applicant has emphasized throughout does not alter this analysis. As established with respect to earlier arguments, periodically reapplying a conventional statistical training process to updated data is itself a well-understood, routine, and conventional practice in the data science and machine learning fields, and is explicitly supported as WURC by the specification's own statement at paragraph [0247] that "the above-described steps can be implemented using standard well-known programming techniques." See also Bancorp Services, LLC v. Sun Life Assurance Co. of Canada, (performing repetitive calculations is well-understood, routine, conventional). Periodic repetition of an abstract process to update stored results does not, by itself or in combination, transform a conventionally implemented abstract process into patent-eligible subject matter. See Intellectual Ventures I LLC v. Capital One Bank (USA).
Regarding Independent Claim 28 and the Dependent Claims
Applicant asserts that independent claim 28 is patent eligible for reasons similar to those articulated for claim 1, and that the dependent claims are patent eligible at least by virtue of their dependence on an independent claim. The Examiner has considered these arguments but finds them unpersuasive. Claim 28 recites the same abstract idea in system claim format, and for all the same reasons articulated with respect to claim 1, claim 28 does not integrate the abstract idea into a practical application and does not recite significantly more than the abstract idea. The system format of the claim does not alter the eligibility analysis; the computer server, computing device, and data store recited in claim 28 are generic computing components performing the same abstract analytical functions analyzed above. See Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1333 (Fed. Cir. 2015) (system claims directed to abstract idea on generic hardware are no more eligible than corresponding method claims). With respect to the dependent claims, the Examiner has considered each in the dependent claims analysis set forth in the Office Action and the current Office Action, and finds that each adds limitations that further narrow the abstract idea without providing additional elements that integrate the judicial exception into a practical application or amount to significantly more. Dependence on an independent claim cannot, by itself, confer eligibility on a dependent claim where the independent claim is ineligible and the dependent claim adds no qualifying additional elements. See MPEP 2106 (each claim must be evaluated individually for eligibility). For all of the foregoing reasons, Applicant's Argument 4 is not persuasive. The rejection of claims 1–30 under 35 U.S.C. 101 is maintained.
The remaining Applicant's arguments filed 30 September 2026 have been fully considered but they are considered moot in view of new grounds of rejection.
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 1–30 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The independent claims recite an abstract idea in the form of a mental process specifically, the concept of collecting, categorizing, and statistically analyzing historical travel-related price data using Naïve Bayes classifiers; comparing the resulting price estimates against Distribution System fare prices; and inferring fare class availability from that comparison together with certain methods of organizing commercial activity in the travel industry. This judicial exception is not integrated into a practical application because the additional elements recited in the claims a computer server, one or more processors, a non-transitory storage medium, a computing device, and a Distribution System communication workflow are generic computer components performing generic computer functions at a high level of generality, which amount to no more than mere instructions to apply the abstract idea using conventional computer technology, and do not impose any meaningful technological limit on the abstract idea. The specification itself confirms at paragraph [0247] that "the novelty of the above-described embodiment lies not in the specific programming techniques but includes the use of the steps described to achieve the described results," establishing that the claimed improvement resides in the abstract analytical process rather than in any computer technology. The claims do not include additional elements sufficient to amount to significantly more than the judicial exception because the additional elements, individually and as an ordered combination, represent routine computer implementation of the abstract process and insignificant extra-solution activity, and do not improve the functioning of a computer or any other technology or technical field.
STEP 1
Regarding Step 1 of the Subject Matter Eligibility Test for Products and Processes (from the January 2019 101 Examination Guidelines, as updated), claims 1–27 are directed to a method, which constitutes a process a statutory category of invention under 35 U.S.C. 101. Claims 28–30 are directed to a system comprising a computing device, a computer data store, and a server, which constitutes a machine also a statutory category of invention under 35 U.S.C. 101. Accordingly, all pending claims 1–30 fall within at least one of the four statutory categories.
STEP 2A, Prong One:
The independent claims (claims 1 and 28) recite a judicial exception. The Examiner identifies the abstract idea as: the concept of (a) collecting and storing an incomplete historical dataset of observable, live bookable travel-related prices that lacks complete fare class information; (b) grouping historical price quotes by category, deriving statistics for each group, and training probabilistic Naïve Bayes classifiers from those statistics to estimate unobserved prices; (c) using the trained classifiers to calculate estimated fare prices for a requested journey; (d) comparing those estimated prices against fare prices received from a Distribution System; (e) inferring fare class availability from that comparison; and (f) periodically repeating this process to update the stored inferred availability. This abstract idea falls within the mental processes grouping, and the classifier-training and posterior probability calculation aspects additionally fall within the mathematical concepts grouping, of abstract ideas. See MPEP 2106.04(a)(2), subsections I and III. Where limitations fall within the same or overlapping groupings, they are considered together as a single abstract idea. See MPEP 2106.04, subsection II.B.
Identification of Specific Claim Limitations Reciting the Abstract Idea
The following limitations of independent claim 1 recite the identified abstract idea:
Step (i): "receiving observable, live bookable prices for travel-related services, including price information, wherein the price information does not include complete fare class information" a data observation and collection step that describes the conceptual act of gathering an incomplete set of price data;
Step (ii): "storing the received observable, live bookable prices… to provide a stored incomplete historical travel-related price dataset which does not include complete fare class information… wherein the stored incomplete historical travel-related price dataset uses a smaller data storage capacity than a complete historical travel-related price dataset" describes the conceptual decision to store an incomplete rather than complete dataset, which is a judgment or organizational choice that can be performed mentally;
Step (iv)(a): "obtaining historical price quotes from the incomplete historical travel-related price dataset" a data retrieval/observation step;
Step (iv)(b): "grouping the historical price quotes by category" an evaluative categorization step constituting a mental process of organization and classification;
Step (iv)(c): "deriving statistics for each group" a mathematical calculation (statistical derivation) from grouped data;
Step (iv)(d): "storing on a computer for each group a plurality of classifiers comprising structured data including the derived statistics, including training the classifiers using the historical price quotes" describes training statistical (Naïve Bayes) classifiers from derived statistics, which is a mathematical/statistical modeling process;
Step (iv)(e): "identifying groups with stored classifiers to which the requested prices correspond" a matching and identification evaluation constituting a mental process of lookup and comparison;
Step (v): "calculate estimates for the requested prices… calculating a set of estimates, for the requested prices on the particular date using statistics from the stored classifiers corresponding to the identified groups" a mathematical calculation of price estimates from statistical classifier outputs;
Step (ix): "the computer server comparing the calculated estimates for the requested fare prices from step (v) with the Distribution System's fare prices received in step (viii) to infer which fare classes are available" a numerical comparison and logical inference constituting a mental process of observation and evaluation; and
Step (xii): "repeating steps (i) and (ii) and (iv) to (xi), including training the classifiers periodically… to infer the fare class availability… periodically, and to store the inferred fare class availability periodically" periodic reapplication of the same abstract analytical and inferential process.
Mental Processes Grouping Analysis (MPEP 2106.04(a)(2)(III))
The above limitations, under their broadest reasonable interpretation, cover concepts that can practically be performed in the human mind including through observation, evaluation, judgment, and comparison with or without pen and paper, but for the recitation of generic computer components. See MPEP 2106.04(a)(2)(III); CyberSource Corp. v. Retail Decisions, Inc.
The act of observing and collecting price information (step (i)) is a classic mental observation. The decision to maintain an incomplete rather than complete dataset (step (ii)) is a conceptual organizational judgment. The grouping of price quotes by category (step (iv)(b)) is an evaluative classification that a person could perform on a smaller dataset with pen and paper. Deriving statistics for each group (step (iv)(c)) is a mathematical calculation. Identifying groups corresponding to a requested price (step (iv)(e)) is a lookup and matching evaluation. Computing estimated prices using statistics from trained classifiers (step (v)) is a mathematical calculation specifically, computing Naïve Bayes posterior probabilities as further described in dependent claim 11 of the type courts have consistently treated as mathematical/mental processes. See Parker v. Flook; Gottschalk v. Benson. Comparing calculated estimates against Distribution System prices to infer fare class availability (step (ix)) is an observation, comparison, and evaluation a classic mental process. See In re Killian. Periodic repetition of these steps (step (xii)) is simply iterative reapplication of the same abstract process.
The Examiner acknowledges the guidance in MPEP 2106.04(a)(2)(III) and the August 4, 2025 Memorandum from the Deputy Commissioner for Patents that the mental processes grouping is not without limits, and that claim limitations that cannot practically be performed in the human mind i.e., operations that the human mind is categorically unequipped to perform (such as GPS signal processing, network packet analysis, or specific data encryption operations) do not fall within this grouping. The Examiner has carefully evaluated each limitation of claim 1 in light of this guidance and confirms that the claimed operations grouping, statistical derivation, probabilistic classification, comparison, and inference are qualitatively operations of the kind the human mind performs routinely. While performing these operations on the claimed scale (up to 365 × 5,000 different route/date requests, per step (iii)) would be practically impractical for a human due to data volume, practical impracticability due to scale is legally distinct from the question of whether the type of operation is one the human mind is categorically unequipped to perform. See Electric Power Group, LLC v. Alstom S.A., (claims directed to monitoring and analyzing complex power grid data remained abstract despite being practically unperformable manually at scale); CyberSource, (large data volumes do not take claims out of the mental processes grouping when the type of operation is mental).
Mathematical Concepts Grouping Analysis (MPEP 2106.04(a)(2)(I))
Additionally, steps (iv)(c), (iv)(d), and (v) expressly recite mathematical concepts specifically, mathematical calculations (deriving statistics; computing Naïve Bayes posterior probabilities) and mathematical relationships (probabilistic models of price ranges as a function of feature values). These fall within the mathematical concepts grouping of abstract ideas. See MPEP 2106.04(a)(2)(I); Parker v. Flook; SAP Am., Inc. v. InvestPic, LLC. An improvement in the mathematical process itself such as improved prediction accuracy from a better statistical model is an improvement to the abstract idea, not to technology. See In re Board of Trustees of Leland Stanford Junior University (Stanford II). The limitations reciting the Naïve Bayes classification approach (reflected in steps (iv)(c)–(v) and further specified in dependent claims 5, 6, 8, 9, 11, and 27) are therefore also abstract ideas on this basis.
Accordingly, the independent claims recite a judicial exception.
STEP 2A, Prong Two:
The Examiner has evaluated the additional elements in the claims those beyond the identified abstract idea both individually and in combination, to determine whether the claim as a whole integrates the judicial exception into a practical application. See MPEP 2106.04(d), 2106.05(a)–(c), 2106.05(e)–(h). Step 2A Prong Two specifically excludes consideration of whether the additional elements are well-understood, routine, or conventional; that analysis is reserved for Step 2B. See MPEP 2106.04(d).
Identification of Additional Elements Beyond the Abstract Idea
The following elements in independent claim 1 are additional elements beyond the identified abstract idea:
"a computer server" performing steps (iii), (vi), (vii), (viii), (ix), and (x) receiving price requests, providing calculated price estimates, querying the Distribution System, receiving Distribution System fare prices, comparing estimates with Distribution System prices, and outputting inferred fare class availability;
"one or more first computer processors" configured to perform step (iv) the classifier training and statistical derivation steps;
"one or more second computer processors" configured to perform step (v) the estimate calculation step;
"a non-transitory storage medium" on which the incomplete historical travel-related price dataset is embodied (steps (ii) and (iv));
"a computing device" to which the inferred fare class availability is output (step (x)) and which stores the inferred fare class availability (step (xi));
"a Distribution System" to which the server sends a request for fare prices and from which it receives fare prices (steps (vii) and (viii)); and
Providing calculated price estimates (step (vi)), outputting inferred fare class availability to the computing device (step (x)), and storing the inferred fare class availability (step (xi)).
The additional elements in independent claim 28 are materially identical, recited in system claim format.
Analysis of Additional Elements
Improvement to Technology or Technical Field (MPEP 2106.05(a))
The claim does not recite an improvement to the functioning of a computer or to any other technology or technical field. Under MPEP 2106.04(d)(1) and 2106.05(a), as updated by the December 5, 2025 Memorandum from the Deputy Commissioner for Patents incorporating Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB) (precedential), the improvement consideration requires: first, that the specification provide sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement in the functioning of a computer or another technology; and second, that the claim itself reflect that disclosed improvement.
Critically, an improvement in the abstract idea itself is not an improvement in the technology. See MPEP 2106.05(a)(II); Stanford II, (improvement in "the accuracy of a mathematically calculated statistical prediction" is an improvement to the abstract idea, not to another technology); Trading Technologies Int'l v. IBG, (user interface that provided a trader with more information improved the business process of trading but did not improve computers or technology).
The specification at paragraphs [0009] and [0089] frames the problem and solution in commercial and operational terms: "the massive increase in the number of queries needed to form a 'complete picture' of a scheduled model route combined with the costs of GDSs means that it is not practical (i.e. it is disadvantageous) to build such a cache" (para. [0089], emphasis added). The driver of the incomplete dataset is GDS query cost an economic constraint not a technical limitation in computer storage or processing capability. The specification does not describe a technical problem with how computers store data, how classifiers function as technology, or how any computer system operates. The specification at paragraph [0247] confirms: "The novelty of the above-described embodiment lies not in the specific programming techniques but includes the use of the steps described to achieve the described results." This concession establishes that the specification itself does not present an improvement to computer technology only an improvement to the analytical workflow.
Unlike Enfish, LLC v. Microsoft Corp., where the claims recited a specific self-referential table data structure that concretely changed how a computer organized and accessed data in memory, claim 1 does not recite any specific data structure architecture that improves computer storage or retrieval operations. The "incomplete historical travel-related price dataset" is a conventional dataset with fewer records. Unlike McRO, Inc. v. Bandai Namco Games Am. Inc., where specific rule structures were the technical innovation enabling automation of animation tasks, claim 1 does not recite a specific classifier architecture or rule structure that constitutes a technical innovation in machine learning technology. Unlike Ex Parte Desjardins, where the improvement was to how the machine learning model itself operated specifically addressing the technical problem of "catastrophic forgetting" in continual learning the claim here does not describe an improvement to how classifiers or machine learning technology operates, but rather applies known Naïve Bayes methods to a new commercial domain. The claimed Naïve Bayes classification is a well-established statistical method; applying it to travel pricing does not improve the technology of machine learning.
Claim 1 is instead more analogous to Example 49, Claim 1 of the 2024 USPTO AI Subject Matter Eligibility Update Examples, which was found ineligible because the claim recited using a deep neural network at a high level of generality "without reflecting the improvement discussed in the disclosure," and thus the "recited generic DNN merely adds a generic computer component to perform the method and therefore fails to provide an improvement to the technology or technical field." Similarly, claim 1 recites training Naïve Bayes classifiers and periodically retraining them without reciting specific technical details about how the classifiers are structured or implemented in a manner that improves classifier technology, data storage technology, or any other technology. The improvement consideration does not weigh in favor of eligibility. See MPEP 2106.05(a).
Particular Machine (MPEP 2106.05(b))
The claims do not recite use of a particular machine that imposes meaningful limits on the claim. The "computer server," "one or more first computer processors," "one or more second computer processors," "non-transitory storage medium," and "computing device" are generic computing components recited at a high level of generality. The specification at paragraph [0247] confirms they are "standard well-known" components. The "Distribution System" is invoked only as an external information source to which a request is sent and from which prices are received, recited at a high level of generality without any particular technical configuration that would impose a meaningful limit. The particular machine consideration does not weigh in favor of eligibility. See MPEP 2106.05(b).
Mere Instructions to Apply the Exception (MPEP 2106.05(f))
The additional elements amount to no more than mere instructions to implement the abstract idea on a computer. The claim instructs that the abstract analytical process of collecting price data, training statistical classifiers, computing estimates, querying a Distribution System, comparing results, and inferring and storing fare class availability shall be performed on generic computer hardware. This is the paradigmatic "apply it" scenario: "simply stating the abstract idea while adding the words 'apply it.'" Alice Corp. v. CLS Bank Int'l. The Examiner is mindful of the caution in the August 4, 2025 Memorandum against oversimplifying claim limitations and expanding the "apply it" consideration; however, after careful evaluation of each limitation, the Examiner finds that the recited computer components individually and in combination do not go beyond generic computer implementation of the identified abstract analytical process. The "apply it" consideration weighs against eligibility. See MPEP 2106.05(f).
Insignificant Extra-Solution Activity (MPEP 2106.05(g))
The steps of receiving price data (step (i)), providing calculated price estimates (step (vi)), outputting inferred fare class availability to a computing device (step (x)), and storing that inferred availability (step (xi)) constitute insignificant extra-solution activity. Specifically, step (i) is mere pre-solution data gathering; steps (vi), (x), and (xi) are post-solution data transmission, output, and storage steps that convey or record the result of the abstract analytical process. These are necessary incidental steps that do not add a meaningful technological limit to the claim. See MPEP 2106.05(g); Electric Power Group, ("Collecting information, analyzing it, and displaying certain results" does not integrate an abstract idea into a practical application). The insignificant extra-solution activity consideration weighs against eligibility.
Considering all additional elements individually and in combination, the claim as a whole does not integrate the judicial exception into a practical application. The additional elements amount to: (1) generic computer implementation of the abstract analytical process ("apply it"); (2) insignificant extra-solution data gathering and output activity; and (3) general linking of the abstract idea to the commercial travel pricing field of use. None of these, individually or in combination, impose a meaningful technological limit on the practice of the abstract idea or reflect a technological improvement to a computer or other technology. Accordingly, the claims are directed to a judicial exception.
STEP 2B
As discussed with respect to Step 2A Prong Two, the additional elements in the claims amount to no more than mere instructions to apply the judicial exception using generic computer components and insignificant extra-solution activity. The same analysis applies in Step 2B: mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. See MPEP 2106.05(f); Alice Corp.
The Examiner has re-evaluated the additional elements as an ordered combination, as required by the USPTO's subject matter eligibility framework. See MPEP 2106.05; Diamond v. Diehr, ("A new combination of steps in a process may be patentable even though all the constituents of the combination were well known and in common use before the combination was made."); BASCOM Global Internet Servs. v. AT&T Mobility LLC, (inventive concept may be found in a non-conventional and non-generic arrangement of elements that are individually well-known). The question is therefore whether the ordered combination of additional elements represents a non-conventional and non-generic arrangement that provides a technical improvement, or whether the combination is merely the expected sequential implementation of the abstract process on generic computing hardware.
The Examiner finds that the ordered combination does not constitute a non-conventional and non-generic arrangement. The sequence a computer server receiving data and storing it conventionally; processors training classifiers using established Naïve Bayes methodology; processors calculating estimates; the server querying a Distribution System in the conventional manner; the server comparing estimates to received prices; the server outputting results to a computing device; the computing device storing results; and the sequence periodically repeating describes the natural logical workflow one would ordinarily implement when applying these known methods on generic computing hardware. Each step is performed in the conventional manner expected for that type of operation. Unlike BASCOM, where the specific and unconventional placement of the content filter at the remote ISP server with per-user customizability was a technically unconventional network architecture, and unlike DDR Holdings, LLC v. Hotels.com, L.P., where the claim specified how Internet interactions were manipulated to override the conventional sequence of events triggered by a hyperlink, claim 1 does not present any technically unconventional arrangement of its computing components. The combination implements the identified abstract process in the expected, conventional sequence on generic computing hardware. Moreover, as noted above, the specification at paragraph [0247] expressly confirms that the novelty of the invention does not reside in the programming techniques i.e., not in any arrangement of computer components but in the analytical steps themselves, which are the abstract idea. Accordingly, the ordered combination does not provide significantly more than the abstract idea.
Furthermore, as established in the analysis of the practical application consideration above, the problem the combination is designed to address the commercial cost and data volume burden of maintaining a complete fare availability/pricing cache is an economic and operational problem, not a technical problem with the operation of any computer or other technology. A combination of conventional computer-implemented steps that produces a commercially useful result by applying an abstract process more efficiently does not constitute an inventive concept when the result is an improvement to the efficiency of the abstract process rather than to any underlying technology. See Stanford II,; SAP Am., Inc. v. InvestPic, LLC; Recentive Analytics, Inc. v. Fox Corp., (steps incidental to automating an abstract idea are not sufficient to confer eligibility even when viewed in combination).
Well-Understood, Routine, and Conventional Activity Analysis (MPEP 2106.05(d))
The Examiner additionally finds, with evidentiary support as set forth below, that the individual additional elements of the claims are well-understood, routine, and conventional activities in the relevant field:
"Receiving observable live bookable prices for travel-related services over a network / from external sources" The courts have recognized that receiving or transmitting data over a network or from external data sources is well-understood, routine, conventional activity. See Symantec Corp.; TLI Communications LLC v. AV Auto. LLC; OIP Techs., Inc. v. Amazon.com, Inc. The specification at paragraph [0006] describes obtaining live bookable prices from a GDS as an established, pre-existing commercial practice: "Flight comparison services, such as Skyscanner, and some airlines, pay a fee to obtain live, bookable prices from a GDS; these prices are the actual bookable prices that a potential passenger can book." (Evidentiary support: Specification para. [0006]; court decisions in MPEP 2106.05(d)(II).)
"Storing data in a computer data store / non-transitory storage medium" The courts have recognized storing and retrieving information in memory as well-understood, routine, conventional activity. See Versata Dev. Group, Inc. v. SAP Am., Inc.; OIP Techs. The specification at paragraph [0247] confirms: "software programming code which embodies or forms part of the present invention is typically stored in permanent, non-transitory storage," confirming conventional storage practices. (Evidentiary support: Specification para. [0247]; court decisions in MPEP 2106.05(d)(II).)
"Using a generic computer server and processors to receive requests, perform calculations, query external systems, and output results" The courts have recognized using a computer to perform calculations and data processing as well-understood, routine, conventional activity when the computer is invoked as a generic tool. See Alice Corp.; Benson. The specification at paragraph [0248] states: "it will be understood that each element of the illustrations… can be implemented by general and/or special purpose hardware-based systems that perform the specified functions or steps, or by combinations of general and/or special-purpose hardware and computer instructions," confirming that the hardware components are generic. (Evidentiary support: Specification paras. [0247], [0248]; court decisions in MPEP 2106.05(d)(II).)
"Sending a request to a Distribution System and receiving a response (steps (vii)–(viii))" Querying an external database or pricing system for information and receiving a response is well-understood, routine, conventional activity, and the specification at paragraphs [0003] and [0181]–[0188] describes querying a GDS as the established pre-existing industry practice for obtaining fare prices. See Content Extraction & Transmission LLC v. Wells Fargo Bank, N.A. (Evidentiary support: Specification paras. [0003], [0181]–[0188]; court decisions in MPEP 2106.05(d)(II).)
"Performing repetitive calculations using a computer (periodic retraining, step (xii))" Performing repetitive calculations is well-understood, routine, conventional computer activity. See Parker v. Flook; Bancorp Services, LLC v. Sun Life Assurance Co. of Canada, ("The computer required by some of Bancorp's claims is employed only for its most basic function, the performance of repetitive calculations"). The specification at paragraph [0247] confirms that all described steps "can be implemented using standard well-known programming techniques." (Evidentiary support: Specification para. [0247]; court decisions in MPEP 2106.05(d)(II)).
"Outputting results to a computing device and storing those results (steps (x)–(xi))" The courts have recognized outputting and storing data as well-understood, routine, conventional activity. See Electric Power Group; Alice Corp., ("creating and maintaining 'shadow accounts'" is conventional). (Evidentiary support: court decisions in MPEP 2106.05(d)(II).)
Even considering these elements as an ordered combination, they do not collectively amount to significantly more. The combination performs these individually conventional steps in the natural sequential order expected for implementing such a workflow on generic computing hardware, without any non-conventional or technically inventive arrangement of the components. Unlike the eligible combination in BASCOM, there is no specific placement of any component that is architecturally unconventional or that overrides the expected conventional sequence of operations. The combination produces a useful commercial result inferring fare class availability without a complete fare cache but as established above, that result is an improvement to the efficiency of the abstract commercial process, not a technological improvement to any computer or other technology. The claims do not include additional elements sufficient to amount to significantly more than the judicial exception, whether evaluated individually or as an ordered combination. Accordingly, the claims are not patent eligible.
Dependent Claims Analysis
The dependent claims do not add limitations that integrate the judicial exception into a practical application or provide significantly more than the judicial exception:
Claims 2, 7, 19, 30 - These claims further define the parameters of the requested travel-related services, specifying activity types (airfare, train fare), desired weather conditions, star ratings, keywords (claims 2 and 30), specific journey features including departure day of week, airline, time to travel, route, and month (claim 7), and train fare prices (claim 19). These limitations specify the type and scope of data inputs to the abstract analytical process field-of-use and subject matter limitations that link the abstract idea to particular types of travel services or query parameters. Limiting an abstract idea to a particular field of use does not integrate it into a practical application. See MPEP 2106.05(h); Affinity Labs of Tex., LLC v. DIRECTV, LLC. These limitations narrow the metes and bounds of the abstract idea but do not add any element that would integrate the judicial exception into a practical application or amount to significantly more.
Claims 3 and 4 - Claim 3 specifies that determination of estimated prices is performed by inferring, deriving, or predicting estimated prices, and claim 4 specifies that step (iv) includes using rules to analyze patterns in the dataset. These limitations further describe the abstract analytical process itself the modes of performing the abstract price estimation and add no element beyond the abstract idea. Describing variations in how an abstract mental or mathematical process is performed does not transform the abstract idea into patent-eligible subject matter.
Claims 5, 6, 8, 9, and 29 - These claims specify the use of a Naïve Bayes classifier machine learning approach producing a probabilistic model of prices (claim 5, mirrored in claim 29 for the system); that classifiers are trained using observed prices and corresponding feature sets (claim 6); that a classifier predicts an unobserved price by returning the most likely price given a feature set (claim 8); and that features are derived by training multiple models with different features and comparing their predictive accuracy (claim 9). These limitations further define the mathematical and statistical process specifically the Naïve Bayes mathematical framework that constitutes the abstract idea itself. Providing additional detail about the mathematical methodology of an abstract idea does not transform it into patent-eligible subject matter. See SAP Am., ("a claim for a new abstract idea is still an abstract idea"). An improvement in the mathematical model or prediction process such as using a Naïve Bayes approach rather than a simpler average is an improvement to the abstract idea, not to technology. See Stanford II.
Claims 10, 11, 12, 13, and 14 - These claims specify that step (iv) includes building a statistical model, identifying missing quote candidates, and pricing them (claim 10); that estimating prices for each candidate involves extracting category feature values, retrieving a trained classifier, extracting all feature values, calculating Bayes posterior probabilities for each price range, choosing the class with the highest probability, and attaching that price class to the candidate quote (claim 11); that statistical model inputs include a list of routes, classifier categorization scheme, historical quotes, and supported features with weights (claim 12); that historical quotes are filtered by age (claim 13); and that inputs include reversed route equivalents (claim 14). These limitations provide additional procedural and mathematical detail about the Naïve Bayes classification process and its inputs. Claim 11, in particular, recites the mathematical calculation of Bayes posterior probabilities explicitly, which falls squarely within the mathematical concepts grouping of abstract ideas. See MPEP 2106.04(a)(2)(I). None of these limitations add elements that reflect a technological improvement to classifier technology, data storage, or any other technology; they add specificity to the abstract mathematical and analytical process.
Claims 15, 16, 17, 18, 20, 21, and 22 - These claims specify including cached fare prices in the set of price estimates (claim 15); that prices are for a one-way journey (claim 16) or return journey (claim 17); that prices include air fare prices (claim 18), car hire prices (claim 20), or hotel prices (claim 21); and that the request comprises a flexible search request (claim 22). These are field-of-use limitations that link the abstract idea to particular types of travel services, journey types, or search request formats. As noted above, limiting an abstract idea to a particular field of use does not integrate it into a practical application. See MPEP 2106.05(h). These limitations narrow the subject matter without adding any qualifying additional element.
Claims 23, 24, 25, and 26 - These claims add the steps of determining confidence ranges of estimated prices from the incomplete historical price dataset (claim 23); providing those confidence ranges together with the price estimates to an end-user computing device (claim 24); using the confidence ranges to decide whether to display a price or a probable range of prices (claim 25); and displaying the probable range as error bars (claim 26). The determination of confidence ranges is itself a mathematical/statistical calculation specifically, computing and bounding Bayesian posterior probability distributions that falls within the mathematical concepts grouping of abstract ideas and/or constitutes a mental process of evaluation and judgment. The display of confidence ranges or error bars to a user is insignificant post-solution activity that outputs the result of the abstract idea. These limitations do not add any element that integrates the judicial exception into a practical application or provides significantly more.
Claim 27 - This claim specifies that the estimation process is parameterized by one or more of: a minimum Bayes posterior probability required to accept a classification result; a maximum number of route operators involved in candidate generation; or a random variation added to Bayes posterior probability to avoid ties. These limitations define specific mathematical parameters of the Naïve Bayes classification process, further instantiating the abstract mathematical idea. Parameterizing a mathematical process with specific threshold values or operational parameters does not transform the abstract idea into patent-eligible subject matter. See Parker v. Flook, (specific mathematical parameters of an abstract algorithm do not make the claim patent-eligible).
Claims 28, 29, and 30 - Claim 28 is an independent system claim reciting the same abstract idea in system format, with the same additional elements (computer server, data store, non-transitory storage medium, computing device, Distribution System) as method claim 1. For all the same reasons articulated above with respect to claim 1, claim 28 does not integrate the abstract idea into a practical application and does not recite significantly more than the abstract idea. The system claim format does not alter the eligibility analysis; the computer server, computing device, and data store are generic computing components performing the same abstract functions. See Versata Dev. Group v. SAP Am., Inc., (system claims directed to abstract ideas on generic hardware are no more eligible than corresponding method claims). Claim 29 depends from claim 28 and adds that the server uses rules to analyze patterns, mirroring method claim 4; for the same reasons as claim 4, this limitation does not remedy the eligibility deficiency. Claim 30 depends from claim 28 and adds specific activity types and parameters, mirroring method claims 2 and 7; for the same reasons as those claims, this limitation does not remedy the eligibility deficiency.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Rowley et al. (U.S. Patent Publication 2018/0053264 A1) discloses systems and methods for promoting customer engagement in travel related programs.
De Marcken (U.S. Patent Publication 2008/0167912 A1) discloses providing travel information using cached summaries of travel options.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to STEPHEN S SWARTZ whose telephone number is (571)270-7789. The examiner can normally be reached Mon-Fri 9:00 - 6:00.
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, Boswell Beth can be reached at 571 272-6737. 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.
/S.S.S/Examiner, Art Unit 3625
/DYLAN C WHITE/Primary Examiner, Art Unit 3625
Prosecution Insights