DETAILED ACTION
This communication is a Final Office Action on the merits in response to communications received on 05/19/2026. Claims 1, 8, and 15 have been amended. Therefore, claims 1, 8, and 15 are pending and have been addressed below. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections – 35 USC §101
1. 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.
2. Claims 1, 8, and 15 are rejected under 35 U.S.C. § 101 because the claimed invention recites an abstract idea without significantly more.
3. Regarding Step One, claim 1 recites a process (i.e., an act or step, or a series of acts or steps), claim 8 recites a machine (i.e., consisting of parts, or of certain devices and combination of devices), claim 15 recites a manufacture (i.e., an article that is given a new form, quality, property, or combination through man-made or artificial means.) Thus, each of these claims fall within one of the four statutory categories.
4. Regarding Step 2A [Prong One], claims 1, 8, and 15 recite:
“extracting features…extract features from data stored in a respective data format…the extracting comprising: extracting intrinsic features comprising market value, total area, age of an associated building, and number of floors of the associated building;”, “extracting neighborhood asset types by extracting asset types for a group of neighboring properties;”, “extracting census information;”, “extracting asset types of other properties by: determining a set of related properties by: identifying a group of neighboring properties based on a map of the each of the one or more properties identifying properties that share a same compound or parcel; and identifying additional properties having a common owner;”, “estimating a multinomial distribution over the asset types for the set of related properties, wherein: the multinomial distribution has k possible results corresponding to a number of the asset types, each possible result has an associated probability that a property is a corresponding asset type, a sum of the associated probabilities is 1, and parameters of the multinomial distribution are estimated using Bayesian inference;”, “extracting…features by: tokenizing NAICS descriptions from… providing the tokenized NAICS descriptions;”, “receiving a set of NAICS embeddings… based on the NAICS tokenized descriptions;” and “averaging the set of NAICS embeddings extracting legal description features by: tokenizing legal descriptions;”, “providing the tokenized legal descriptions to the model;”, “receiving a set of legal embeddings…based on the tokenized legal descriptions averaging the set of legal embeddings and…to the set of legal embeddings;”, “using the features, wherein each of…corresponds to a separate asset type in a set of asset types,…selecting a portion of the features: generating an artificial feature comprising random numbers; computing a feature importance score for each feature and for the artificial feature…; and excluding features whose feature importance score falls below the feature importance score of the artificial feature to obtain the portion of the features; and…using the portion of the features corresponding to the separate asset type;”, “determining an asset type of the set of asset types for each of the one or more properties using the features as an input;” and “outputting…the asset type for each of the one or more properties; and”
Under the broadest reasonable interpretation, the limitations recite an abstract idea for extracting property data, determining, and outputting classification types for one or more real estate properties which is/are a concept that encompasses fundamental economic principles or practices (i.e., hedging, insurance, mitigating risk), commercial interactions (i.e., marketing or sales activities, business relations), mental concepts (i.e., observations, evaluations, judgments, opinions), mathematical concepts (i.e., mathematical calculations) and falls within the certain methods of organizing human activity, mental processes, and mathematical concepts groupings discussed in MPEP 2106.04(a)(2).
The Applicant’s Specification in at least [0002-0005, 0065] Referring to Fig. 2, Fig. 2 is a flow diagram for a process 200 of determining asset types for one or more properties. The process 200 may be implemented to effectuate commercial research on properties in a large area. The process 200 may determine multiple asset types for a single property. Additionally, the process 200 may determine that a property does not have any asset types. The process 200 includes steps of collecting data from a large number of databases. [0071] At step 220, the process may output each asset type of the one or more properties. Each of the determined asset types may be displayed on a list viewable to a user when the property is selected. In various embodiments, the property is displayed on a map at an accurate position relative to other properties. The user may select one of the displayed properties to display the list of its determined asset types.
Consistent with the disclosure, the limitations are considered fundamental economic practices, sales or marketing activities, and/or business relations because the steps of “extracting”, “determining”, “outputting” in the context of the claim may be performed by a user that has access to property data the ability to evaluate and interact with one or more properties that have been classified, i.e., businesses or residences, according to a particular category. In this way, the limitations cover subject matter that may be reasonably characterized as falling within the certain methods of organizing human activity grouping. The limitations are considered mental processes because the steps of “extracting”, “determining”, “outputting” in the context of the claim involve the collection of information, i.e., property data, and processes for understanding the meaning of the information (such as determining an asset type for one or more properties) where these steps can be performed by a human using evaluation and because they involve determinations which are mental tasks humans routinely do and thus can be practically performed in the human mind. In this way, the limitations may be reasonably characterized as falling within the mental processes grouping. The limitations are also considered mathematical calculations because the steps of “estimating” and “computing” in the context of the claim recites mathematical relationships/operations. It is important to note that a mathematical concept need not be expressed in mathematical symbols, because "[w]ords used in a claim operating on data to solve a problem can serve the same purpose as a formula. Accordingly, the claim recites an abstract idea.
5. Regarding Step 2A [Prong Two], the recited additional elements include:
“a processor in communication with a memory”, “from a plurality of databases”, “by the processor”, “using a plurality of format-specific feature extraction components, wherein each format-specific feature extraction component is configured to”, “of a corresponding one of the plurality of databases”, “North American Industry Classification System (NAICS)”, “a NAICS database”, “to a model”, “from the model”, “applying a term frequency-inverse document frequency algorithm”, “training, using a tree-based gradient boosting algorithm, a set of binary classifiers”, “the binary classifiers”, “wherein training comprises: training each of the binary classifiers by:”, “for each of the binary classifiers”, “to a one-versus-rest model and the set of binary classifiers”, “on a user interface of a client device:”, “selectable markers”, “a geographic map”, “each marker”, “on the geographic map”, “one of the selectable markers”, “a detail panel to be displayed”, “displayed on the geographic map”, “the user interface to display”, “a computing system”, “a processing server”, “the processing server”, “a non-transitory computer readable storage medium, with a processor in communication with a memory through a bus, having data stored therein representing a software executable by a computer, the software comprising instructions that, when executed, cause the computer to perform:” – see claims 1, 8, 15 which are features recited at a high-level of generality in light of the specification [Figs. 1, 8-11, ¶ 0029-0049, 0093-0096]. Since the specification describes the additional elements in general terms, without describing the particulars, the additional elements may be broadly but reasonably construed as generic computer components being used in their ordinary capacity to perform the abstract idea. The additional elements merely add the words “apply it” with the judicial exception, or mere instructions to implement the abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea, as discussed in MPEP 2106.05 (f)
The other additional elements of: “collecting data…having dissimilar data types and dissimilar data formats, the data related to the one or more properties;”, “causing to be displayed,…the one or more properties…positioned at a geographic location of the respective property relative to other properties…;”, “in response to a user selecting…identifying the determined asset type of the selected property alongside one or more property attributes of the selected property;” and “the properties…to be filterable by asset type, such that selection of a particular asset type from the set of asset types causes…only those properties having the particular asset type.” merely add insignificant extra-solution activity, i.e., data gathering/output, to the judicial exception, as discussed in MPEP 2106.05(g).
The other additional element of: “a method for determining asset types of one or more properties, the method comprising:” recited in the preamble merely indicates a particular field of use or technological environment in which to apply the judicial exception, as discussed in MPEP 2106.05(h).
Thus, the additional claim elements are not indicative of integration into a practical application, because the claims do not involve improvements to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)), the claims do not apply or use the abstract idea to effect a particular treatment or prophylaxis for a disease or medical condition (Vanda Memo), the claims do not apply the abstract idea with, or by use of, a particular machine (MPEP 2106.05(b)), the claims do not effect a transformation or reduction of a particular article to a different state or thing (MPEP 2106.05(c)), and the claims do not apply or use the abstract idea in some other meaningful way beyond generally linking the use of the abstract idea to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception (MPEP 2106.05(e) and Vanda Memo). Therefore, the claims do not, for example, purport to improve the functioning of a computer. Nor do they effect an improvement in any other technology or technical field. Accordingly, the additional elements do not impose any meaningful limits on practicing the abstract idea and the claims are directed to an abstract idea.
6. 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 integration of the abstract idea into a practical application, the additional element(s) of: “a processor in communication with a memory”, “from a plurality of databases”, “by the processor”, “using a plurality of format-specific feature extraction components, wherein each format-specific feature extraction component is configured to”, “of a corresponding one of the plurality of databases”, “North American Industry Classification System (NAICS)”, “a NAICS database”, “to a model”, “from the model”, “applying a term frequency-inverse document frequency algorithm”, “training, using a tree-based gradient boosting algorithm, a set of binary classifiers”, “the binary classifiers”, “wherein training comprises: training each of the binary classifiers by:”, “for each of the binary classifiers”, “to a one-versus-rest model and the set of binary classifiers”, “on a user interface of a client device:”, “selectable markers”, “a geographic map”, “each marker”, “on the geographic map”, “one of the selectable markers”, “a detail panel to be displayed”, “displayed on the geographic map”, “the user interface to display”, “a computing system”, “a processing server”, “the processing server”, “a non-transitory computer readable storage medium, with a processor in communication with a memory through a bus, having data stored therein representing a software executable by a computer, the software comprising instructions that, when executed, cause the computer to perform:” – see claims 1, 8, 15 at best add the words “apply it” to the judicial exception and mere instructions to apply the judicial exception does not provide an inventive concept at Step 2B.
The other additional element of: “collecting data…having dissimilar data types and dissimilar data formats, the data related to the one or more properties;”, “causing to be displayed,…the one or more properties…positioned at a geographic location of the respective property relative to other properties…;”, “in response to a user selecting…identifying the determined asset type of the selected property alongside one or more property attributes of the selected property;” and “the properties…to be filterable by asset type, such that selection of a particular asset type from the set of asset types causes…only those properties having the particular asset type.” were considered to be insignificant extra-solution activity in Step 2A, and thus re-evaluated in Step 2B to determine if it is more than well-understood, routine, conventional activity in the field. The Symantec, TLI, OIP Techs, Versata Dev. Group court decisions cited in MPEP 2106.05(d)(II) indicate: “receiving or transmitting data over a network”, “storing and retrieving information in memory”, and “presenting offers and gathering statistics” are well-understood, routine, conventional activity. Thus, when viewing these additional elements individually and as a whole in combination with the judicial exception, these additional elements do not provide an inventive concept at Step 2B.
Response to Arguments
7. Applicant's arguments filed 05/19/2026 have been fully considered but they are not persuasive.
With Respect to Rejections Under 35 USC 101
Applicant argues “Applicant respectfully asserts that no independent claim explicitly sets forth an abstract idea in such a manner. The MPEP provides, as an example of "describe", the description of the concept if intermediated settlement in Alice Corp. v. CLS Bank without including the words intermediated or settlement. Id. Applicant respectfully asserts that there is no such description of an abstract idea present in any independent claim. MPEP § 2106 goes on to explain if a claim merely involves a judicial exception, rather than reciting it (or setting forth or describing it), then the "claim is eligible without further analysis". MPEP § 2106. As an example, the MPEP provides a claim that recites a judicial exception as one that sets forth "A machine comprising elements that operate in accordance with F=ma". Id. Such a claim requires further analysis under Prong Two of Step 2A. The MPEP further provides an example of a claim, which merely involves an exception: "A teeter-totter comprising an elongated member pivotably attached to a base member, having seats and handles attached at opposing sides of the elongated member". Id. The MPEP goes on to explain that "This claim is based on the concept of a lever pivoting on a fulcrum, which involves the natural principles of mechanical advantage and the law of the lever. However, this claim does not recite these natural principles and therefore is not directed to a judicial exception". Id. Therefore, the MPEP concludes that the hypothetical claim is eligible without further analysis. Id. In addition, the USPTO has provided additional guidance in the form of a series of subject matter eligibility examples.”
“In one such example, Example 39, the USPTO provided an example for training a neural network for facial detection with a claim that provides for a series of data collection and modification steps to generate training data and then generically training the neural network using the training data. See 2019 Revised Patent Subject Matter Eligibility Guidance (2019 PEG), Example 39. The guidance continues, stating that this claim is eligible because it does not recite a judicial exception under Prong 1. See id. Further, the guidance states that this claim is based on mathematical concepts, those concepts are not recited in the claims and, further, the claims does not recite a mental process nor does it recite any method of organizing human activity.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The remarks restate guidance from the MPEP and training Example 39 but do not sufficiently explain how the guidance applies to the claims in the present case. At best, the remarks are attempting to use the guidance to establish that there is no abstract idea recited, however, these conclusory statements made by Applicant do not alter the previous 101 analysis or make the claimed invention any less abstract. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Independent claim 1, as amended, requires a series of data preparation steps (i.e., the steps of collecting and the detailed set of extracting steps), a step of training multiple classifiers, using the classifiers along with a one-versus-rest model to generate a prediction using the data prepared in the data preparation steps using models to obtain a prediction, and a step of presenting the data on a map in a particular manner. As such, the claims provide a specific method of data manipulation and use of models to obtain a specific type of prediction data.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. Here, the claim is written in “result-based functional language” that does sufficiently describe how to generate a prediction or present data in a map in a non-abstract way. The series of steps of claim merely describe collecting and extracting the data necessary to generate a prediction and then presenting data on a map which is all abstract. As discussed in MPEP 2106.04 (a)(2)(III)(A) - claims recite a mental process when they contain limitations that can practically be performed in the human mind, including for example, observations, evaluations, judgments, and opinions. For example, a claim to "collecting information, analyzing it, and displaying certain results of the collection and analysis," where the data analysis steps are recited at a high level of generality such that they could practically be performed in the human mind, Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016). The additional elements of “training multiple classifiers”, “using the classifiers along with a one-versus-rest model” do not preclude the identified limitations under Prong One of the analysis from being within the certain methods of organizing human activity, mental processes, and/or mathematical concepts groupings. The limitation for “presenting data on a map” as recited in claim 1 also adds insignificant extra solution activity to the abstract idea. See MPEP 2106.05(g). For these reasons, the rejections under 101 are being maintained.
Applicant further argues “This claimed method is highly analogous to the Example 39 described above, which also provides for a method of generating a model that is able to provide a certain type of prediction data (i.e., facial detection). In particular, the claims provide for a set of data manipulation steps, like Example 39, and training of models, also like Example 39. Further, the current recitations include a much more detailed set of steps for preparing the data than those provided in Example 39. In addition, the present claims provide for actually using the trained models, which is an additional step that is not present in Example 39. Moreover, the present claims provide for displaying the generated data. Thus, for at least the same reasons that Example 39 is eligible, the pending claims should also be found to be eligible under Prong 1.” The Examiner respectfully disagrees.
The Applicant’s argument are not persuasive. For example, the asserted claims or remarks do not discuss any training of a neural network as in Example 39. The Examiner asserts the cited training example is directed to a specific technological improvement, whereas the present claims are merely directed to applying an abstract idea collecting, analyzing, and displaying information with generic computing components. The trained models are simply being used as a tool to aid in performing the abstract idea. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Further, as amended, independent claims 1, 8, and 15 include three additional sets of limitations that further remove them from any abstract-idea characterization. First, the claims recite collecting data from a plurality of databases having dissimilar data types and dissimilar data formats, using a plurality of format-specific feature extraction components, wherein each format-specific feature extraction component is configured to extract features from data stored in a respective data format of a corresponding one of the plurality of databases. This limitation describes a concrete computing architecture and does not set forth any mathematical relationship, formula, or equation.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The step of “collecting data” is performing the necessary data gathering to aid in performance of the abstract idea. As previously explained the step adds pre-solution activity to the abstract idea. See MPEP 2106.05(g) The specific type of data, data format, and features being collected and extracted further describes the type of information that may be used but does not alter the analysis or make the claimed invention any less abstract. Merely reciting generic computing components in a claim such “a plurality of databases” and “a plurality of format-specific feature extraction components” does not lead towards eligibility. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Second, the claims now recite selecting features for each binary classifier by generating an artificial feature comprising random numbers, computing a feature importance score for each feature and for the artificial feature using the tree-based gradient boosting algorithm, and excluding features whose score falls below the artificial feature's score. This limitation does not name any mathematical algorithm such as backpropagation or gradient descent as an essential claim element. Like the "training the neural network" step in Example 39, it describes machine- executable steps that do not set forth mathematical relationships.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. As for “selecting”, “generating”, “excluding” steps, the courts have previously held merely selecting information, by content or source, for collection, analysis, and display does nothing significant to differentiate a process from ordinary mental processes, whose implicit exclusion from § 101 undergirds the information-based category of abstract ideas. As for computing a feature score, the claim now recites a mathematical calculation. See MPEP 2106.04(a)(2)(I)(C) - a mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number, e.g., performing an arithmetic operation such as exponentiation. There is no particular word or set of words that indicates a claim recites a mathematical calculation. That is, a claim does not have to recite the word "calculating" in order to be considered a mathematical calculation. At best understood, the remarks directed toward these limitations are combining one abstract idea (mental processes) with another (mathematical concepts) which does not make the claimed invention any less abstract.
Applicant further argues “Third, the claims now recite a filterable geographic map display with selectable property markers positioned at geographic coordinates, detail panels triggered by user selection, and asset-type-based filtering. These limitations describe a specific interactive user interface architecture and do not set forth or describe any mathematical concept, mental process, or method of organizing human activity.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. Here, the remarks do not change the analysis under Step 2A Prong One and rely upon features related to the interactive user interface and map that were considered under Step 2A Prong Two of the analysis. In the instant case, a processor is used in conjunction with an interface to "filter" or “select” data within the map, but "mere automation of manual processes using generic computers does not constitute a patentable improvement in computer technology. Credit Acceptance Corp. v. Westlake Servs., 859 F.3d 1044, 1055 (Fed. Cir. 2017). For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Accordingly, the amended claims are even more strongly analogous to Example 39 than the prior claims. In addition, as a practical matter, the human mind is not equipped to perform this method. See MPEP § 2106.04(a)(2)(III)(A). The method requires a process for training models, performing computer-based manipulations to data, and using both inputs and outputs of models, which a human mind is unable to perform without the use of a computer. See e.g., Research Corp. Techs., 627 F.3d at 868, 97 USPQ2d at 1280. In particular, the amended claims' artificial-feature-based feature selection step requires training a gradient boosting model, computing feature importance scores across a large feature matrix of potentially millions of data points, and comparing each score to a synthetic noise baseline, machine-scale operations that cannot practically be performed in the human mind or with pen and paper. Subsequently, because the claims, as amended, cannot be performed in the human mind or by a human using pen and paper, the amended claims fail to recite a mental process (i.e., an abstract idea), and the analysis under the Mayo/Alice Test should end with a conclusion that the claims are directed to patent eligible subject matter, and the independent claims should be considered patent eligible.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The Examiner maintains the claims are directed to an abstract idea and the discussion points above have already addressed and explained why training Example 39 is not applicable. At best, the remarks rely upon results-based functional claim language such as (artificial-feature-based feature selection step for computing feature importance scores across a large feature matrix of potentially millions of data points, and comparing each score to a synthetic noise baseline) which is a frequent feature of ineligible claims. As discussed in MPEP 2106.04(a)(2) – claims can recite a mental processes even though they are being performed by a computer. The inability for the human mind to perform each claim step does not alone confer patentability because the courts have previously held, "the fact that the required calculations could be performed more efficiently via a computer does not materially alter the patent eligibility of the claimed subject matter." Bancorp Servs., 687 F.3d at 1278. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Similar to McRO, where the Federal Circuit found claims eligible because they were "limited to rules with specific characteristics" that improved computer animation technology, the present claims are limited to a specific, technical implementation for analyzing property data through a particular ordered combination of, at least: (i) Collecting data from a plurality of databases having dissimilar data types and
dissimilar data formats; (ii) Extracting multiple categories of features using a plurality of format-specific feature extraction components, wherein each component is configured to extract features from data stored in a respective data format of a corresponding database, the features including intrinsic features, neighborhood asset types, census
information, and related property distributions; (iii) Applying specific technical processes including tokenization of NAICS and legal descriptions; (iv) Generating embeddings through models and averaging techniques; (v) Estimating multinomial distributions using Bayesian inference with specific constraints; (vi) Training binary classifiers using an artificial-feature-based feature selection methodology, comprising generating a synthetic noise feature of random numbers, computing a feature importance score for each real feature and for the artificial feature using the tree-based gradient boosting algorithm, and excluding features whose importance score fails to surpass the noise baseline, ensuring only informative features are used to train each binary classifier; and
(vii) Presenting the classified results on a filterable geographic map interface, displaying properties as selectable markers at their geographic locations, showing a detail panel identifying the asset type and property attributes in response to user selection, and dynamically filtering the displayed properties by asset type.
“Like the claims in McRO that used specific rules to achieve improved lip synchronization rather than automating animator tasks, the present claims use specific technical processes and techniques to achieve improved property classification rather than merely automating human classification tasks. The claim requires "specific features" and "specific technical implementations" that parallel the "specific... rules defining outputs based on timing" that rendered the McRO claims eligible.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The Examiner asserts the claims in McRO were directed to specific technological improvements in computer technology, whereas, the asserted claims are directed to an abstract idea for determining and displaying property asset types using generic computing components recited at a high-level of generality. The asserted claims do not recite a comparable technological improvement. Although the remarks restate various steps and features recited in the claim as an ordered combination, claims are generally not saved from abstraction merely because they recite components more specific than a generic computer. BSG Tech LLC v. BuySeasons, Inc., 899 F.3d 1281, 1286 (Fed. Cir. 2018) For these reasons, the rejections under 101 are being maintained.
Applicant further argues “In addition, like the claims in McRO, the present claims do not preempt all ways of determining property asset types. Rather, they are limited to the specific technical implementation involving the particular combination of format-specific feature extraction, embedding generation, statistical modeling, artificial-feature-based classifier training, and filterable geographic map presentation recited in the claims. These specific technical elements place the claims squarely within the realm of patent-eligible subject matter under McRo and its progeny.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The specificity of the presently recited techniques here does not lead towards eligibility. As for remarks directed towards preemption, the Examiner asserts the courts have previously held that claims that are otherwise directed to patent-ineligible subject matter cannot be saved by arguing the absence of complete preemption. See, e.g., Synopsys, 839 F.3d at 1150 (holding that an argument about the absence of complete preemption "misses the mark"); FairWarning, 839 F.3d at 1098 ("But even assuming that the ... patent does not preempt the field, its lack of preemption does not save these claims.") "While preemption may signal patent ineligible subject matter, the absence of complete preemption does not demonstrate patent eligibility." In the instance case, arguments about the lack of preemption risk cannot save claims that are deemed to only be directed to patent-ineligible subject matter. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Moreover, similar to the claims in Desjardins, the claims provide for an improvement to the functionality of machine learning models themselves. The technical field of machine learning cannot be directly or generically applied to data about properties. Instead, the field of machine learning faces specific problems based on the real-world collection and use of data about properties. These problems are addressed in the Specification and the improvements to the field of machine learning are provided in the claimed invention. In particular, the claims provide for a highly-specific set of steps for extracting features, training a set of binary classifiers using the features, and determining an asset type using the trained binary classifiers and a one-versus-rest model.”
“More specifically, as provided in the Specification, the underlying data is sourced from various databases that have dissimilar types of data and store the data in different formats, thus requiring the claimed set of feature extraction steps that must each be customized to the specific type of data sought to be extracted. See Specification, para. [0040]. This problem faced by the technical field and the improvement thereto is provided by the plurality of format-specific feature
extraction components, each configured to extract features from data stored in a respective data format, now explicitly recited in the amended claims. Further, the binary classifiers must be able to scale from millions of data points, be parallelizable, run fast, and use low amounts of memory and the training of the binary classifiers must also be able to scale to millions of data points and handle missing values. See Specification, para. [0047]. In the current response, the amended claims directly address these requirements by adding the artificial-feature-based feature selection step. Specifically, by generating a synthetic feature of random numbers as a noise baseline, computing a feature importance score for each real feature and for the artificial feature using the tree-based gradient boosting algorithm, and excluding any feature whose score fails to surpass the noise baseline, the amended claims ensure that binary classifiers are trained on only the most informative features. This noise-based pruning reduces the feature set to those providing more signal than random noise, directly enabling the classifiers to run faster, use less memory, and scale to millions of data points, the precise technical constraints identified in the Specification.
This is an improvement to how the machine learning classifiers operate. The Examiner characterized the prior claims as calling for "generic use of binary classifiers." The amended claims directly address that characterization: they specify, in technically concrete terms, how the classifiers are trained in a way that improves their performance and scalability. As can be seen, the technical field of machine learning faces specific problems when applied to data related to property. The claims address these specific problems and presents an improvement to the field of machine learning to enable the technology of machine learning to be properly applied to data related to property. As such, like Desjardins, the claimed invention is directed to an improvement in computer functionality.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. In the Ex Parte Desjardins court decision, the court cited to the Specification and limitations reflected in the claim. The court held that the claimed invention improves the operation of a machine learning system, such as by enhancing its training efficiency or preserving prior learning, it is not “directed to” an abstract idea under Alice Step 1. The Applicant’s presently recited claim limitations do not recite a comparable technological solution. The passages from the Applicant’s Specification [¶ 0040, 0047] at best discuss the type/format of the data and extraction components necessary to collect the data along with a general explanation relating to benefits for using the classifiers. The cited passages do not support a finding that the claims are directed to a technological improvement in machine learning functionality. The Examiner asserts if there are any improvements they are within the recited abstract idea and not the generic computing components or machinery being used to carry out the abstract idea. For these reasons, the rejections under 101 are being maintained.
Applicant further argues “Evaluating additional elements to determine whether they amount to an inventive concept requires considering them both individually and in combination to ensure that they amount to significantly more than the judicial exception itself. Because this approach considers all claim elements, the Supreme Court has noted that "it is consistent with the general rule that patent claims 'must be considered as a whole."' [citation omitted] Consideration of the elements in combination is particularly important, because even if an additional element does not amount to significantly more on its own, it can still amount to significantly more when considered in combination with the other elements of the claim". MPEP § 2106.”
“The Examiner has failed to consider the additional elements as a whole. For example. See MPEP § 2106(I)(B). The amended claims' ordered combination of (i) format-specific feature extraction components tailored to each database's data format; (ii) artificial-feature-based feature selection that uses the gradient boosting algorithm's own importance scores to exclude features that fail to outperform random noise; and (iii) a filterable geographic map interface with selectable markers, detail panels, and asset-type-driven display filtering, together impose meaningful limits on any recited exception. This ordered combination is not routine, conventional, or well-understood activity, and the Examiner has not provided evidence in the record demonstrating otherwise.” The Examiner respectfully disagrees.
The Applicant’s arguments are not persuasive. The Examiner maintains whether viewing the claim limitations individually or as an ordered combination, the asserted claims that encompass (i.e., (i) format-specific feature extraction components tailored to each database's data format; (ii) artificial-feature-based feature selection that uses the gradient boosting algorithm's own importance scores to exclude features that fail to outperform random noise; and (iii) a filterable geographic map interface with selectable markers, detail panels, and asset-type-driven display filtering) do not add an inventive concept. The remarks rely upon these three different steps and purport these elements are not routine, conventional, or well-understood activity.
Here, the Applicant does not specifically point to anything inventive about the ordered combination of elements. Merely narrowing or reformulating an abstract idea does not add "significantly more" to it. The conclusory allegations that the prior art lacked elements of the asserted claims is insufficient to demonstrate an inventive concept. see also Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 716 (Fed. Cir. 2014) ("That some of the eleven steps were not previously employed in this art is not enough—standing alone—to confer patent eligibility upon the claims at issue."); Elec. Power Grp., 830 F.3d at 1355 (using off-the-shelf computer, network and display technology to gather, analyze, send, and present data is not inventive) As for the remarks directed towards evidence, they do not properly address the Examiner’s previous findings. The last Non-Final Office Action, pg. 8 relies upon MPEP 2106.05(d)(II) and cites to court decisions discussed in this section to further support why the “collecting” and “displaying” steps recited in the claim are computer functions that are considered well-understood, routine, and conventional under Berkheimer Option 2. For these reasons, the rejections under 101 are being maintained.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EHRIN PRATT whose telephone number is (571)270-3184. The examiner can normally be reached 8-5 EST Monday-Friday.
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, Lynda Jasmin can be reached at 571-272-6782. 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.
/EHRIN L PRATT/Examiner, Art Unit 3629
/ANDREW B WHITAKER/Primary Examiner, Art Unit 3629
Prosecution Insights