Prosecution Insights
Last updated: November 20, 2026
Application No. 18/972,562

Automated Tool For Determining And Using Dwelling Attribute Utility Data For Dwelling Similarity Rankings

Non-Final OA §101§103
Filed
Dec 06, 2024
Examiner
ALLEN, BRITTANY N
Art Unit
2169
Tech Center
2100 — Computer Architecture & Software
Assignee
MFTB Holdco Inc.
OA Round
3 (Non-Final)
42%
Grant Probability
Moderate
3-4
OA Rounds
2y 4m
Est. Remaining
80%
With Interview

Examiner Intelligence

Grants 42% of resolved cases
42%
Career Allowance Rate
170 granted / 403 resolved
-12.8% vs TC avg
Strong +38% interview lift
Without
With
+37.8%
Interview Lift
resolved cases with interview
Typical timeline
4y 4m
Avg Prosecution
25 currently pending
Career history
443
Total Applications
across all art units

Statute-Specific Performance

§101
17.8%
-22.2% vs TC avg
§103
52.9%
+12.9% vs TC avg
§102
13.0%
-27.0% vs TC avg
§112
13.4%
-26.6% vs TC avg
Black line = Tech Center average estimate • Based on career data from 403 resolved cases

Office Action

§101 §103
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 . 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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 6/3/26 has been entered. Remarks This action is in response to the application received on 6/3/26. Claims 1-19 and 21-23 are pending in the application. Claim 20 has been cancelled. Applicants' arguments have been carefully and respectfully considered. Claims 1-19 and 21-23 are rejected under 35 U.S.C. 101. Claims 1-19, 21, and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Warnaar (US 11,436,240), and further in view of Zhang et al. (US 2022/0277015). Claim 22 is rejected under 35 U.S.C. 103 as being unpatentable over Warnaar in view of Zhang, and further in view of Carney et al. (US 2025/0021882). 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 1-19 and 21-23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception without significantly more. Step 2A, Prong One asks: Is the claim directed to a law of nature, a natural phenomenon (product of nature) or an abstract idea? See MPEP 2106.04 Part I. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. See MPEP 2106.04(a). With respect to claim 1, the limitation of “generating, by the one or more computing systems, search results”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “by the one or more computing systems,” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “by the one or more computing systems” language, “generating” in the context of this claim encompasses the user mentally determining data return to the user. The limitations of “determining” other data such as candidate houses, scores, quantifications of differences, etc., as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, “determining” in the context of this claim encompasses the user analyzing data. Similarly, the limitations of “combining the differential utility scores”, “ranking the multiple candidate houses”, and “selecting some of the ranked candidate houses as the search results”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, “combining”, “ranking”, and “selecting” in the context of this claim encompasses the user analyzing data. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. At step 2a, prong two, this judicial exception is not integrated into a practical application. The claims recite one or more computing systems, a predictive feed forward neural network model, and a displayed graphical user interface, however, this is recited as a high-level of generality (i.e., as a generic processor performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Additionally, the claim recites “receiving, by one or more computing systems, a search query” and “presenting, by the one or more computing systems and in a displayed graphical user interface, information.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. With respect to “receiving, by one or more computing systems, a search query”, the courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). With respect to “presenting, by the one or more computing systems and in a displayed graphical user interface, information”, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. Considering the additional elements individually and in combination and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. The claim is not patent eligible. Claims 2 and 3 further define elements discussed above and do not provide additional elements. With respect to claim 4, the limitations of “determining” other data such as candidate houses, scores, quantifications of differences, etc., as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “by the one or more computing systems” language, “determining” in the context of this claim encompasses the user analyzing data. Similarly, the limitations of “combining, by the one or more computing systems, the differential utility scores” and “ranking, by the one or more computing systems, the multiple candidate houses”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “by the one or more computing systems” language, “combining” and “ranking” in the context of this claim encompasses the user analyzing data. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. At step 2a, prong two, this judicial exception is not integrated into a practical application. The claims recite one or more computing systems and a predictive neural network model, however, this is recited as a high-level of generality (i.e., as a generic processor performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Additionally, the claim recites “receiving, by one or more computing systems, a search query” and “providing, by the one or more computing systems and as search results based on the target dwelling, at least some of the multiple candidate dwellings.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. With respect to “providing, by the one or more computing systems and as search results based on the target dwelling, at least some of the multiple candidate dwellings”, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. With respect to claim 5, the claim recites “receiving, by one or more computing systems, a search query.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Claims 6-11 further define elements discussed above and do not provide additional elements. With respect to claim 12, the limitations of “determining” other data such as candidate houses, scores, quantifications of differences, etc., as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, “determining” in the context of this claim encompasses the user analyzing data. Similarly, the limitations of “generating the determined quantification of difference between that candidate dwelling and the target dwelling”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, “generating the determined quantification of difference between that candidate dwelling and the target dwelling” in the context of this claim encompasses the user analyzing data. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. At step 2a, prong two, this judicial exception is not integrated into a practical application. The claims recite one or more hardware processors, one or more memories, and a predictive model, however, this is recited as a high-level of generality (i.e., as a generic processor performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Additionally, the claim recites “providing indications of at least some of the multiple candidate dwellings.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. With respect to “providing indications of at least some of the multiple candidate dwellings”, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. With respect to claim 13, the claim recites “receiving a search query” and “presenting, by the one or more computing systems and in a displayed graphical user interface, information.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Additionally, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. Claims 14-17 further define elements discussed above and do not provide additional elements. With respect to claim 18, the limitations of “determining” other data such as candidate houses, scores, quantifications of differences, etc., as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “by the one or more computing systems” language, “determining” in the context of this claim encompasses the user analyzing data. Similarly, the limitations of “generating the determined quantification of difference between that candidate dwelling and the target dwelling”, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the “by the one or more computing systems” language, “generating the determined quantification of difference between that candidate dwelling and the target dwelling” in the context of this claim encompasses the user analyzing data. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. At step 2a, prong two, this judicial exception is not integrated into a practical application. The claims recite one or more computing devices, however, this is recited as a high-level of generality (i.e., as a generic processor performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Additionally, the claim recites “providing indications of at least some of the multiple candidate dwellings.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. With respect to “providing indications of at least some of the multiple candidate dwellings”, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. With respect to claim 19, the claim recites “receiving a search query” and “presenting, in a displayed graphical user interface, information.” These elements do not integrate the abstract idea into a practical application because they do not impose a meaningful limit on the judicial exception and provide only insignificant extra solution activity that is mere data gathering in conjunction with the abstract idea. The courts have found limitations directed towards data gathering to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Additionally, the courts have found limitations directed towards outputting information to be well-understood, routine, and conventional. See MPEP 2106.05(d)(II). Presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93. Claims 21-23 further define elements discussed above and do not provide additional elements. Claim Rejections - 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claims 1-19, 21, and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Warnaar (US 11,436,240), and further in view of Zhang et al. (US 2022/0277015). With respect to claim 1, Warnaar teaches a computer-implemented method comprising: receiving, by one or more computing systems, a search query indicating a target house having multiple dwelling attributes each with one or more indicated values (Warnaar, Col. 9 Li. 1-3, At step 304, seeking user's real estate needs (i.e., explicit needs and implicit needs) may be identified from the user data collected from seeking user 204. & Col. 9 Li. 11-12, At step 306, attributes of each real estate need of seeking user 204 may be identified & Col. 9 Li. 34-35, At step 310, explicit preferences and implicit preferences of seeking user 204 for each attribute may be identified); generating, by the one or more computing systems, search results in response to the search query (Warnaar, Col. 18 Li. 4-5, identifying a set of comparable real estate properties) by: determining multiple candidate houses that are located within a defined distance of a geographical location of the target house, each of the multiple candidate houses having one or more associated values for each of the multiple dwelling attributes (Warnaar, Col. 12 Li. 62-67, At step 616, the real estate attributes may be analyzed to identify explicit attributes 618 and to identify implicit attributes 620. Furthermore, geographic maps and images that show proximity of natural and man-made features and accurate traffic estimations can provide real estate attributes relative to a location & Col. 13 Li. 19-22, each proximity attribute is described by primary characteristics of static data such as footprint, height, type, age, and elevation difference with and distance to real estate property 702); determining using a predictive … model trained to provide overall utility scores for houses based on the multiple dwelling attributes, an overall target utility score for the target house, and respective overall utility scores for the multiple candidate houses (Warnaar, Col. 18 Li. 11-14, At step 1206, the aggregate score for each real estate property selected at step 1204 may be computed using the process depicted in FIG. 15 and described in detail below. & Col. 19 Li. 54-58, each attribute in collection of primary attributes 1502 and collection of secondary attributes 1504 of real estate property 1404 may be assigned a single numeric score based on how well the attribute conforms to the metrics that indicate desirability & Col. 20 Li. 15-21, the attributes, prioritized and weighted, may be further analyzed and combined into relevant groups of attributes by a decision network 1522. Decision network 1522 may be based on machine learning and artificial intelligence tools to create scoring that employs probability to predict user satisfaction. Decision network 1522 can result in a plurality of component scores 1524); determining respective quantifications of differences between the multiple candidate houses and the target house based on the multiple dwelling attributes (Warnaar, Col. 18 Li. 16-21, At step 1210, primary attributes may be identified in the set of properties selected in step 1208 where multiple properties in the set have primary attributes in common and the primary attributes make a significant contribution in the aggregate score of at least one property in the set.), by, for each of the multiple candidate houses: determining two or more differing dwelling attributes of the multiple dwelling attributes for the candidate house relative to the target house, each of the two or more differing dwelling attributes having one or more indicated values for the target house that differ from the one or more associated values for the differing dwelling attribute for the candidate house (Warnaar, Col. 19 Li. 54-61, each attribute in collection of primary attributes 1502 and collection of secondary attributes 1504 of real estate property 1404 may be assigned a single numeric score based on how well the attribute conforms to the metrics that indicate desirability. For example, attribute 1516 can have a metric of "unfulfilled", "partially fulfilled" or "fulfilled". The "fulfilled" state may not change no matter how many instances of the attribute there are.); determining for each of the two or more differing dwelling attributes, a differential utility score for the dwelling attribute of the candidate house by identifying and using weighting data internal to the trained predictive feed forward neural network model associated with the differing dwelling attribute (Warnaar, Col. 20 Li. 13-21, the attributes, now prioritized by desirability, are associated with imputed weighted scores 1520. In some embodiments, the attributes, prioritized and weighted, may be further analyzed and combined into relevant groups of attributes by a decision network 1522. Decision network 1522 may be based on machine learning and artificial intelligence tools to create scoring that employs probability to predict user satisfaction. Decision network 1522 can result in a plurality of component scores 1524), wherein the differential utility score is: associated with a difference between the one or more indicated values of the dwelling attribute for the target house and the one or more associated values of the dwelling attribute for the candidate house; and represents a percentage contribution of the dwelling attribute to an overall difference between the respective overall utility score for the candidate house and the overall target utility score for the target house (Warnaar, Col. 19 Li. 61-67, As an example, the attribute may be a 2-car garage. If real estate property 1404 includes no garage, the awarded value may be 0. If real estate property 1404 includes a 1-car garage, the attribute may be awarded a value of 1. & Col. 20 Li. 20-24, Decision network 1522 can result in a plurality of component scores 1524 that are combined in a single score 1528. For example, decision network 1522 can determine the contributions of primary and secondary scores separately.) combining the differential utility scores for the two or more differing dwelling attributes for the candidate house (Warnaar, Col. 20 Li. 20-22, Decision network 1522 can result in a plurality of component scores 1524 that are combined in a single score 1528.); ranking the multiple candidate houses based on the respective quantifications of differences (Warnaar, Col. 21 Li. 32-36, A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204); and selecting some of the multiple candidate houses as the search results based on the ranking, relative to a threshold (Warnaar, Col. 21 Li. 38-41, seeking user 204 may set a threshold 1812 for the size of subset 1810, the minimum listing score, or any other criteria that limits the set of real estate properties offered to seeking user 204 for review.); and presenting, by the one or more computing systems …, information about the some of the multiple candidate houses (Warnaar, Col. 14 Li. 34-36, real estate properties with implicit scores above a predetermined threshold may be presented to seeking user 204.). Warnaar doesn't expressly discuss determining using a predictive feed forward neural network model trained to provide overall utility scores for houses based on the multiple dwelling attributes, an overall target utility score for the target house, and respective overall utility scores for the multiple candidate houses and presenting, by the one or more computing systems and in a displayed graphical user interface, information about the some of the multiple candidate houses. Zhang teaches determining, by the one or more computing systems and using a predictive feed forward neural network model trained to provide overall utility scores for houses based on values for those houses of the multiple dwelling attributes (Zhang, pa 0068, FIG. 6 shows one non-limiting implementation of the second attention-based neural network 132. From top to bottom, a feed-forward neural network 602 can use any number of fully-connected layers to map a plurality of instances of enhanced similarity information into a plurality of instances of projected feature information.), an overall target utility score for the target house and respective overall utility scores for the multiple candidate houses (Zhang, pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network that includes any number of layers, and which can use any activation function(s) (such as ReLU, etc.). The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. & pa 0091, generating (e.g., in block 708), using a ranking subsystem (e.g., ranking subsystem 128), plural ranking scores based on the plurality of instances of query-to-result information); determining two or more differing dwelling attributes of the multiple dwelling attributes for the candidate house relative to the target house, each of the two or more differing dwelling attributes having one or more indicated values for the target house that differ from the one or more associated values for the differing dwelling attribute for the candidate house (Zhang, pa 0047, the first attention-based neural network 126 maps instances of original feature information to original Q-to-R similarity information. The original feature information may include text-based features and structure-based features. The combination component 130 combines supplemental feature information with each instance of Q-to-R similarity information, e.g., by concatenating these two pieces of information together… The combination component 130 produces a plurality of instances of enhanced similarity information.); determining for each of the two or more differing dwelling attributes, a differential utility score for the dwelling attribute of the candidate house by identifying and using weighting data internal to the trained predictive feed forward neural network model associated with the differing dwelling attribute (Zhang, pa 0055, the second attention based neural network 132 packs the instances of enhanced similarity information into a matrix Eenhanced and multiples that same matrix Eenhanced by three respective machine trained weighting matrices (wQ2, WK2 , WV2) to produce the Q, K, and V matrices. & pa 0069, An attention processing component then maps the plurality of instances of projected feature information into a plurality of corresponding instances of attention feature information.), … ranking the multiple candidate houses based on the respective quantifications of differences (Zhang, pa 0091, generating (e.g., in block 708), using a ranking subsystem (e.g., ranking subsystem 128), plural ranking scores based on the plurality of instances of query-to-result information); selecting some of the multiple candidate houses as the search results based on the ranking, relative to a threshold (Zhang, pa 0091, The method then includes identifying (e.g., in block 710) at least a top-ranking candidate result item to be presented to a user based on the plural scores.); presenting, by the one or more computing systems and in a displayed graphical user interface, information about the some candidate houses (Zhang, pa 0091, The method then includes identifying (e.g., in block 710) at least a top-ranking candidate result item to be presented to a user based on the plural scores.). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Zhang because it allows eliminates or reduces manually crafted features while enabling single pass computation for efficient use of computing resources (Zhang, pa 0006). With respect to claim 2, Warnaar in view of Wang teaches the computer-implemented method of claim 1 wherein the determining for each of the multiple candidate houses of the differential utility score for each of the two or more differing dwelling attributes includes extracting the weighting data internal to the trained predictive feed forward neural network model associated with the differing dwelling attribute, and using the extracted weighting data to determine the differential utility score for the differing dwelling attribute (Zhang, pa 0065, the attention processing component produces the query matrix Q based on the query embedding information in conjunction with a first machine Trained weighting matrix WQ1 & pa 0070, A concatenation component optionally concatenates the attention feature information with corresponding instances of the projected feature information, to produce a plurality of instances of result-to-result (R-to-R) information. & pa 0076, In block 808, the ranking subsystem 128 maps, using the scoring neural network 134, the plurality of instances of result-to-result context information into the plural respective ranking scores.). With respect to claim 3, Warnaar in view of Wang teaches the computer-implemented method of claim 1 wherein the defined distance of the geographical location of the target house includes one or more geographical regions, wherein the predictive feed forward neural network model is trained to predict the overall utility scores for houses in the one or more geographical regions (Warnaar, Col. 11 Li. 15-19, information indicative of geographic locations is stored in geographic feature database (GFD) 432. The geographic location information may be used in determining real estate location and relative distance to landmarks), wherein the differential utility scores for the two or more differing dwelling attributes of the multiple candidate houses include negative utility scores and positive utility scores, and wherein the combining of the differential utility scores for each of the multiple candidate houses includes adding absolute values for the differential utility scores for the two or more differing dwelling attributes of the candidate house (Warnaar, Col. 19 Li. 65 – Col. 20 Li. 12, If real estate property 1404 includes a 2-car garage, the attribute may be awarded value is 2 and if real estate property 1404 includes a 3-car garage, the attribute awarded value remains 2. In the case of primary attribute 1518, the attribute metric is "not present", "partially meeting the goal", "fully meeting the goal", and "too many instances of the attribute present", with corresponding desirability. An example of this may be a user that desires real estate property 1404 which is a down-town real estate property. In this case the attribute of a number of parking garages in proximity to real estate property 1404 may be primary attribute 1502. If there is 1 parking garage 10 within a half-mile of real estate property 1404, the attribute value may decline because the specification was for at least 2, not 1 parking garage.). With respect to claim 4, Warnaar teaches a computer-implemented method comprising: determining, by one or more computing systems and for an indicated target dwelling having multiple dwelling attributes each with one or more indicated values, multiple candidate dwellings for comparison to the target dwelling, wherein the multiple candidate dwellings are located in one or more selected geographical areas associated with the target dwelling and each has one or more associated values for each of the multiple dwelling attributes (Warnaar, Col. 9 Li. 1-3, At step 304, seeking user's real estate needs (i.e., explicit needs and implicit needs) may be identified from the user data collected from seeking user 204. & Col. 9 Li. 11-12, At step 306, attributes of each real estate need of seeking user 204 may be identified & Col. 12 Li. 62-67, At step 616, the real estate attributes may be analyzed to identify explicit attributes 618 and to identify implicit attributes 620. Furthermore, geographic maps and images that show proximity of natural and man-made features and accurate traffic estimations can provide real estate attributes relative to a location & Col. 13 Li. 19-22, each proximity attribute is described by primary characteristics of static data such as footprint, height, type, age, and elevation difference with and distance to real estate property 702); determining, by the one or more computing systems, respective quantifications of differences between the multiple candidate dwellings and the target dwelling based on the multiple dwelling attributes (Warnaar, Col. 18 Li. 16-21, At step 1210, primary attributes may be identified in the set of properties selected in step 1208 where multiple properties in the set have primary attributes in common and the primary attributes make a significant contribution in the aggregate score of at least one property in the set.), by, for each of the multiple candidate dwellings: determining, for each of two or more dwelling attributes of the multiple dwelling attributes, that the one or more associated values of the dwelling attribute for the candidate dwelling differ from the one or more indicated values of the dwelling attribute for the target dwelling (Warnaar, Col. 19 Li. 54-61, each attribute in collection of primary attributes 1502 and collection of secondary attributes 1504 of real estate property 1404 may be assigned a single numeric score based on how well the attribute conforms to the metrics that indicate desirability. For example, attribute 1516 can have a metric of "unfulfilled", "partially fulfilled" or "fulfilled". The "fulfilled" state may not change no matter how many instances of the attribute there are.); determining for each of the two or more dwelling attributes, and using a predictive neural network model trained to provide overall utility scores for dwellings based on combinations of values of the multiple dwelling attributes for the dwellings, a differential utility score for the dwelling attribute of the candidate dwelling (Warnaar, Col. 20 Li. 13-21, the attributes, now prioritized by desirability, are associated with imputed weighted scores 1520. In some embodiments, the attributes, prioritized and weighted, may be further analyzed and combined into relevant groups of attributes by a decision network 1522. Decision network 1522 may be based on machine learning and artificial intelligence tools to create scoring that employs probability to predict user satisfaction. Decision network 1522 can result in a plurality of component scores 1524), wherein the differential utility score is: associated with a difference between the one or more indicated values of the dwelling attribute for the target dwelling and the one or more associated values of the dwelling attribute for the candidate dwelling; and represents a contribution of the dwelling attribute to an overall difference between an overall utility score for the candidate dwelling and an overall utility score for the target dwelling (Warnaar, Col. 19 Li. 61-67, As an example, the attribute may be a 2-car garage. If real estate property 1404 includes no garage, the awarded value may be 0. If real estate property 1404 includes a 1-car garage, the attribute may be awarded a value of 1. & Col. 20 Li. 20-24, Decision network 1522 can result in a plurality of component scores 1524 that are combined in a single score 1528. For example, decision network 1522 can determine the contributions of primary and secondary scores separately.); and combining the differential utility scores for the two or more dwelling attributes for the candidate dwelling (Warnaar, Col. 20 Li. 20-22, Decision network 1522 can result in a plurality of component scores 1524 that are combined in a single score 1528.); ranking the multiple candidate dwellings based on the respective quantifications of differences (Warnaar, Col. 21 Li. 32-36, A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204); and providing, by the one or more computing systems and as search results based on the target dwelling, at least some of the multiple candidate dwellings that are selected based on the ranking (Warnaar, Col. 14 Li. 34-36, real estate properties with implicit scores above a predetermined threshold may be presented to seeking user 204.). Warnaar doesn't expressly discuss determining for each of the two or more dwelling attributes, and using a predictive neural network model trained to provide overall utility scores for dwellings based on combinations of values of the multiple dwelling attributes for the dwellings, a differential utility score for the dwelling attribute of the candidate dwelling by determining and using differences between activation values of nodes of the trained predictive neural network model for the one or more associated values of the dwelling attribute for the candidate dwelling and for the one or more indicated values of the dwelling attribute for the target dwelling. Zhang teaches determining for each of the two or more dwelling attributes, and using a predictive neural network model trained to provide overall utility scores for dwellings based on combinations of values of the multiple dwelling attributes for the dwellings (Zhang, pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network that includes any number of layers, and which can use any activation function(s) (such as ReLU, etc.). The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. & pa 0091, generating (e.g., in block 708), using a ranking subsystem (e.g., ranking subsystem 128), plural ranking scores based on the plurality of instances of query-to-result information & pa 0068, FIG. 6 shows one non-limiting implementation of the second attention-based neural network 132. From top to bottom, a feed-forward neural network 602 can use any number of fully-connected layers to map a plurality of instances of enhanced similarity information into a plurality of instances of projected feature information.), a differential utility score for the dwelling attribute of the candidate dwelling by determining and using differences between activation values of nodes of the trained predictive neural network model for the one or more associated values of the dwelling attribute for the candidate dwelling and for the one or more indicated values of the dwelling attribute for the target dwelling (Zhang, pa 0040, pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network that includes any number of layers, and which can use any activation function(s) (such as ReLU, etc.). The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. Examiner note: activation functions of neural networks calculate the output of the node based on its individual inputs and their weights. A scoring neural network using the output of the second neural network as input and outputting a score therefrom would determine and use the differences of the nodes due to how neural network processing functions). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Zhang because it allows eliminates or reduces manually crafted features while enabling single pass computation for efficient use of computing resources (Zhang, pa 0006). With respect to claim 5, Warnaar in view of Wang teaches the computer-implemented method of claim 4 further comprising receiving, by the one or more computing systems and before the determining of the respective quantifications of differences between the multiple candidate dwellings and the target dwelling, a search query that indicates the target dwelling and one or more additional search criteria (Warnaar, Col. 9 Li. 1-3, At step 304, seeking user's real estate needs (i.e., explicit needs and implicit needs) may be identified from the user data collected from seeking user 204. & Col. 9 Li. 11-12, At step 306, attributes of each real estate need of seeking user 204 may be identified & Col. 9 Li. 34-35, At step 310, explicit preferences and implicit preferences of seeking user 204 for each attribute may be identified), and wherein at least one of the determining of the multiple candidate dwellings or the ranking of the multiple candidate dwellings is performed based at least in part on the one or more additional search criteria (Warnaar, Col. 20 Li. 30-31, calculated aggregate score may be indicative of the ability of real estate property 1404 to meet the user's real estate needs. & Col. 21 Li. 27-36, The process uses the score of each real estate property 1802 in the real estate collection 1804 of real estate properties offered for sale, rent, or occupation based on the customized profile of seeking user 204 and optimization as described in embodiments above. A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204). With respect to claim 6, Warnaar in view of Wang teaches the computer-implemented method of claim 4 wherein the one or more geographical areas are within a geographical region in which the target dwelling is located (Warnaar, Col. 11 Li. 10-16, The database of real estate attributes 428 may be obtained from the real estate database… information indicative of geographic locations is stored in geographic feature database (GFD)), wherein the target dwelling has a dwelling type that is one of a house, a home, an apartment, or a condominium (Warnaar, Col. 18 Li. 59-64, estate listing services may provide historical information entered by real estate professionals indicative of real estate property which, in some embodiments, may be a home. The historical information may comprise a year built, a size, a number of rooms, and a type of rooms of the home), wherein the determining of the multiple candidate dwellings includes selecting some dwellings in the one or more geographical areas that are of the dwelling type (Warnaar, Col. 12 Li. 62-67, At step 616, the real estate attributes may be analyzed to identify explicit attributes 618 and to identify implicit attributes 620. Furthermore, geographic maps and images that show proximity of natural and man-made features and accurate traffic estimations can provide real estate attributes relative to a location & Col. 13 Li. 19-22, each proximity attribute is described by primary characteristics of static data such as footprint, height, type, age, and elevation difference with and distance to real estate property 702). With respect to claim 7, Warnaar in view of Wang teaches the computer-implemented method of claim 6 wherein selecting some dwellings in the one or more geographical areas that are of the dwelling type includes selecting a subset of some of the dwellings in the one or more geographical areas that are of the dwelling type (Warnaar, Col. 12 Li. 62-67, At step 616, the real estate attributes may be analyzed to identify explicit attributes 618 and to identify implicit attributes 620. Furthermore, geographic maps and images that show proximity of natural and man-made features and accurate traffic estimations can provide real estate attributes relative to a location & Col. 13 Li. 19-22, each proximity attribute is described by primary characteristics of static data such as footprint, height, type, age, and elevation difference with and distance to real estate property 702), including, for one or more selected dwelling attributes of the multiple dwelling attributes, excluding dwellings in the one or more geographical areas that are of the dwelling type and have associated values for the one or more selected dwelling attributes that differ from the indicated values for the one or more selected dwelling attributes of the target dwelling (Warnaar, Col. 21 Li. 32-41, A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204 … seeking user 204 may set a threshold 1812 for the size of subset 1810, the minimum listing score, or any other criteria that limits the set of real estate properties offered to seeking user 204 for review.), and wherein the two or more dwelling attributes are different from the one or more selected dwelling attributes (Warnaar, Col. 12 Li. 35-61, real estate attributes). With respect to claim 8, Warnaar in view of Wang teaches the computer-implemented method of claim 4 wherein determining the differential utility score for each of the two or more dwelling attributes of each of the multiple candidate dwellings using the trained predictive neural network model includes identifying and using weighting data of the trained predictive neural network model associated with the dwelling attribute to determine the differential utility score for the dwelling attribute (Zhang, pa 0055, the second attention based neural network 132 packs the instances of enhanced similarity information into a matrix Eenhanced and multiples that same matrix Eenhanced by three respective machine trained weighting matrices (wQ2, WK2 , WV2) to produce the Q, K, and V matrices. & pa 0069, An attention processing component then maps the plurality of instances of projected feature information into a plurality of corresponding instances of attention feature information.). With respect to claim 9, Warnaar in view of Wang teaches the computer-implemented method of claim 4, wherein determining the differential utility score for each of the two or more dwelling attributes of each of the multiple candidate dwellings using the trained predictive neural network model includes determining and using differences between activation values of nodes of the trained predictive neural network model for the one or more associated values of that dwelling attribute for the candidate dwelling and for the one or more indicated values of the dwelling attribute for the target dwelling (Zhang, pa 0040, pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network that includes any number of layers, and which can use any activation function(s) (such as ReLU, etc.). The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. Examiner note: activation functions of neural networks calculate the output of the node based on its individual inputs and their weights. A scoring neural network using the output of the second neural network as input and outputting a score therefrom would determine and use the differences of the nodes due to how neural network processing functions). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Zhang because it allows eliminates or reduces manually crafted features while enabling single pass computation for efficient use of computing resources (Zhang, pa 0006). With respect to claim 10, Warnaar in view of Wang teaches the computer-implemented method of claim 4 wherein determining the differential utility score for each of the two or more dwelling attributes of each of the multiple candidate dwellings using the trained predictive neural network model further includes using at least one of: perturbating input to the trained predictive neural network model for a plurality of values of the dwelling attribute and determining the resulting differences in the overall utility score that is output by the trained predictive neural network model as part of determining the contribution of the dwelling attribute to the overall difference between the overall utility score for the candidate dwelling and the overall utility score for the target dwelling; using shapley values to determine the contribution of the dwelling attribute to the overall difference between the overall utility score for the candidate dwelling and the overall utility score for the target dwelling; or using conditional expectations for multiple values of the dwelling attribute to determine the contribution of the dwelling attribute to the overall difference between the overall utility score for the candidate dwelling and the overall utility score for the target dwelling (Warnaar, Col. 18 Li. 21-24, At step 1212, a subset of properties may be selected that have primary attributes in common and that have the most distinct aggregate score contribution in the set.). With respect to claim 11, Warnaar in view of Wang teaches the computer-implemented method of claim 4 wherein the predictive neural network model is trained to provide the overall utility scores for dwellings in the one or more geographical areas (Zhang, , pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network … The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. & pa 0091, The method then includes identifying (e.g., in block 710) at least a top-ranking candidate result item to be presented to a user based on the plural scores.), the method further comprising receiving a search query from a client device that indicates the target dwelling (Zhang, pa 0091, obtaining (e.g., in block 704) the query and the plurality of candidate result items that match the query, the query having one or more tokens, and each candidate result item having one or more tokens.), and wherein providing the at least some of the multiple candidate dwellings that are selected based on the ranking as the search results includes transmitting the search results over one or more computer networks to the client device for display (Zhang, pa 0091, The method then includes identifying (e.g., in block 710) at least a top-ranking candidate result item to be presented to a user based on the plural scores.). With respect to claim 12, Warnaar teaches a system comprising: one or more hardware processors of one or more computing devices (Warnaar, Col. 7 Li. 16, CPU); and one or more memories (Warnaar, Col. 7 Li. 18, RAM) with stored instructions that, when executed by the one or more hardware processors, cause the one or more computing devices to perform automated operations including: determining, for an indicated target dwelling having multiple dwelling attributes each with one or more indicated values, multiple candidate dwellings for comparison to the target dwelling, the multiple candidate dwellings each having one or more associated values for each of the multiple dwelling attributes (Warnaar, Col. 21 Li. 32-36, A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204); determining respective quantifications of differences between the multiple candidate dwellings and the target dwelling (Warnaar, Col. 19 Li. 54-61, each attribute in collection of primary attributes 1502 and collection of secondary attributes 1504 of real estate property 1404 may be assigned a single numeric score based on how well the attribute conforms to the metrics that indicate desirability. For example, attribute 1516 can have a metric of "unfulfilled", "partially fulfilled" or "fulfilled". The "fulfilled" state may not change no matter how many instances of the attribute there are.), by, for each of the multiple candidate dwellings: determining, for each of one or more dwelling attributes of the multiple dwelling attributes, that the one or more associated values of the dwelling attribute for that candidate dwelling differ from the one or more indicated values of the dwelling attribute for the target dwelling (Warnaar, Col. 20 Li. 1-12, In the case of primary attribute 1518, the attribute metric is "not present", "partially meeting the goal", "fully meeting the goal", and "too many instances of the attribute present", with corresponding desirability. An example of this may be a user that desires real estate property 1404 which is a down-town real estate property. In this case the attribute of a number of parking garages in proximity to real estate property 1404 may be primary attribute 1502. If there is 1 parking garage within a half-mile of real estate property 1404, the attribute value may decline because the specification was for at least 2, not 1 parking garage.); determining, for each of the one or more dwelling attributes, …, a differential utility score for the dwelling attribute of the candidate dwelling is associated with a difference between the one or more associated values of the dwelling attribute for the candidate dwelling and the one or more indicated values of the dwelling attribute for the target dwelling (Warnaar, Col. 20 Li. 13-17, the attributes, now prioritized by desirability, are associated with imputed weighted scores 1520. In some embodiments, the attributes, prioritized and weighted, may be further analyzed and combined into relevant groups of attributes by a decision network 1522.); and represents a contribution of that dwelling attribute to an overall difference between an overall utility score for the candidate dwelling and an overall utility score for the target dwelling (Warnaar, Col. 20 Li. 20-22, Decision network 1522 can result in a plurality of component scores 1524 that are combined in a single score 1528. For example, decision network 1522 can determine the contributions of primary and secondary scores separately); or generating the quantification of difference between that candidate dwelling and the target dwelling based on the differential utility scores for the one or more dwelling attributes for the candidate dwelling (Warnaar, Col. 21 Li. 25-36, Turning now to FIG. 18, where a process for matching real estate properties with a user is depicted and referred to generally by reference numeral 1800. The process uses the score of each real estate property 1802 in the real estate collection 1804 of real estate properties offered for sale, rent, or occupation based on the customized profile of seeking user 204 and optimization as described in embodiments above. A rank ordered list 1808 of real estate properties may be generated in the real estate collection 1804 such that a subset 1810 of real estate properties at top of rank ordered list 1808 achieves the highest and best use of real estate currently offered for seeking user 204); and providing indications of at least some of the multiple candidate dwellings that are selected based on the respective quantifications of differences for the at least some candidate dwellings (Warnaar, Col. 21 Li. 38-41, seeking user 204 may set a threshold 1812 for the size of subset 1810, the minimum listing score, or any other criteria that limits the set of real estate properties offered to seeking user 204 for review.). Warnaar doesn't expressly discuss using a predictive neural network model trained to provide overall utility scores for dwellings based on combinations of values of the multiple dwelling attributes for the dwellings, a differential utility score for the dwelling attribute of the candidate dwelling by perturbating input to the trained predictive neural network model for a plurality of values of the dwelling attribute and determining resulting differences in the overall utility score output by the trained predictive neural network model. Wang teaches using a predictive neural network model trained to provide overall utility scores for dwellings based on combinations of values of the multiple dwelling attributes for the dwellings (Zhang, pa 0068, FIG. 6 shows one non-limiting implementation of the second attention-based neural network 132. From top to bottom, a feed-forward neural network 602 can use any number of fully-connected layers to map a plurality of instances of enhanced similarity information into a plurality of instances of projected feature information.), a differential utility score for the dwelling attribute of the candidate dwelling (Zhang, pa 0040, the ranking subsystem 128 includes a second attention-based neural network 132 in combination with a scoring neural network 134… the scoring neural network 134 can correspond to a fully-connected neural network that includes any number of layers, and which can use any activation function(s) (such as ReLU, etc.). The scoring neural network 134 maps the output of the second attention-based neural network 132 into the plural ranking scores. & pa 0091, generating (e.g., in block 708), using a ranking subsystem (e.g., ranking subsystem 128), plural ranking scores based on the plurality of instances of query-to-result information) by perturbating input to the trained predictive neural network model for a plurality of values of the dwelling attribute and determining resulting differences in the overall utility score output by the trained predictive neural network model (Zhang, pa 0044, The training system 136 iteratively modifies its model(s) in the course of processing a set of training examples with the objective of optimizing the above-described NDCG@ 1 metric. Once fully trained, the resultant machine-trained model(s) will enable the ranking system 104 to correctly identify the top-ranking result item in most cases.). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Zhang because it allows eliminates or reduces manually crafted features while enabling single pass computation for efficient use of computing resources (Zhang, pa 0006). With respect to claim 13, Warnaar in view of Wang teaches the system of claim 12, the automated operations further including receiving a search query that indicates the target dwelling (Warnaar, Col. 9 Li. 1-3, At step 304, seeking user's real estate needs (i.e., explicit needs and implicit needs) may be identified from the user data collected from seeking user 204. & Col. 9 Li. 11-12, At step 306, attributes of each real estate need of seeking user 204 may be identified & Col. 9 Li. 34-35, At step 310, explicit preferences and implicit preferences of seeking user 204 for each attribute may be identified). Warnaar doesn't expressly discuss wherein providing the indications of the at least some candidate dwellings includes presenting, in a displayed graphical user interface, search results for the search query that include information about the at least some candidate dwellings. Zhang teaches wherein providing the indications of the at least some candidate dwellings includes presenting, in a displayed graphical user interface, search results for the search query that include information about the at least some candidate dwellings (Zhang, pa 0091, The method then includes identifying (e.g., in block 710) at least a top-ranking candidate result item to be presented to a user based on the plural scores.). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Zhang because it allows eliminates or reduces manually crafted features while enabling single pass computation for efficient use of computing resources (Zhang, pa 0006). With respect to claim 14, Warnaar in view of Wang teaches the system of claim 12 wherein the one or more dwelling attributes for each of the at least some candidate dwellings include two or more dwelling attributes, the automated operations further including selecting each of the at least some candidate dwellings based on a combination of the differential utility scores for the two or more dwelling attributes for the candidate dwelling (Warnaar, Col. 21 Li. 38-41, seeking user 204 may set a threshold 1812 for the size of subset 1810, the minimum listing score, or any other criteria that limits the set of real estate properties offered to seeking user 204 for review.). With respect to claim 15, Warnaar in view of Wang teaches the system of claim 12, the automated operations further including at least one of: selecting dwellings that include the at least some candidate dwellings and providing the indications about the at least some candidate dwellings by presenting the selected dwellings as a group of related dwellings; or excluding a subset of the multiple candidate dwellings that are not the at least some candidate dwellings and providing the indications about the at least some candidate dwellings by presenting only the at least some candidate dwellings as available options for filtering or selection (Warnaar, Col. 14 Li. 34-36, real estate properties with implicit scores above a predetermined threshold may be presented to seeking user 204.). With respect to claim 16, Warnaar in view of Wang teaches the system of claim 12, the automated operations further including determining, for each of at least one of the multiple dwelling attributes, an aggregate utility score for at least one value of the dwelling attribute by combining determined utility scores for the dwelling attribute from two or more of the multiple candidate dwellings (Warnaar, Col. 18 Li. 16-21, At step 1210, primary attributes may be identified in the set of properties selected in step 1208 where multiple properties in the set have primary attributes in common and the primary attributes make a significant contribution in the aggregate score of at least one property in the set.), and providing one or more indications of the determined aggregate utility score for the at least one value of each of the at least one dwelling attributes (Warnaar, Col. 14 Li. 34-36, real estate properties with implicit scores above a predetermined threshold may be presented to seeking user 204.). With respect to claim 17, Warnaar in view of Wang teaches the system of claim 12, wherein the target dwelling has a dwelling type that is one of a house, a home, an apartment, or a condominium (Warnaar, Col. 18 Li. 59-64, estate listing services may provide historical information entered by real estate professionals indicative of real estate property which, in some embodiments, may be a home. The historical information may comprise a year built, a size, a number of rooms, and a type of rooms of the home), wherein determining the multiple candidate dwellings includes selecting candidate dwellings that are in a geographical region associated with the target dwelling and that each are of the dwelling type (Warnaar, Col. 12 Li. 62-67, At step 616, the real estate attributes may be analyzed to identify explicit attributes 618 and to identify implicit attributes 620. Furthermore, geographic maps and images that show proximity of natural and man-made features and accurate traffic estimations can provide real estate attributes relative to a location & Col. 13 Li. 19-22, each proximity attribute is described by primary characteristics of static data such as footprint, height, type, age, and elevation difference with and distance to real estate property 702). With respect to claims 18, 19, 21, and 23, the limitations are essentially the same as 12-14 and 17, and are rejected for the same reasons. Claim 22 is rejected under 35 U.S.C. 103 as being unpatentable over Warnaar in view of Wang, and further in view of Carney et al. (US 2025/0021882). With respect to claim 22, Warnaar in view of Wang teaches the non-transitory computer-readable medium of claim 18, as discussed above. Warner teaches the automated operations further comprising associating values of the multiple dwelling attributes with overall utility scores of dwellings by generating a linear model and using normalized coefficients of the linear model (Carney, pa 0082, One way of achieving this is by training a simpler model (like a linear regression) on the outputs of the predictive model. The coefficients of this simpler model can then be interpreted as a measure of the importance of a feature's presence for the likelihood of success). It would have been obvious at the effective filing date of the invention to a person having ordinary skill in the art to which said subject matter pertains to have modified Warnaar with the teachings of Carney because it allows indicating of the performance of a particular metric when predicting outcomes (Carney, pa 0082). Response to Amendment 35 U.S.C. 112 With regard to claims 1-3, the amendments to the claims have overcome the 35 U.S.C. 112 rejection. The Examiner withdraws the 35 U.S.C. 112 rejection to claims 1-3. Response to Arguments 35 U.S.C. 103 With respect to claims 1-19 and 21-23, Applicant seems to argue a newly amended limitation. Applicant’s amendment has rendered the previous rejection moot. Upon further consideration of the amendment, a new grounds of rejection is made in view of Zhang et al. (US 2022/0277015). 35 U.S.C. 101 Applicant argues that the amendments clarifying specific neural network internal operations for determining differential scores involve accessing and manipulating the internal architecture of a neural network, providing specific computational operations that cannot be practically performed in human mind or with pen and paper. The Examiner respectfully disagrees. Claim 1 does not require manipulating the internal architecture of a neural network but rather identifying and using weighting data internal to the neural network model. The neural network claimed is recited as a high-level of generality such that it amounts to no more than mere instructions to apply the exception. The “determining” by “identifying and using” is still a mental processes of identifying and using data under Step 2A Prong One. Applicant argues that the specific use of the internal structure of a trained neural network model integrates the alleged abstract idea into a practical application. The Examiner respectfully disagrees. Under Step 2A Prong Two, the recited abstract idea should be integrated into a practical application through recitation of additional elements. While the recited neural network is an element additional to the abstract idea, it is recited as a high-level of generality such that it amounts to no more than mere instructions to apply the exception. Even though the claim recites weighting data internal to the model as being identified and used, this is describing how neural networks are typically used and does not provide an improvement in the functioning of a computer, or an improvement to other technology or technical field. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRITTANY N ALLEN whose telephone number is (571)270-3566. The examiner can normally be reached M-F 9 am - 5:00 pm EST. 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, Sherief Badawi can be reached at 571-272-9782. 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. /BRITTANY N ALLEN/ Primary Examiner, Art Unit 2169
Read full office action

Prosecution Timeline

Show 4 earlier events
Jan 15, 2026
Applicant Interview (Telephonic)
Jan 22, 2026
Response Filed
Mar 06, 2026
Final Rejection mailed — §101, §103
Jun 03, 2026
Request for Continued Examination
Jun 05, 2026
Response after Non-Final Action
Aug 18, 2026
Non-Final Rejection mailed — §101, §103
Nov 17, 2026
Applicant Interview (Telephonic)
Nov 17, 2026
Examiner Interview Summary

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12783557
SKILL-CENTRIC ENTITY EMBEDDING AND RANKING
2y 8m to grant Granted Nov 03, 2026
Patent 12769759
MULTI-CLUSTER DATABASE MANAGEMENT SYSTEM
1y 11m to grant Granted Oct 20, 2026
Patent 12737358
Query Plan Adaptation Using Query Plan Fragments
2y 7m to grant Granted Sep 15, 2026
Patent 12711105
THREAD SAFE LOCK-FREE CONCURRENT WRITE OPERATIONS FOR USE WITH MULTI-THREADED IN-LINE LOGGING
2y 10m to grant Granted Aug 18, 2026
Patent 12705232
SYSTEMS AND METHODS FOR MANAGING OFFLINE DATABASE ACCESS
2y 11m to grant Granted Aug 11, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
42%
Grant Probability
80%
With Interview (+37.8%)
4y 4m (~2y 4m remaining)
Median Time to Grant
High
PTA Risk
Based on 403 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month