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 .
The following is a Final Office Action in response to communications received on 12/12/2025. Claims 21-26 and 30-43 are currently pending and have been examined. Claims 21 and 40 have been amended. Claims 1-20 and 27-29 are cancelled. Claims 41-43 have been added.
Priority
Applicant's claim for the benefit of a prior-filed application under 35 U.S.C. 121 is acknowledged. Applicant has not complied with one or more conditions for receiving the benefit of an earlier filing date under 35 U.S.C. 121 as follows:
The later-filed application must be an application for a patent for an invention which is also disclosed in the prior application (the parent or original non-provisional application or provisional application). The disclosure of the invention in the parent application and in the later-filed application must be sufficient to comply with the requirements of the first paragraph of 35 U.S.C. 112. See Transco Products, Inc. v. Performance Contracting, Inc., 38 F.3d 551,32 USPQ2d 1077 (Fed. Cir. 1994). It is Newly added claims 42-43 are not supported in the provisional application 62/837064, however are supported in the parent applications 16/855929 and 17/308674 and such have been treated with the priority date of 4/22/2020. All other pending claims are supported in the provisional and given the priority date of 4/22/2019.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 21, 22, 23, 30, 35, 36, and 38-40 rejected on the ground of nonstatutory double patenting as being unpatentable over claims 21, 22, 23, 24, 25, 30, and 35-40 rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1, 8, 10, 12, and 21 of U.S. Patent No. US PAT 11922370 in view of Fredrich US 20180144389.
Although the claims at issue are not identical, they are not patentably distinct from each other because the claim limitations of the instant application are obvious over U.S. Patent No. 11922370 in view of Fredrich (US 20180144389) by the claim limitations of the patent as shown in the comparison below.
Instant Application
US PAT 11922370
21. (New) One or more non-transitory computer-readable storage media comprising computer program instructions that when executed by one or more processors effectuate operations comprising:
1. A non-transitory computer-readable storage medium comprising computer program instructions that when executed by one or more processors effectuate operations comprising:
obtaining, in a database, a plurality of first user records, each first user record comprising past consumption amounts for goods;
obtaining, in a database, a plurality of user records, each record comprising household properties having corresponding values and consumption amounts for goods;
forming, from the plurality of first user records, a training set;
forming, from the plurality of user records, a training set and a validation set;
training, using the training set, a consumption model configured to output predicted future consumption amounts for respective goods represented in the training set, wherein the consumption model comprises a machine learning model;
training, based on the training set, a consumption model configured to output, based on input values for a set of a plurality of household properties represented in the training records, determined consumption amounts for respective goods represented in the training records, wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the user records in the validation set;
obtaining a second user record comprising consumption data for a user;
obtaining a new user record comprising at least some household properties having corresponding values;
determining, with the consumption model based on the second user record, a predicted consumption amount for each good in a set of goods;
determining, with the consumption model based on the new user record, a predicted consumption amount for each good in a set of goods;
obtaining, for a good in the set of goods, at least one stock record indicative of at least one identifier having a count of the good that satisfies in whole or in part the predicted consumption amount for the good, wherein at least one identifier is obtained for each good in the set of goods;
obtaining, for a good in the set of goods, at least one stock record indicative of at least one stock keeping unit (SKU) having a count of the good that satisfies in whole or in part the predicted consumption amount for the good, wherein at least one SKU is obtained for each good in the set of goods and a combination of SKUs is obtained for at least some of the goods in the set of goods;
generating, for the set of goods, an order comprising a plurality of identifiers indicated in obtained stock records for which a total count of one or more identifiers corresponding to respective ones of the goods meet the predicted consumption amount for the good without exceeding a threshold; and
generating, for the set of goods, an order comprising a plurality of SKUs indicated in obtained stock records for which a total count of one or more SKUs corresponding to respective ones of the goods meet the predicted consumption amount for the good without exceeding a threshold; and
submitting the generated order for fulfillment.
submitting the generated order for fulfillment.
22. (New) The media of claim 21, the operations comprising:
determining a first feedback prompt corresponding to a first good in the set of goods; and
updating the second user record based on received user feedback responsive to the prompt.
8. The medium of claim 1, the operations further comprising:
updating the new user record based on the generated order;
determining a first feedback prompt corresponding to a first good in the set of goods;
updating the new user record based on received user feedback responsive to the prompt;
determining, with the consumption model based on the updated user record, a next predicted consumption amount for each good in a set of goods during a next consumption period;
updating the new user record based on a next generated order submitted for fulfilment for the next consumption period; and
determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount.
23. (New) The media of claim 22, the operations comprising:
determining a measure of accuracy of current consumption amounts of goods determined for the updated second user record based at least in part on the user feedback;
and identifying the updated second user record to the training set based on the measure of accuracy.
12. The medium of claim 1, the operations further comprising:
determine one or more prompts for user feedback corresponding to household properties different from the at least some household properties included in the new user record;
transmitting the one or more prompts for user feedback to a user device of the user associated with the new user record;
updating the new user record based on received user feedback:
determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback;
and identifying the updated user record to the training set or the validation set based on the measure of accuracy.
24. (New) The media of claim 22, the operations comprising:
determining, with the consumption model based on the updated second user record, a next predicted consumption amount for each good in a second set of goods for a next consumption period; and
updating the second user record based on a next generated order submitted for fulfilment for the next consumption period.
8. The medium of claim 1, the operations further comprising:
updating the new user record based on the generated order;
determining a first feedback prompt corresponding to a first good in the set of goods;
updating the new user record based on received user feedback responsive to the prompt;
determining, with the consumption model based on the updated user record, a next predicted consumption amount for each good in a set of goods during a next consumption period;
updating the new user record based on a next generated order submitted for fulfilment for the next consumption period;
and determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount.
25. (New) The media of claim 24, the operations comprising:
determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount.
Claim 8 continued
and determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount.
30. (New) The media of claim 21, the operations comprising:
determining a first feedback prompt corresponding to a first good in the set of goods;
receiving user feedback responsive to the prompt; and
determining the predicted consumption amount for the good based in part on the user feedback.
8. The medium of claim 1, the operations further comprising:
updating the new user record based on the generated order;
determining a first feedback prompt corresponding to a first good in the set of goods;
updating the new user record based on received user feedback responsive to the prompt;
determining, with the consumption model based on the updated user record, a next predicted consumption amount for each good in a set of goods during a next consumption period;
updating the new user record based on a next generated order submitted for fulfilment for the next consumption period;
and determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount.
35. (New) The media of claim 21, the operations comprising:
steps for forming the training set from the first plurality of user records.
Claim 1 continued
forming, from the plurality of user records, a training set and a validation set;
36. (New) The media of claim 21, the operations comprising:
forming a validation set from the first plurality of user records.
forming, from the plurality of user records, a training set and a validation set;
37. (New) The media of claim 36, the operations comprising:
determining one or more prompts for user feedback corresponding to household properties different from at least some household properties included in the second user record;
transmitting the one or more prompts for user feedback to a user device of the user;
updating the second user record based on received user feedback:
determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and
identifying the updated user record to the training set or the validation set based on the measure of accuracy.
12. The medium of claim 1, the operations further comprising:
determine one or more prompts for user feedback corresponding to household properties different from the at least some household properties included in the new user record;
transmitting the one or more prompts for user feedback to a user device of the user associated with the new user record;
updating the new user record based on received user feedback:
determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and
identifying the updated user record to the training set or the validation set based on the measure of accuracy.
38. (New) The media of claim 36, wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the first user records in the validation set.
10. The medium of claim 1, the operations further comprising:
training a plurality of consumption models;
determining that a given one of the consumption models minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some different collections of user records in the validation set, wherein the different collections of user records correspond to different clusters of user records; and
determining, based on the respective values of the at least some household properties in the new user record and respective values representative of the at least some household properties of the different clusters, distances between the user record and the different clusters to identify a closest one of the clusters based on the distances.
39. (New) The media of claim 21, wherein the identifiers comprise stock keeping units (SKUs).
Claim 1 continued
at least one stock record indicative of at least one stock keeping unit (SKU)
Claim 40 recites parallel claim language of a method
Claim 21 recites parallel claim language of a method
US PAT 11922370 does not disclose wherein the consumption model comprises machine learning. However Fredrich teaches [0113] Information feedback predictor 914a may be configured to implement any analytical determination to correlate and classify users and/or items for predicting samples, according to various examples. According to some embodiments, information feedback predictor 914a may be configured to classify and/or quantify various user and item attributes by, for example, applying machine learning or deep learning techniques, or the like. In one example, information feedback predictor 914a may be configured to segregate, separate, or distinguish a number of data points representing similar (or statistically similar) attributes, thereby forming one or more clusters or groups of data (not shown). The clustered data may be grouped or clustered about a particular attribute of the data, such as a source of data (e.g., a channel of data), a type of language, a degree of similarity with synonyms or other words, etc., or any other attribute, characteristic, parameter or the like. While any number of techniques may be implemented, information feedback predictor 914a may apply “k-means clustering,” or any other known clustering data identification techniques. In some examples, information feedback predictor 914a may be configured to detect patterns or classifications among datasets and other data through the use of Bayesian networks, clustering analysis, as well as other known machine learning techniques or deep-learning techniques (e.g., including any known artificial intelligence techniques, or any of k-NN algorithms, regression, Bayesian inferences and the like, including classification algorithms, such as Naïve Bayes classifiers, or any other statistical or empirical technique). Also see [0114, 0115, 0118 and 0119]. Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the training of US PAT 11922370 to include wherein the consumption model comprises machine learning, as taught in Fredrich, in order to detect patterns or classifications among datasets.
For these reasons it is determined while the conflicting claims are not identical, at least one examined application claim is not patentably distinct from the reference claims(s) because the examined application claim is obvious over US PAT 11922370 in view of Fredrich as shown above.
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.
Step 1: The claims 21-26 and 30-39 are a computer readable medium and claim 40 is a method. Thus, each independent claim, on its face, is directed to one of the statutory categories of 35 U.S.C. §101. However, the claims 21-26 and 30-43 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 2A Prong 1: The independent claims (21 and 40, taking claim 21 as a representative claim) recite:
One or more non-transitory computer-readable storage media comprising computer program instructions that when executed by one or more processors effectuate operations comprising: (claim 21)
obtaining, in a database, a plurality of first user records, each first user record comprising past consumption amounts for goods;
forming, from the plurality of first user records, a training set;
training, using the training set, a consumption model configured to output predicted future consumption amounts for respective goods represented in the training records, wherein the consumption model comprises a machine learning model;
obtaining a second user record comprising consumption data for a user;
determining, with the consumption model based on the second user record, a predicted consumption amount for each good in a set of goods;
obtaining, for a good in the set of goods, at least one stock record indicative of at least one identifier having a count of the good that satisfies in whole or in part the predicted consumption amount for the good, wherein at least one identifier is obtained for each good in the set of goods;
generating, for the set of goods, an order comprising a plurality of identifiers indicated in obtained stock records for which a total count of one or more identifiers corresponding to respective ones of the goods meet the predicted consumption amount for the good without exceeding a threshold; and
submitting the generated order for fulfillment.
These limitations, except for the italicized portions, under their broadest reasonable interpretations, recite certain methods of organizing human activity for managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) as well as commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations). The claimed invention recites steps determining a predicted consumption amount of a set of goods, identifying for a set of goods at least one stock record, determining an amount to order based on the predicted consumption and submitting the order to be filled. The steps under its broadest reasonable interpretation specifically fall under sales activities. The Examiner notes that although the claim limitations are summarized, the analysis regarding subject matter eligibility considers the entirety of the claim and all of the claim elements individually, as a whole, and in ordered combination.
Prong 2: This judicial exception is not integrated into a practical application. In particular, the claims recite the additional elements of
One or more non-transitory computer-readable storage media comprising computer program instructions that when executed by one or more processors effectuate operations comprising: (claim 21)
A computer-implemented method comprising: (claim 40)
in a database
training, using the training set, a consumption model, wherein the consumption model comprises a machine learning model;
The emphasized additional elements above are recited at a high-level of generality (i.e., as a generic processor performing a generic computer function of processing data) such that it amounts no more than mere instructions to apply the exception using a generic computer component. The limitations do not impose any meaningful limits on practicing the abstract idea, and therefore do not integrate the abstract idea into a practical application – MPEP 2106.05(f).
Accordingly, these additional elements when considered individually or as a whole do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The independent claims are directed to an abstract idea.
Step 2B: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed with respect to Step 2A Prong two, the additional elements in the claims amount to no more than mere instructions to apply the judicial exception using a generic computer component.
Even when considered as an ordered combination, the additional elements of claims 21 and 40 do not add anything that is not already present when they are considered individually. Therefore, under Step 2B, there are no meaningful limitations in claims 21 and 40 that transform the judicial exception into a patent eligible application such that the claims amount to significantly more than the judicial exception itself (see MPEP 2106.05).
As such, independent claims 21 and 40 are ineligible.
Dependent claims 22-26, 30-43 when analyzed as a whole, are held to be patent ineligible under 35 U.S.C. §101 because the additional recited limitations fail to establish that the claims are not directed to the same abstract idea of Independent Claims 21 and 40 without significantly more.
Claim 22 recites the operations comprising: determining a first feedback prompt corresponding to a first good in the set of goods; and updating the second user record based on received user feedback responsive to the prompt. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 23 recites the operations comprising: determining a measure of accuracy of current consumption amounts of goods determined for the updated second user record based at least in part on the user feedback; and identifying the updated second user record to the training set based on the measure of accuracy. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 24 recites the operations comprising: determining, with the consumption model based on the updated second user record, a next predicted consumption amount for each good in a second set of goods for a next consumption period; and updating the second user record based on a next generated order submitted for fulfilment for the next consumption period. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 25 recites the operations comprising: determining a second feedback prompt corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 26 recites the operations comprising: receiving an input from the user; and generating the order based at least in part on the input. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 30 recites the operations comprising: determining a first feedback prompt corresponding to a first good in the set of goods; receiving user feedback responsive to the prompt; and determining the predicted consumption amount for the good based in part on the user feedback. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 31 recites determining that predicted consumption amounts, including the predicted consumption amount, are more accurate when the predicted consumption amounts are based on user responses, including the user feedback, to the first feedback prompt; and based on determining that the predicted consumption amounts are more accurate, increasing a frequency at which the first feedback prompt is provided. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 32 recites the operations comprising: determining, based on the received user feedback, that a household of the user has diverged from past consumption habits. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 33 recite the operations comprising: determining additional feedback prompts based on the divergence from the past consumption habits. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 34 recites the operations comprising: providing the additional feedback prompts to the user; receiving additional feedback responsive to the additional feedback prompts; and generating and submitting one or more additional orders based at least in part on the additional feedback. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 35 recites the operations comprising: steps for forming the training set from the first plurality of user records. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 36 recites the operations comprising: forming a validation set from the first plurality of user records. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 37 recites the operations comprising: determining one or more prompts for user feedback corresponding to household properties different from at least some household properties included in the second user record; transmitting the one or more prompts for user feedback to a user device of the user; updating the second user record based on received user feedback: determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and identifying the updated user record to the training set or the validation set based on the measure of accuracy. The limitation merely further limits the abstract idea and recites the user device at a high level of generality. Therefore it does not integrate the judicial exception into a practical application.
Claim 38 recites wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the first user records in the validation set. The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 39 recites wherein the identifiers comprise stock keeping units (SKUs). The limitation merely further limits the abstract idea and does not integrate the judicial exception into a practical application.
Claim 41 recites the operations further comprising: training a second machine learning model based on orders received from the user; and determining, with the second machine learning model, a predicted consumption amount for a second good; and submitting a generated order for fulfillment based on the predicted consumption amount for the second good. The limitations recite the additional elements of the second machine learning model, however is recited at a high level of generality and does not integrate the judicial exception into a practical application.
Claim 42 recites wherein the machine learning model comprises a feedforward neural network (FFNN). The limitations recite the additional elements of the FFNN, however is recited at a high level of generality and does not integrate the judicial exception into a practical application.
Claim 43 recites the operations comprising: processing an input feature vector with the FFNN, wherein: the input feature vector comprises information regarding previous purchases of the user, and determining the predicted consumption amount for the good in the set of goods is based on the input feature vector, including the information regarding the previous purchases of the user. The limitations recite the additional elements of the FFNN, however is recited at a high level of generality and does not integrate the judicial exception into a practical application.
For at least these reasons claims 21-26 and 30-43 are rejected under 35 USC 101.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 21-26, 30, 31, 35, 36, 39, and 40 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Fredrich (US 20180144389).
Regarding claims 21 and 40, Fredrich discloses:
One or more non-transitory computer-readable storage media comprising computer program instructions that when executed by one or more processors effectuate operations comprising: (claim 21) [0025]
A computer-implemented method comprising (claim 40)
obtaining, in a database, a plurality of first user records, each first user record comprising past consumption amounts for goods; [0038] User profile and account data may be stored in user repository 170, which may include data relating to one or more users 144, or stored in a platform repository 172, which may include data relating to any aspect of data transactions among users 144. [0040] For example, distribution optimizer 118 may be configured to analyze data representing purchasing patterns related to a particular item for a specific user 144. and see [0117]
forming, from the plurality of first user records, a training set; training, using the training set, a consumption model configured to output predicted future consumption amounts for respective goods represented in the training records; [0080] To illustrate, consider that item characteristic correlator 717 is configured to identify usage rates 752a to 752c (e.g., a rate at which a product or service is reordered, or consumed or depleted) for corresponding user accounts 742a to 742c (e.g., associated with user phone numbers). In this example, consider that users 742a to 742c purchase a “laundry detergent” having usage rates 752a to 752c. Usage rates between “0” and “1” (e.g., usage amounts during a spring season), usage rates between “1” and “2” (e.g., usage amounts during a summer season), and usage rates between “2” and “3” (e.g., usage amounts during a fall season). It may be that users 742a to 742c play football during the fall, and consequently use more laundry detergent due to football practices and games in inclement weather (e.g., due to muddy fields, etc.) Thus, distribution predictor 714 may be able to discern patterns 750 of usage. Further, distribution predictor 714 may aggregate the usage rates to form an aggregated usage rate pattern 730 for a group of users 740. Based on aggregated usage rate pattern 730, distribution predictor 714 may be able to generate or predict an aggregated distribution event or an aggregated time of distribution 790, wherein the consumption model comprises a machine learning model; [0113] Information feedback predictor 914a may be configured to implement any analytical determination to correlate and classify users and/or items for predicting samples, according to various examples. According to some embodiments, information feedback predictor 914a may be configured to classify and/or quantify various user and item attributes by, for example, applying machine learning or deep learning techniques, or the like. In one example, information feedback predictor 914a may be configured to segregate, separate, or distinguish a number of data points representing similar (or statistically similar) attributes, thereby forming one or more clusters or groups of data (not shown). The clustered data may be grouped or clustered about a particular attribute of the data, such as a source of data (e.g., a channel of data), a type of language, a degree of similarity with synonyms or other words, etc., or any other attribute, characteristic, parameter or the like. While any number of techniques may be implemented, information feedback predictor 914a may apply “k-means clustering,” or any other known clustering data identification techniques. In some examples, information feedback predictor 914a may be configured to detect patterns or classifications among datasets and other data through the use of Bayesian networks, clustering analysis, as well as other known machine learning techniques or deep-learning techniques (e.g., including any known artificial intelligence techniques, or any of k-NN algorithms, regression, Bayesian inferences and the like, including classification algorithms, such as Naïve Bayes classifiers, or any other statistical or empirical technique). Also see [0114, 0115, 0118 and 0119]
obtaining a second user record comprising consumption data for a user; determining, with the consumption model based on the second user record, a predicted consumption amount for each good in a set of goods; [0082] Based on the above, distribution predictor 714 may be configured to identify a usage rate 720 of a new user 799, and further configured to match the new usage rate 720 against aggregated usage rate 729 of aggregated usage rate pattern 730 to predict, for example, that user 799 “plays football,” as well as other characteristics of the user with which to derive an optimized predicted time of distribution. According to some examples, distribution predictor 714 may predict future participation in an activity or an increase in usage rate during interval 728. Thus, distribution predictor 714 may adapt a predicted time of distribution so as to prepare a user for increased usage rates by adjusting the periods of time prior to a modified time of distribution to reflect an increased laundry detergent amount or a decreased amount of time between shipments. Note that the example described in diagram 700 is not intended to be limiting to laundry detergent, but may be applicable to any characteristic of an item or other items.
obtaining, for a good in the set of goods, at least one stock record indicative of at least one identifier having a count of the good that satisfies in whole or in part the predicted consumption amount for the good, wherein at least one identifier is obtained for each good in the set of goods; [0040] Distribution predictor 114 may include a distribution calculator 116, a distribution optimizer 118, and a zone generator 119. Distribution calculator 116 may be configured to calculate one or more predicted distribution events or replenishment-related data to form an adaptive schedule (e.g., an adaptive shipping schedule). Distribution optimizer 118 may be configured to optimize values of predicted distribution events to, for example, adapt scheduling of distributed items (i.e., product shipments) to conform (or substantially conform) to delivery or usage preferences of user 144 or a group of users 144. For example, distribution optimizer 118 may be configured to analyze data representing purchasing patterns related to a particular item for a specific user 144. Based on the results of such an analysis, distribution optimizer 118 may be configured to emphasize certain item characteristics (or values thereof) that may align more closely to a user's ordering or reordering patterns. For example, replenishment of an exact brand name at a later date may be preferred by user 144 over substitution of a comparable other brand at an earlier date. Zone generator 119 may be configured to define a zone of time, which may be configurable or adjusted based on, for example, one or more of user preferences, an amount of time since a prior distribution (e.g., a prior purchase), one or more usage rates, units of depletion or depletion rate, etc. An example of a depletion rate is the rate at which 2 units of a product are depleted per unit time. To illustrate another example, consider that a bottle of vitamins has 180 tablets and is reordered or depleted every 72 days. Thus, a predicted rate of depletion may be 2.5 per day (e.g., 180 units/72 days). For example, a user may consume 2 to 3 tablets per every other day). [0041] Distribution calculator 116 may be configured to receive data representing item characteristics data 102, according to some embodiments, and may be configured further to determine (e.g., identify, calculate, derive, etc.) one or more distribution events based on one or more item characteristics 102, or combinations thereof (e.g., based on derived item characteristics).
generating, for the set of goods, an order comprising a plurality of identifiers indicated in obtained stock records for which a total count of one or more identifiers corresponding to respective ones of the goods meet the predicted consumption amount for the good without exceeding a threshold; and [0043] A predicted distribution event for an item may be based on a usage rate of the item (e.g., a calculated usage rate), whereby a usage rate may be a rate at which a product or service is distributed (e.g., ordered or reordered), consumed, or depleted. In one example, predicted distribution of an item for a user 144 may be based on a predicted time of exhaustion, such as exemplified in the above example in which a distribution event for a bottle of 30 vitamin tablets is predicted to occur at the 30th day (e.g., relative to a previous delivery). In another example, predicted distribution of an item for user 144 may be based on the user's pattern of purchasing, using, ordering, or reordering the item (or generically similar or complementary items). For example, a predicted time of distribution to replenish an item, such as a bottle of ketchup, may be based on a user's past rates of replenishment (e.g., shipment rates), such as a median or average time between successive requests to distribute reordered items. A distribution event may be predicted or supplemented by predicting a time of distribution for ketchup based on rates of past replenishment of mustard, a complementary product having a usage rates that may correlate to that of ketchup as both items may be used together (and thus consumed at similar depletion rates). Other users' patterns of purchasing, using, ordering, or reordering of the same item (e.g., same brand of vitamin A at the same merchant) or equivalent item (e.g., different brands of vitamin A at the same or different merchants) may also be used to predict a time of distribution. For example, consider that user 144 is replenishing an item, such a vitamin A tablets, at a merchant X. However, there may be negligible information to predict a usage rate (or a time of distribution) for that item at merchant X. Therefore, other users' patterns of reordering vitamin A at another merchant, merchant Y, may be used to form a predicted time of distribution for use in purchasing vitamin A tablets at merchant X. and see [0044-45]
submitting the generated order for fulfillment. [0055] Responsive to receiving “YES,” conversation platform controller 115 may generate a control message 121 for transmission to, for example, merchant computing system 130a to initiate completion of the transaction.
The examiner notes as set forth in [0070], the distribution predictor 314 may be configured to receive and/or determine data for one or more item characteristics that may include, but are not limited to, data representing one or more characteristics of an item, shipment rate-related data 302, indicator-related data 304, and usage-related data 308, and the like. The analysis is not limited to a single item or single iteration.
Regarding claim 22, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining a first feedback prompt (see example prompt in Fig. 1 152b) corresponding to a first good in the set of goods; and [0080] To illustrate, consider that item characteristic correlator 717 is configured to identify usage rates 752a to 752c (e.g., a rate at which a product or service is reordered, or consumed or depleted) for corresponding user accounts 742a to 742c (e.g., associated with user phone numbers) and see [0067]
updating the second user record based on received user feedback responsive to the prompt. [0080] Thus, distribution predictor 714 may be able to discern patterns 750 of usage. Further, distribution predictor 714 may aggregate the usage rates to form an aggregated usage rate pattern 730 for a group of users 740. Based on aggregated usage rate pattern 730, distribution predictor 714 may be able to generate or predict an aggregated distribution event or an aggregated time of distribution 790.
Regarding claim 23, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining a measure of accuracy of current consumption amounts of goods determined for the updated second user record based at least in part on the user feedback; and identifying the updated second user record to the training set based on the measure of accuracy. [0044] In some examples, a usage rate to determine a predicted time of distribution may be based on identifying distribution rates of an item relative to one or more other accounts associated with one or more other users or other user computing systems to form an aggregate usage rate. An aggregated usage rate for an item may express, for example, a nominal usage rate that may be used (at least initially) to ascertain predicted time of reorder with a relatively high degree of confidence.
Regarding claim 24, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining, with the consumption model based on the updated second user record, a next predicted consumption amount for each good in a second set of goods for a next consumption period; and updating the second user record based on a next generated order submitted for fulfilment for the next consumption period. [0040] For example, distribution optimizer 118 may be configured to analyze data representing purchasing patterns related to a particular item for a specific user 144. Based on the results of such an analysis, distribution optimizer 118 may be configured to emphasize certain item characteristics (or values thereof) that may align more closely to a user's ordering or reordering patterns. [0043] Other users' patterns of purchasing, using, ordering, or reordering of the same item (e.g., same brand of vitamin A at the same merchant) or equivalent item (e.g., different brands of vitamin A at the same or different merchants) may also be used to predict a time of distribution. For example, consider that user 144 is replenishing an item, such a vitamin A tablets, at a merchant X. However, there may be negligible information to predict a usage rate (or a time of distribution) for that item at merchant X. Therefore, other users' patterns of reordering vitamin A at another merchant, merchant Y, may be used to form a predicted time of distribution for use in purchasing vitamin A tablets at merchant X. And see example in [0052], [0066], [0071]
Regarding claim 25, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining a second feedback prompt (electronic message 124a) corresponding to a different good in the set of goods based on an amount of change in the next predicted consumption amount for the different good relative to the previously predicted consumption amount. [0043]In another example, predicted distribution of an item for user 144 may be based on the user's pattern of purchasing, using, ordering, or reordering the item (or generically similar or complementary items). For example, a predicted time of distribution to replenish an item, such as a bottle of ketchup, may be based on a user's past rates of replenishment (e.g., shipment rates), such as a median or average time between successive requests to distribute reordered items. A distribution event may be predicted or supplemented by predicting a time of distribution for ketchup based on rates of past replenishment of mustard, a complementary product having a usage rates that may correlate to that of ketchup as both items may be used together (and thus consumed at similar depletion rates).
Regarding claim 26, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: receiving an input from the user; and generating the order based at least in part on the input. [0060]Thus, an adaptive distribution platform including conversation platform controller 215 may adapt presentation of user inputs to accommodate user purchasing and scheduling patterns and preferences to enhance, among other things, users' experiences.
Regarding claim 27, Fredrich discloses the limitations set forth above and further discloses:
wherein the input is received from the user without prompting the user to provide the input. [0053] In other examples, a user computing device 152a may initiate item replenishment. For example, user interface 156a may receive a user input 158a of “paper towels,” and, optionally, a destination account identifier (“774169”) 154a, such as a shortcode (or a phone number, an email address, a URL, etc.). User computing device 152a may be configured to transmit data representing user input 158a as electronic message 122b, which may also include data representing an account identifier (e.g., a mobile phone number) associated with user 144.
Regarding claim 28, Fredrich discloses the limitations set forth above and further discloses:
wherein the input indicates a status or a quantity of the good. (see prompt message in Figure 2 element 249a and 249b "order now", "order in x" days and element 210 interface with time and "d units")
Regarding claim 29, Fredrich discloses the limitations set forth above and further discloses:
wherein the input is received in natural-language text or speech. [0085] Computing platform 800 exchanges data representing inputs and outputs via input-and-output devices 801, including, but not limited to, keyboards, mice, audio inputs (e.g., speech-to-text driven devices)
Regarding claim 30, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining a first feedback prompt corresponding to a first good in the set of goods; receiving user feedback responsive to the prompt; and determining the predicted consumption amount for the good based in part on the user feedback. [0067]In some examples, distribution optimizer 318 may receive feedback relating to ordering patterns of user 331 (e.g., user 331 typically requests shipments prior to a date of exhaustion). Based on the feedback, distribution optimizer 318 may be configured to adjust a predicted distribution event at time 363d and a period 362c of time, which is shorter than periods 362a and 362b of time. And see [0052], [0063]
Regarding claim 31, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: determining that predicted consumption amounts, including the predicted consumption amount, are more accurate when the predicted consumption amounts are based on user responses, including the user feedback, to the first feedback prompt; and based on determining that the predicted consumption amounts are more accurate, increasing a frequency at which the first feedback prompt is provided.[0115] By enhancing accuracy of relevant information pertaining to a sample, information feedback predictor 914a may generate additional analytic information and insights into whether a particular sample is, for example, enhancing a “sample-to-purchase” conversion metric that describes, for example, a ratio of a number of items purchased against a number of samples sent. With improvements in said metric, fewer resources may be consumed or expended unnecessarily without a return on investment. According to various examples, adaptive distributive platform 910 and/or information feedback predictor 914a, or any other element, may be configured to store data memorializing electronic transactions and messages associated with predicting consumption of samples and feedback about the samples. Therefore, stored electronic transaction and electronic message data may be used to further refine a predicted date of consumption, as well as refinements in selecting a sample and modifying a feedback response interval of time, according to various examples
Regarding claim 35, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: steps for forming the training set from the first plurality of user records. [0080] To illustrate, consider that item characteristic correlator 717 is configured to identify usage rates 752a to 752c (e.g., a rate at which a product or service is reordered, or consumed or depleted) for corresponding user accounts 742a to 742c (e.g., associated with user phone numbers). In this example, consider that users 742a to 742c purchase a “laundry detergent” having usage rates 752a to 752c. Usage rates between “0” and “1” (e.g., usage amounts during a spring season), usage rates between “1” and “2” (e.g., usage amounts during a summer season), and usage rates between “2” and “3” (e.g., usage amounts during a fall season). It may be that users 742a to 742c play football during the fall, and consequently use more laundry detergent due to football practices and games in inclement weather (e.g., due to muddy fields, etc.) Thus, distribution predictor 714 may be able to discern patterns 750 of usage. Further, distribution predictor 714 may aggregate the usage rates to form an aggregated usage rate pattern 730 for a group of users 740. Based on aggregated usage rate pattern 730, distribution predictor 714 may be able to generate or predict an aggregated distribution event or an aggregated time of distribution 790.
Regarding claim 36, Fredrich discloses the limitations set forth above and further discloses:
the operations comprising: forming a validation set from the first plurality of user records. [0044] In some examples, a usage rate to determine a predicted time of distribution may be based on identifying distribution rates of an item relative to one or more other accounts associated with one or more other users or other user computing systems to form an aggregate usage rate. An aggregated usage rate for an item may express, for example, a nominal usage rate that may be used (at least initially) to ascertain predicted time of reorder with a relatively high degree of confidence.
Regarding claim 39, Fredrich discloses the limitations set forth above and further discloses:
wherein the identifiers comprise stock keeping units (SKUs). [0070] Examples of some item characteristics may include a product or product type, a service or service type, SKU data, UPC data, etc. for the same or similar items, or complementary and different items
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 32, 33, 34, and 37 are rejected under 35 U.S.C. 103 as being unpatentable over Fredrich (US 20180144389) in view of Wu (US 20170053516).
Regarding claim 32, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
determining, based on the received user feedback, that a household of the user has diverged from past consumption habits.
However Wu teaches:
determining, based on the received user feedback, that a household of the user has diverged from past consumption habits. [0085] Since the number of diners at home is relatively fixed, accordingly, the food consumption rate of the food in the refrigerator is relatively fixed, and thus the user may set the predetermined number of diners, and the food consumption rate determined according to the food habit information of the user is the food consumption rate of the predetermined number of diners. However, when there is a visitor at home or some family member is on a business trip, the actual number of diners will be changed, and the food consumption rate will be changed accordingly. In order to further increase the accuracy of the estimated serving times, the terminal also needs to further acquire the actual number of diners, and modify the food consumption rate according to the actual number of diners. And see [0097]
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the analysis of the usage patterns in Fredrich to include determining, based on the received user feedback, that a household of the user has diverged from past consumption habits, as taught in Wu, in order to further increase the accuracy (see paragraph 0085).
Regarding claim 33, Fredrich in view of Wu teaches the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
determining additional feedback prompts based on the divergence from the past consumption habits
However Wu teaches:
determining additional feedback prompts based on the divergence from the past consumption habits [0113] In order to facilitate the user to supply food in time, when detecting the lack of food, the terminal further acquires a purchase linkage (e.g., online purchasing information including price) of the food, and adds the purchase linkage to the reminder information. As shown in FIG. 3E, step 306 may include steps 306A and 306B. [0114] In step 306A, the purchase linkage of the food is acquired.[0115] The terminal acquires at least two prices for the food, compares these prices and acquires a purchase linkage corresponding to the lowest price. It should be noted that, the terminal may also acquire a corresponding purchase linkage according to a sales volume of the food or a degree of good comments, which is not limited by the present disclosure. [0116] In step 306B, the purchase linkage is added to the reminder information. [0117] The terminal adds the purchase linkage acquired to the reminder information and reminds the user to purchase. And see [0119]
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the analysis of the usage patterns in Fredrich to include determining additional feedback prompts based on the divergence from the past consumption habits, as taught in Wu, in order to further increase the accuracy (see paragraph 0085).
Regarding claim 34, Fredrich in view of Wu teaches the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
providing the additional feedback prompts to the user; receiving additional feedback responsive to the additional feedback prompts; and generating and submitting one or more additional orders based at least in part on the additional feedback
However Wu teaches:
providing the additional feedback prompts to the user; receiving additional feedback responsive to the additional feedback prompts; and generating and submitting one or more additional orders based at least in part on the additional feedback [0113] In order to facilitate the user to supply food in time, when detecting the lack of food, the terminal further acquires a purchase linkage (e.g., online purchasing information including price) of the food, and adds the purchase linkage to the reminder information. As shown in FIG. 3E, step 306 may include steps 306A and 306B. [0114] In step 306A, the purchase linkage of the food is acquired.[0115] The terminal acquires at least two prices for the food, compares these prices and acquires a purchase linkage corresponding to the lowest price. It should be noted that, the terminal may also acquire a corresponding purchase linkage according to a sales volume of the food or a degree of good comments, which is not limited by the present disclosure. [0116] In step 306B, the purchase linkage is added to the reminder information. [0117] The terminal adds the purchase linkage acquired to the reminder information and reminds the user to purchase. And see [0119] "reminders"
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the analysis of the usage patterns in Fredrich to include providing the additional feedback prompts to the user; receiving additional feedback responsive to the additional feedback prompts; and generating and submitting one or more additional orders based at least in part on the additional feedback, as taught in Wu, in order to further increase the accuracy (see paragraph 0085).
Regarding claim 37, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
determining one or more prompts for user feedback corresponding to household properties different from at least some household properties included in the second user record;
transmitting the one or more prompts for user feedback to a user device of the user;
updating the second user record based on received user feedback:
determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and
identifying the updated user record to the training set or the validation set based on the measure of accuracy.
However Wu teaches:
determining one or more prompts for user feedback corresponding to household properties different from at least some household properties included in the second user record; transmitting the one or more prompts for user feedback to a user device of the user; [0091] In order to ensure the accuracy of the actual number of diners, after determining the actual number of diners, the terminal may also send inquiry information to enquire the user whether the actual number of diners is accurate, and modify the actual number of diners according to a feedback from the user.
updating the second user record based on received user feedback; [0080]For example, the terminal may determines the food habit information of the user according to the history record of food consumption recorded in Table Two, and the food habit information may be stored in a storage structure as shown in Table Three.
determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and [0091] In order to ensure the accuracy of the actual number of diners, after determining the actual number of diners, the terminal may also send inquiry information to enquire the user whether the actual number of diners is accurate, and modify the actual number of diners according to a feedback from the user.
identifying the updated user record to the training set or the validation set based on the measure of accuracy. [0112] Alternatively, after determining the actual number of diners, the terminal further needs to detect whether the difference between the actual number of diners and the predetermined number of diners is greater than a predetermined threshold. If the difference is greater than the predetermined threshold, a visiting mode is entered, and the predetermined serving times is adjusted. For example, if the terminal detects that the actual number of diners is 6 people, the predetermined number of diners is 3 people, and the difference between the actual number of diners and the predetermined number of diners is 3 people (3 is larger than the predetermined threshold 2), then the visiting mode is entered, and the predetermined serving times is adjusted as 1, indicating that the foods in the refrigerator satisfy a food amount required for this visiting, that is, none reminder information is required to generate.
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the analysis of the usage patterns in Fredrich to include determining one or more prompts for user feedback corresponding to household properties different from at least some household properties included in the second user record; transmitting the one or more prompts for user feedback to a user device of the user; updating the second user record based on received user feedback: determining a measure of accuracy of current consumption amounts of goods determined for the updated user record based in part on the feedback; and identifying the updated user record to the training set or the validation set based on the measure of accuracy, as taught in Wu, in order to further increase the accuracy (see paragraph 0085).
Claim 38 is rejected under 35 U.S.C. 103 as being unpatentable over Fredrich (US 20180144389) in view of Bera (US 6523015).
Regarding claim 38, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the first user records in the validation set.
However Bera teaches:
wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the first user records in the validation set. [Col. 3 lines 30-40] A robust model is generated using a technique that optimizes the complexity of the model based on data obtained from the system being modeled. Data is split into a training data set and a generalization or cross validation data set. For a given complexity, weights are determined so that the error between the model output and the training data set is minimized.
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the analysis of the usage patterns in Fredrich to include wherein the model minimizes error of output determined consumption amounts relative to the consumption amounts for goods in at least some of the first user records in the validation set, as taught in Bera, so that the error between the model output and the training data set is minimized (Col. 3 lines 30-40).
Claim 41 is rejected under 35 U.S.C. 103 as being unpatentable over Fredrich (US 20180144389) in view of Danducci (US 11244280).
Regarding claim 41, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
training a second machine learning model based on orders received from the user;
and determining, with the second machine learning model, a predicted consumption amount for a second good; and
submitting a generated order for fulfillment based on the predicted consumption amount for the second good.
However Danducci teaches:
training a second machine learning model based on orders received from the user; [Col. 4 lines 49-55] In one embodiment, food utilization recommendation system 104 generates an estimation model that uses a machine learning tool that employs visual recognition in a training phase to obtain images (i.e., training images) of a food item at different times and classify consumption states of the food item at the different times based on the appearance of the food item in the images,
and determining, with the second machine learning model, a predicted consumption amount for a second good; and [Col. 5 lines 1-10] During the training phase, food utilization recommendation system 104 also obtains the measurements of the environmental conditions and attributes of the food item from sensors 108. The estimation model generated by food utilization recommendation system 104 uses the classification of the consumption states and the measurements obtained during the training phase as input. After the training phase, for a given food item, the estimation model outputs a current consumption state of the given food item and a remaining amount of time before the food item is unsuitable for consumption (i.e., the remaining shelf life of the food item) based on matching an image of the given food item to one of the training images and see Figure 4 Example
submitting a generated order for fulfillment based on the predicted consumption amount for the second good. [Col. 6 lines 33-35] Food utilization recommendation system 104 sends the recommendation for reordering ingredients to inventory system 116.
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the machine learning of Fredrich to include training a second machine learning model based on orders received from the user; and determining, with the second machine learning model, a predicted consumption amount for a second good; and submitting a generated order for fulfillment based on the predicted consumption amount for the second good, as taught in Danducci, in order to provide a recommendation for reorder based on utilization (Col. 6 lines 33-35)
Claim 42 is rejected under 35 U.S.C. 103 as being unpatentable over Fredrich (US 20180144389) in view of Tong (US 5359699).
Regarding claim 42, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
wherein the machine learning model comprises a feedforward neural network (FFNN).
However Tong teaches:
wherein the machine learning model comprises a feedforward neural network (FFNN). [Col. 1 lines 15-30] In a three-layer feedforward neural network, the first layer consists of input units where each unit simply receives a single component, i.e., data feature, of the input vector and transmits it to all units in the next layer called the hidden layer. Each unit in the hidden layer receives input from all input units weighed by connection weights, processes this input, and transmits an output to each unit of the output layer, again via weighted connections. The same processing of inputs occurs in the output units, resulting in a final output vector. Typically each neural unit, except the input units, sums the weighted inputs, passes the sum through a sigmoidal function, and outputs the result to the next layer of units.
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the machine learning of Fredrich to include wherein the machine learning model comprises a feedforward neural network (FFNN), as taught in Tong, in order to provide automated classification for heuristic rules (Col. 1 lines 60-65).
Claim 43 is rejected under 35 U.S.C. 103 as being unpatentable over Fredrich (US 20180144389) in view of Tong (US 5359699) in further view of Dirac (US 10635973).
Regarding claim 43, Fredrich discloses the limitations set forth above. While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), the reference does not expressly disclose:
processing an input feature vector with the FFNN, wherein:
the input feature vector comprises information regarding previous purchases of the user, and
determining the predicted consumption amount for the good in the set of goods is based on the input feature vector, including the information regarding the previous purchases of the user.
However Tong teaches:
processing an input feature vector with the FFNN, wherein: [Col. 1 lines 15-30] In a three-layer feedforward neural network, the first layer consists of input units where each unit simply receives a single component, i.e., data feature, of the input vector and transmits it to all units in the next layer called the hidden layer. Each unit in the hidden layer receives input from all input units weighed by connection weights, processes this input, and transmits an output to each unit of the output layer, again via weighted connections. The same processing of inputs occurs in the output units, resulting in a final output vector. Typically each neural unit, except the input units, sums the weighted inputs, passes the sum through a sigmoidal function, and outputs the result to the next layer of units.
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the machine learning of Fredrich to include processing an input feature vector with the FFNN, wherein, as taught in Tong, in order to provide automated classification for heuristic rules (Col. 1 lines 60-65).
While Fredrich discloses looking at usage patterns of associated users such as family members [0070] and sending prompt messages based on the determined patterns (Figure 2), and Tong discloses using FFNN for processing input vectors, the references do not expressly disclose:
the input feature vector comprises information regarding previous purchases of the user, and
determining the predicted consumption amount for the good in the set of goods is based on the input feature vector, including the information regarding the previous purchases of the user.
However Dirac teaches:
the input feature vector comprises information regarding previous purchases of the user, and [Col. 7 lines 20-35] In some embodiments, the set of input values 304 may comprise item consumption events that fall within a predetermined category or meet a specified condition. In some embodiments, the set of input values may be formatted as a matrix/vector, or a set of matrices/vectors. And see Col. 2 lines 30-45 for purchase information
determining the predicted consumption amount for the good in the set of goods is based on the input feature vector, including the information regarding the previous purchases of the user. [Col. 3 lines 13-35] The recommendation system 102 may receive an indication of user-specific item consumption events 108 (e.g., item consumption events that are each associated with a single user). Each of the user-specific item consumption events 108 may be stored in, and maintained by, the item consumption data 106 with respect to a timeline 110. For example, each item consumption event in item consumption data 106 may be associated with a timestamp or other suitable date/time indicator. The recommendation system 102 may receive user-specific item consumption events 108 that occur before a target date 112 in order to generate a consumption event prediction 114. In some embodiments, the target date 112 may be a current date, such that the consumption event prediction 114 represents a future item consumption event by the user associated with the user-specific item consumption events 108. In some embodiments, the target date 112 may be a past date, such that the consumption event prediction 114 may be compared to actual item consumption events occurring after the target date 112 in order to hone the prediction model 104 (e.g., adjust variables and/or assumptions used by the machine learning technique in generating a prediction model 104).
Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the machine learning of Fredrich in view of Tong to include the input feature vector comprises information regarding previous purchases of the user, and determining the predicted consumption amount for the good in the set of goods is based on the input feature vector, including the information regarding the previous purchases of the user, as taught in Dirac, in order to improve accuracy of the recommendations based on consumption events (abstract).
Relevant Art Not Cited
Concannon (US 10817885) discloses a system for determining the amount remaining of an item based on consumption rate data and initiating a subsequent replacement for the item.
Response to Arguments
The Examiner acknowledges the filed terminal disclaimer. However, the terminal disclaimer only addressed US PAT 11030577 and did not address conflicting US PAT US11922370. Therefore the double patenting rejection still stands in view of the amended claims.
Applicant's arguments filed 12/12/2025 have been fully considered but they are not persuasive.
With respect to the remarks directed to the 35 USC 101, the examiner asserts that the machine learning recited in the amended claims is recited at a high level of generality. The fact pattern of the instant application differs from that of the Desjardins decision. In the instant application that alleged improvement is to the accuracy of the predictions which at most improves the business process rooted in the abstract idea (predicting the frequency of the good to be delivered) and does not improve the technical aspects or the machine learning itself as was found in Desjardins. For at least these reasons the claims remain rejected under 35 USC 101.
With respect to the remarks directed to 35 USC 102. The amended independent claims remain anticipated by Fredrich. As claimed, the training of the claim, the teachings of detecting patterns of usage in [0080] of Fredrich is interpreted to teach the training the set of data. The patterns of usage are then used to predict a distribution event. The claim requires no more than training a set of data of a consumption model to predict what might be needed by the user in the future. The aggregated usage rate pattern 730, distribution predictor 714 that may be able to generate or predict an aggregated distribution event or an aggregated time of distribution 790 (see [0080]) teaches this limitation. The amended language of wherein the consumption model comprises a machine learning model is also taught in Fredrich. The reference discloses the use of machine learning for predicting feedback information for item attributes [newly cited [0113, 0114, 0115, 0118 and 0119]]. The use of the machine learning is recited at a high level in the claims without further detailing the manner in which the machine learning model is implemented. For at least reasons the rejection under 35 USC 102 is maintained.
The newly added claims are addressed under 35 USC 103 with newly cited prior art Tong, Danducci, and Dirac. All pending claims remain rejected under prior art.
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 VICTORIA E. FRUNZI whose telephone number is (571)270-1031. The examiner can normally be reached Monday- Friday 7-4 (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, Marissa Thein can be reached at (571) 272-6764. 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.
VICTORIA E. FRUNZI
Primary Examiner
Art Unit TC 3689
/VICTORIA E. FRUNZI/Primary Examiner, Art Unit 3689 1/8/2026