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 .
Claims 1-20 are pending.
Claims 1-11 are withdrawn from consideration.
No claims are amended. No claims are cancelled. No claims are added.
Election/Restrictions
Applicant’s election without traverse of Group II including claims 12-20 in the reply filed on 12/09/2025 is acknowledged.
Drawings
The drawings are objected to because each of Figure 3A-3H include text that is unreadable and boxes that are located on top of other information, Figs. 4A-4C includes text in gray that is difficult to read, Figs. 6A includes white text that is laid over images, and Figs. 6A-6C, 6E, 6F, 7A-7M, 7O, and 9A-10 includes gray on gray labeling or small text that is illegible.
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
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.
Claim 19 is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Representative claim 19 recites “obtaining an intended demographic focus for the machine learning system; obtaining demographic information for each of a plurality of training data elements; and …”. Therefore, the claim as a whole is directed to “Obtaining demographic information”, which is an abstract idea because it is a method of organizing human activity, including commercial interactions (including marketing or sales activities or behaviors; business relations); managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) and mental process including concepts performed in the human mind (including an observation, evaluation, judgment, opinion). “Obtaining demographic information” is considered to be is a method of organizing human activity and mental process because the selection of an intended design goal and selection of data to be used for business decision making. The claimed obtaining steps are purely mental decisions used in the course of business evaluations.
This judicial exception is not integrated into a practical application. In particular, claim 19 recites the following additional element(s): training the machine learning system based on weighting the plurality of training data elements based on the demographic information and the intended demographic focus. This additional element does not integrate the exception into a practical application. That is, the recitations of additional elements amount merely reciting the words ‘‘apply it’’ (or an equivalent) with the judicial exception, or merely including instructions to implement an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea (see MPEP 2106.05(f)). That is, the selection of data to be used as training data for machine learning system does not alter the machine learning system, does not address a technical problem, and does not provide any technical solution to solve a problem. Rather, certain data is applied in generically recited machine learning model, which is commonly commercially available software. As such, the recitation of training a machine learning model does no more than generally link the use of a judicial exception to a particular technological environment or field of use (see MPEP 2106.05(h)). Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. Claim 19 is directed to an abstract idea.
Claim 19 does 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 are merely being used to apply the abstract idea to a technological environment. As noted above, the claimed additional element does not address a technical problem, and does not provide any technical solution. Rather, the claim merely places obtained data in a generally recited technological environment. Accordingly, claim 19 is ineligible.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of pre-AIA 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 12 and 14-18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by U.S. Patent Application Publication No. 20190228105 to Ma et al.
With regards to claim 12, Ma et al. teaches
identifying user options provided by the e-commerce application (paragraph [0002], “Website visitors sometimes have to navigate through many linked webpages of a website to accomplish goals they have when accessing the website. For example, a user shopping for shoes on an online retail website may have to click through a succession of hyperlinks to access a webpage displaying women's, size 9, silver shoes, with a ½ inch heel. ");
generating a probabilistic network having nodes corresponding to the identified options and edges defining transition dependencies between the identified options (paragraph [0004], “An online system allows a website to be personalized according to the attributes and intents of specific visiting users. Hidden Markov models may be used to model transitions between user intents that are possible in view of the structure of the website. The online system develops probability distributions that are generally applicable to groups of website users and the online system can also develop customized probability distributions that account for particular user attributes and website interaction sequences. A Hidden Markov model structure uses three kinds of probability distributions: a transition probability distribution, an emission probability distribution, and an initial probability distribution.”; paragraph [0040], “Transition probabilities, shown in FIG. 3 as values associated with directed edges between hidden states 310, represent probabilities of a user transitioning from one intent to another intent. For example, a user with a first intent of browsing available shoes on a website may have a high probability of intending to view his shopping cart next. Thus, a HMM can model likely sequences of user intents, and the likelihood of a particular sequence of intents can be predicted based on the transition probabilities and associated sets of observation likelihood values 320.’);
assigning transition probabilities to the edges based on details of the identified options (paragraph [0038], “FIG. 3 illustrates a representation of an example Hidden Markov model loaded with transition probabilities, emission likelihoods 320, and initial probabilities 330.”);
associating at least one of the nodes with an action step (paragraph [0041], “A HMM may also incorporate an initial probability distribution that represents a probability that for each hidden state the state is the first state in a sequence of user intents. Initial probabilities 330 are represented in FIG. 3 as arrows that end at nodes representing states 310, but that do not originate from another node. It can also be possible for a HMM to have starting and ending states in lieu of initial probabilities 330 for each state 310.”; paragraph [0052], “That is, the implementation module 150 can affect the way a user interacts with a website architecture such that the user can access webpages and content associated with predicted intents easily. The implementation module 150 may show hyperlinks, images, text, or other content based on a best guess of a user's future intent. … In such an embodiment, other static graphical elements may be displayed for all users of the webpage. As one example, the implementation module 150 may include a hyperlink on a webpage the user is currently viewing that leads to a specific webpage that is associated with a predicted user intent. Such a hyperlink can allow the user to choose to bypass other intermediate webpages within the website architecture to access the webpage associated with a user intent quickly and easily. In another embodiment, the implementation module 150 may redirect a user to a webpage associated with a predicted user intent.”); and
traversing through the probabilistic network based on the transition probabilities and the action steps to simulate the user experience on the e-commerce application (paragraph [0053], “The implementation module 150 uses a ranking of the likely next intents of the user to determine what content to present to the user. The order of appearance or size of content items may be based on such a ranking. For example, if a user is predicted to be likely to have a next intent of browsing sale items, and if the user is also predicted to be almost as likely to want to view his shopping cart next, the implementation module may include both options in a list with a link to the sales webpage appearing first and a link to the user's shopping cart appearing second, and so on. In some cases, content related to only a certain number of the most highly ranked likely next user intents is presented to a user. For example, only the top three most likely next user intents may be presented as links.”).
With regards to claim 14, Ma et al. teaches dynamically updating the transition probabilities during the traversal through the probabilistic network (paragraph [0055], “As a user interacts with a website, observable actions are added to an observation sequence. The updated sequence of observations can then be re-applied by the ranking module 145 to determine additional future intents of the user. As is shown in FIG. 4, the process of updating the observation sequence to include a new user action, using the updated observation sequence as input to a HMM to determine a likely current user intent, applying a deterministic or Monte Carlo Tree search algorithm to the states of the HMM starting with a likely current user intent to predict a future user intent, and presenting content to the user at the client device 110 based on the predicted future user intent can be repeated cyclically while the user continues to interact with the website.”).
With regards to claim 15, Ma et al. teaches the details of the identified options comprise one or more of: option visibility, option prominence, and emotional context associated with the option (paragraph [0023], “Thus the crawler module receives information about the architecture of the website (e.g., a graph of interactive components of the website). The website graph and accompanying information is stored in the website graph store 165. The crawler module may periodically repeat a website traversal to capture any changes to the website content or structure over time.”; paragraph [0034], “This data may provide insight into how the online system 130 can be adjusted to work more effectively. For example, if a user navigates to a webpage that is associated with a predicted likely intent for that user, but does not access the webpage via the content presented by the implementation module 150, then the online system 130 may need to change the way content is presented to the user.”; paragraph [0052], “The implementation module 150 uses information generated by the ranking module 145 to determine both what type of content to show to the user and to determine how the content can be best presented within the graphical interface elements. In one embodiment, the implementation module 150 may control a subset of dynamic graphical interface elements within webpages of the website (i.e., elements that can be changed for different users). In such an embodiment, other static graphical elements may be displayed for all users of the webpage. As one example, the implementation module 150 may include a hyperlink on a webpage the user is currently viewing that leads to a specific webpage that is associated with a predicted user intent.”; paragraph [0054], “Similarly, a system administrator may provide additional configuration instructions that affect the sizing and ordering of website content in conjunction with considerations related to predicted next user intents. The implementation module 150 may affect different aspects of a website including specific images, specific links (e.g., suppressing or promoting certain links based on system administrator defined preferences), ordering of content, changing fonts, formatting content in a particular way, offering personalized incentives, customizing the HTML, CSS, and JS of the webpage, and so forth.”).
With regards to claim 16, Ma et al. teaches the action steps comprise one or more of adding a product to cart and completing a checkout (paragraph [0025], “For example, accessing a webpage displaying shoes may be categorized as “browse shoes” and the action of clicking on a button for submitting an order may be categorized as “make purchase.” In some cases, such categorization of observable user actions as relating to intents may be performed manually by a system administrator of the online system 130.”; paragraph [0053], “For example, if a user is predicted to be likely to have a next intent of browsing sale items, and if the user is also predicted to be almost as likely to want to view his shopping cart next, the implementation module may include both options in a list with a link to the sales webpage appearing first and a link to the user's shopping cart appearing second, and so on. In some cases, content related to only a certain number of the most highly ranked likely next user intents is presented to a user.”).
With regards to claim 17, Ma et al. teaches that the probabilistic network is a three-dimensional Markov network (paragraph [0004], “An online system allows a website to be personalized according to the attributes and intents of specific visiting users. Hidden Markov models may be used to model transitions between user intents that are possible in view of the structure of the website. The online system develops probability distributions that are generally applicable to groups of website users and the online system can also develop customized probability distributions that account for particular user attributes and website interaction sequences. A Hidden Markov model structure uses three kinds of probability distributions: a transition probability distribution, an emission probability distribution, and an initial probability distribution.”; paragraph [0005], “When a user accesses the website the three different types of probability distributions for that website are loaded into the Hidden Markov model structure. Values of each of the three probability distributions may be selected for a particular user based on the amount and type of data the online system has stored about past visits by the particular user to the website.”).
With regards to claim 18, Ma et al. teaches edges connecting nodes in first and second dimensions correspond to dependencies between user options in a first page of the e-commerce application (paragraph [0040], “Transition probabilities, shown in FIG. 3 as values associated with directed edges between hidden states 310, represent probabilities of a user transitioning from one intent to another intent. For example, a user with a first intent of browsing available shoes on a website may have a high probability of intending to view his shopping cart next. Thus, a HMM can model likely sequences of user intents, and the likelihood of a particular sequence of intents can be predicted based on the transition probabilities and associated sets of observation likelihood values 320.”), and
edges connecting nodes in a third dimension with nodes in the first or second dimensions correspond to dependencies between user options across multiple pages of the e-commerce application (paragraph [0005], “When a user accesses the website the three different types of probability distributions for that website are loaded into the Hidden Markov model structure. Values of each of the three probability distributions may be selected for a particular user based on the amount and type of data the online system has stored about past visits by the particular user to the website. As the user progressively interacts with the website, user actions are provided to the online system.”; paragraph [0046], “A variety of probability distributions may be generated for a website to be loaded into a HMM in different situations. For example, the set of probability distributions that are loaded into a HMM may depend on how much user profile data the online system 130 has stored about a user accessing a website. The pre-processing module 140 may train all three types of probability distributions to be based on observation sequences from all users, observation sequences from a particular classification of users, and based on sequences of observable actions performed by individual users.”).
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.
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20190228105 to Ma et al. as applied to claims 12 and 14-18 above, and further in view of U.S. Patent Application Publication No. 20200118145 to Jain et al.
With regards to claim 13, Ma et al. teaches repeating the step of traversing through the probabilistic network, but fails to explicitly teach measuring aggregate simulation performance results. However Jain et al. teaches ing the step of traversing through the probabilistic network and measuring aggregate simulation performance results (paragraph [0053], “To further explain the function of states for the user experience valuation techniques described herein, it would be helpful to understand the framework of the techniques. The browsing behavior of a user 104 of an online platform 122 may be modeled as a first-order Markov process to simulate the dynamics of the interactive computing environment.”; paragraph [0053], “The event analysis system 110 simulates the dynamics of the interactive computing environment and applies one or more techniques for exploiting the simulated environment to extract latent experience values. For example, the state prediction system 202 may include a recurrent neural network (RNN) trained to predict the next event in the user's journey.”; paragraph [0056], “The event analysis system 110 simulates the dynamics of the interactive computing environment and applies one or more techniques for exploiting the simulated environment to extract latent experience values. …he fully connected output layer produces an output of size |ϵ| through, for example, a softmax activation at each time-step of the sequence provided to the input layer. An example of the output for each time-step is a probability distribution vector over multiple possible next events.”).
This part of Jain et al. is applicable to the system of Ma et al. as they both share characteristics and capabilities, namely, they are directed to using probabilistic networks for modeling interactions with a system. 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 system of Ma et al. to include the aggregated simulation as taught by Jain et al. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to modify Ma et al. in order to better modifying the online computing platform based on these events to improve future consumer experiences (see paragraph [0002] of Jain et al.).
Claims 19 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20190228105 to Ma et al. as applied to claims 12 and 14-18 above, and further in view of U.S. Patent Application Publication No. 20230011954 to Elserafy et al.
With regards to claim 19, Ma et al. teaches
obtaining an intended … focus for the machine learning system (paragraph [0046], “Additionally, the pre-processing module 140 may use models trained with user attribute data to generate probability distribution values for subgroups of users. In these ways, the online system 130 can store HMM probability distributions that are customized for repeat users of a website and that are generalized for users who have never visited the website before.”);
obtaining …. information for each of a plurality of training data elements (paragraph [0046], “Additionally, the pre-processing module 140 may use models trained with user attribute data to generate probability distribution values for subgroups of users. In these ways, the online system 130 can store HMM probability distributions that are customized for repeat users of a website and that are generalized for users who have never visited the website before.”); and
training the machine learning system based on weighting the plurality of training data elements based on the … information and the intended … focus (paragraph [0030], “To use a HMM, the ranking module 145 populates the HMM with values from probability distributions that are optimized for making predictions about a particular user or for a group of users. In some cases, the ranking module 145 may differentiate between users for whom existing records are stored (e.g., in the user profile store 170) and users who are new to a website or for whom the online system 130 has no data. After predicting a current user intent, the ranking module 145 selects a likely next user intent (or future intent) using a deterministic algorithm or a Monte Carlo Tree search which can examine possible future user intents when applied to a graph structure of an HMM. Additional details about the ranking module will be included in the description of FIG. 4. More information about the HMMs is included in the description of FIG. 3.”; paragraph [0056], ) but fails to explicitly teach that the information is demographic information. However Elserafy et al. teaches
obtaining an intended demographic focus for the machine learning system (paragraph [0139], “each of the identified competing businesses based on business field, target customer demographics, products and services, size, and the like, and assign a weighting value based on the determined overall similarity (e.g., businesses with greater similarity to the first business are allocated greater weighting values).”);
obtaining demographic information for each of a plurality of training data elements (paragraph [0139], “each of the identified competing businesses based on business field, target customer demographics, products and services, size, and the like, and assign a weighting value based on the determined overall similarity (e.g., businesses with greater similarity to the first business are allocated greater weighting values).”); and
training the machine learning system based on weighting the plurality of training data elements based on the demographic information and the intended demographic focus (paragraph [0139], “The weighting function 1046 provides functionality for assigning a weighting value to each business identified as a potential competitor based on the collected competitor business information. The weighting value refers to a quantitative measure of the relative importance or relevance, or impact of a competing business on the first business. In embodiments, the weighting value may be expressed as an integer value between 0 and 10, where lower values indicate lower relevance, and higher values indicate higher relevance. Here, the weighting function 1046 may include a machine learning technique trained to evaluate the overall similarity between the first business and each of the identified competing businesses based on business field, target customer demographics, products and services, size, and the like, and assign a weighting value based on the determined overall similarity (e.g., businesses with greater similarity to the first business are allocated greater weighting values). In embodiments, the weighting value may be assigned or revised by a user.”).
This part of Elserafy et al. is applicable to the system of Ma et al. as they both share characteristics and capabilities, namely, they are directed to using probabilistic networks for modeling interactions with a system. 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 system of Ma et al. to include the demographic data training as taught by Elserafy et al. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to modify Ma et al. in order to provide business specific analysis based on the collected competitor business information (see paragraph [0139] of Elserafy et al.).
With regards to claim 20, Ma et al. teaches assigning transition probabilities to the edges, but fails to explicitly teach training in accordance with the method of claim 19. However Elserafy et al. teaches
comprises conducting multiversal training in accordance with the method of claim 19 (paragraph [0139], “The weighting function 1046 provides functionality for assigning a weighting value to each business identified as a potential competitor based on the collected competitor business information. The weighting value refers to a quantitative measure of the relative importance or relevance, or impact of a competing business on the first business. In embodiments, the weighting value may be expressed as an integer value between 0 and 10, where lower values indicate lower relevance, and higher values indicate higher relevance. Here, the weighting function 1046 may include a machine learning technique trained to evaluate the overall similarity between the first business and each of the identified competing businesses based on business field, target customer demographics, products and services, size, and the like, and assign a weighting value based on the determined overall similarity (e.g., businesses with greater similarity to the first business are allocated greater weighting values). In embodiments, the weighting value may be assigned or revised by a user.”).
This part of Elserafy et al. is applicable to the system of Ma et al. as they both share characteristics and capabilities, namely, they are directed to using probabilistic networks for modeling interactions with a system. 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 system of Ma et al. to include the demographic data training as taught by Elserafy et al. One of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to modify Ma et al. in order to provide business specific analysis based on the collected competitor business information (see paragraph [0139] of Elserafy et al.).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
U.S. Patent Application Publication No. 20190294633 to Dembo et al. discusses visual mining of user behavior patterns are disclosed. A plurality of clickstreams may be received. Each clickstream may represent a corresponding sequence of user actions. A visual representation of each clickstream may be caused to be displayed on a device. Each visual representation may comprise a distinct graphical element for each user action of the corresponding clickstream, and each visual representation may be configured to indicate a frequency level of the corresponding clickstream. Each distinct graphical element may comprise a geometric shape and a corresponding color that distinctly represents the corresponding user action. Graphical elements of each visual representation may comprise a size that is proportional to the frequency level of the corresponding clickstream..
U.S. Patent Application Publication No. 20210342247 to Ciolek et al. discusses a graph model of a graphical user interface (GUI) may be generated by processing usage data of the GUI where the usage data comprises sequences of GUI pages and actions between GUI pages. The nodes of the graph model may be determined by obtaining GUI pages from the usage data, identifying dynamic GUI elements in the GUI pages, generating canonical GUI pages by modifying the GUI pages using the dynamic GUI elements, and creating graph nodes using the canonical GUI pages. The edges of the graph may be determined by processing actions from the GUI data that were performed by users to transition from one GUI page to another GUI page. The graph model of the GUI may be used for any appropriate application, such as determining statistics relating to the GUI or statistics relating to individual users of the GUI.
U.S. Patent Application Publication No. 20210275911 to Padmanabhan et al. discusses determining the cultural style of a participant of an organization by accessing records of organizational data and participant data to generate an instance of simulation, and initiating a game simulation session. It includes receiving inputs from the participant to automatically calculate and generate a first set of simulation events, and presents to the participant a first set of events, and receives a response as textual or spoken inputs in an unstructured form. It further performs text analytics on the inputs, identifies themes and clusters based on frequency and meaning of inputs, and calculates probabilities and generates a further set of simulation events for a changed state of organizational model. It further includes presenting a further set of simulation events to the participant for inputs.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Joshua D Schneider whose telephone number is (571)270-7120. The examiner can normally be reached on Monday - Friday, 9am-5pm.
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, Jessica Lemieux can be reached on (571)270-3445. 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.
/J.D.S./Examiner, Art Unit 3626
/JESSICA LEMIEUX/Supervisory Patent Examiner, Art Unit 3626
