CTFR 18/250,089 CTFR 98442 DETAILED ACTION This action is responsive to the claims filed on 03/27/2026. Claims 1-2, 4-5, and 7 are pending for examination. This action is Final. Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Response to Arguments Applicant’s arguments regarding the 101 rejection of the claims have been fully considered but are not persuasive. Although claim 1 has been amended to recite extracting stored behavior history, identifying threshold-crossing time intervals, evaluating effort and habituation features, training a prediction model, predicting a next behavior time, and outputting a result to a display, the claim remains directed to an abstract idea, namely collecting/analyzing behavior-history data, calculating numerical features and threshold relationships, and using those calculated values in a predictive model. The recited processor, memory, storage units, prediction model storage unit, and display are invoked at a high level of generality and do not impose any particular improvement to computer hardware, data storage architecture, model architecture, training algorithm, loss function, or other computer technology. Applicant’s asserted improvement in prediction accuracy or reduced computational burden is not commensurate with the claim scope because the claim does not recite any specific technological mechanism that improves computer functionality; rather, it recites selecting and calculating particular input variables for a generic prediction model. Similarly, outputting the prediction result to a display is merely insignificant post-solution activity. Even if the particular feature selection were alleged to be unconventional, the additional elements, considered individually and as an ordered combination, amount to no more than generic computer implementation of the abstract data-analysis/prediction concept and therefore do not integrate the abstract idea into a practical application or add significantly more. Applicant’s arguments with respect to the 103 rejection of the claims have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Claim Rejections - 35 USC § 101 07-04-01 AIA 07-04 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1-2, 4-5, and 7 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Statutory Categories Claims 1-2 are directed to a method. Claims 4-5 are directed to an apparatus. Claim 7 is directed to a computer-readable medium. Independent Claims – Claims 1, 4, and 7 Step 2A Prong 1: Does the claim recite an abstract idea, law of nature, or natural phenomenon? Yes. Independent claims 1, 4, and 7 recites limitations that are abstract ideas in the form of mental processes: Claim 1 recites : Identifying… a specific time interval from a first point in time at which the numerical value exceeds a previous threshold to a second point in time at which the numerical value exceeds a current threshold different from the previous threshold, based on the first behavior history of the person ( this limitation merely amounts to identifying a time interval based on given data at a high level of generality being interpreted as a mental process of evaluation which can reasonably be permed in human mind or with aid of pen and paper ) evaluating… a first feature indicating a first amount of effort the person makes until the numerical value exceeds the current threshold, based on the first behavior history of the person ( this limitation merely amounts to evaluating values from a storage unit at a high level of generality and is being interpreted as a mental process of evaluation which can reasonably be permed in human mind or with aid of pen and paper ) evaluating… a second feature indicating a degree of habituation of the person determined by how many times in the past the person has exceeded the current threshold in a positive direction as of the second point in time due to a change in rating of the person caused by the behaviors of the person, based on the first behavior history ( this limitation merely amounts to evaluating values from a storage unit at a high level of generality and is being interpreted as a mental process of evaluation which can reasonably be permed in human mind or with aid of pen and paper ) predicting a time of a next behavior of the person in a second behavior history ( this limitation merely amounts to predicting a time at a high level of generality and is being interpreted as a mental process of evaluation which can reasonably be permed in human mind or with aid of pen and paper ) Claim 1 also recites the following additional elements for the purposes of Step 2A Prong Two analysis: A behavior prediction method executed by a behavior prediction apparatus including a processor and a memory, the method comprising: (the recitation of a computer used at a high level of generality is being considered as mere instructions to apply an exception, see MPEP 2106.05(f)) by the processor (the recitation of a processor used at a high level of generality is being considered as mere instructions to apply an exception using generic computer components as a tool see MPEP 2106.05(f)) extracting, by the processor, a first behavior history including, for each of a plurality of behaviors of a person, a time of each behavior and a numerical value indicating a state of the person after each behavior, from behavior history information stored in a reference score/behavior history storage unit ( extracting the model in storage is being considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g) ) training a prediction model in which the first feature and the second feature are used as explanatory variables and the specific time interval is used as an explained variable, based on the first feature, the second feature, and the specific time interval (the recitation of a prediction model, trained at a high level of generality is being considered as mere instructions to apply an exception under Step 2A prong 2, see MPEP 2106.05(f)) storing the trained prediction model in a prediction model storage unit ( storing the model in storage is being considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g) ) by using the stored prediction model (the recitation of a prediction model, trained at a high level of generality is being considered as mere instructions to apply an exception under Step 2A prong 2, see MPEP 2106.05(f)) and outputting a result of the predicting to a display connected to the processor ( outputting predetermined results is considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g) ) The additional limitations fail step 2A Prong 2 of the 101 analysis because they do not transform the claim into a practical application. These limitations are too abstract or lack technical improvement that would make the concept practically useful. Without clear utility or integration into a specific field, the claim does not relate to any particular application. It does not meet the requirements of Step 2A Prong 2, as it fails to make the concept meaningfully applicable in practice. Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea. This claim recites the following additional elements for the purposes of Step 2B analysis: A behavior prediction method executed by a behavior prediction apparatus including a processor and a memory, the method comprising: (the recitation of a computer used at a high level of generality is being considered as mere instructions to apply an exception, see MPEP 2106.05(f)) extracting, by the processor, a first behavior history including, for each of a plurality of behaviors of a person, a time of each behavior and a numerical value indicating a state of the person after each behavior, from behavior history information stored in a reference score/behavior history storage unit ( extracting the model in storage is being considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g), for the purposes of step 2B it should be noted that the courts have recognized storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015) as well-understood, routine, and conventional activity. ) training a prediction model in which the first feature and the second feature are used as explanatory variables and the specific time interval is used as an explained variable, based on the first feature, the second feature, and the specific time interval (the recitation of a prediction model, trained at a high level of generality is being considered as mere instructions to apply an exception, see MPEP 2106.05(f)) storing the trained prediction model in a prediction model storage unit ( storing the model in storage is being considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g), for the purposes of step 2B it should be noted that the courts have recognized storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015) as well-understood, routine, and conventional data. ) by using the stored prediction model (the recitation of a prediction model, trained at a high level of generality is being considered as mere instructions to apply an exception, see MPEP 2106.05(f)) and outputting a result of the predicting to a display connected to the processor ( outputting predetermined results is considered as mere data gathering which is considered insignificant extra-solution activity, see MPEP 2106.05(g), for the purposes of step 2B it should be noted that the courts have recognized presenting offers and gathering statistics, OIP Techs., 788 F.3d at 1362-63, 115 USPQ2d at 1092-93 as well-understood, routine, and conventional data. ) The claim also fails Step 2B of the analysis because the additional limitations do not amount to significantly more than the abstract idea itself. The additional limitations do not enhance the claim in a way that would move it beyond its abstract ideas as they minimally elaborate on the core concept without adding any inventive or technical substance. Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible. Claims 4 and 7 recite limitations substantially similar to claim 1, as such a similar analysis applies. Claim 4 recites an additional limitation for consideration: a processor; and a memory storing executable instructions which, when executed by the processor cause the processor to (Under step 2A prong II and step 2B, this limitation is invoking computers or other machinery merely as a tool to perform an existing process, see MPEP 2106.05(f)) Claim 7 recites an additional limitation for consideration: A non-transitory computer-readable recording medium storing a program that causes the processor to execute the behavior prediction method (Under step 2A prong II and step 2B, this limitation is invoking computers or other machinery merely as a tool to perform an existing process, see MPEP 2106.05(f)) Dependents of Claim 1 and 4 The remaining dependent claims corresponding to independent claims 1 and 4 do not recite additional elements, whether considered individually or in combination, that are sufficient to integrate the judicial exception into a practical application or amount to significantly more than the judicial exception. The analysis of which is shown below: The claims below recite additional limitations which fail step 2A Prong 2 of the 101 analysis because they do not transform the claim into a practical application. These limitations are too abstract or lack technical improvement that would make the concept practically useful. Without clear utility or integration into a specific field, the claim does not relate to any particular application. It does not meet the requirements of Step 2A Prong 2, as it fails to make the concept meaningfully applicable in practice. The claims also fails Step 2B of the analysis because the additional limitations do not amount to significantly more than the abstract idea itself. The additional limitations do not enhance the claim in a way that would move it beyond its abstract ideas as they minimally elaborate on the core concept without adding any inventive or technical substance. The claims are unpatentable. Claim 2 recites the additional limitation of : The behavior prediction method according to claim 1, wherein the previous threshold is one stage lower than the current threshold ( this limitation is merely additional information for an aforementioned abstract idea and is still considered a mental process of evaluation which can reasonably be performed in human mind or with aid of pen and paper ) wherein the first feature further indicates a second amount of effort the person makes from the first point in time to the second point in time ( Under step 2A prong II and step 2B analysis: this limitation amount to merely indicating a field of use or technological environment in which to apply a judicial exception, which does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application ) wherein the second feature further indicates a degree of habituation of the person to the current threshold ( Under step 2A prong II and step 2B analysis: this limitation amount to merely indicating a field of use or technological environment in which to apply a judicial exception, which does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application ) Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible. Claims 5 recite limitations substantially similar to claims 2, as such a similar analysis applies. Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 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. 07-20-aia AIA 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. 07-103 AIA The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action. 07-23-aia AIA 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 non-obviousness. 07-20-02-aia AIA This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention. 07-21-aia AIA Claim s 1-2, 4-5 and 7 are rejected under 35 U.S.C. 103 as being unpatentable over Keim et al., ( US 20090061407 A1 ), hereafter referred to as Keim in view of Gerber ( WO 2014036396 A1 ), hereafter referred to as Gerber, and in further view of Du et al. ( Du, N., Dai, H., Trivedi, R., Upadhyay, U., Gomez-Rodriguez, M., & Song, L. (2016, August). Recurrent marked temporal point processes: Embedding event history to vector. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1555-1564). ), hereafter referred to as Du . Claim 1 : Keim teaches: A behavior prediction method executed by a behavior prediction apparatus including a processor and a memory, the method comprising: ( Keim, paragraph 13, “ In accordance with yet another aspect of the present invention, a system is disclosed for adaptive recall including at least a processor; at least a memory coupled to the processor; at least an input device coupled to the computer system; and one or more programs encoded by the memory, the one or more programs causing the processor to determine a user's knowledge of a first item of information ”, Keim teaches a computer-implemented adaptive learning/recall system. The claimed “behavior prediction apparatus including a processor and a memory” reads on Keim’s processor/memory system executing software for determining user knowledge, calculating lag/adaptive recall time, and selecting review/testing items. ) extracting, by the processor, a first behavior history including, for each of a plurality of behaviors of a person, a time of each behavior and a numerical value indicating a state of the person after each behavior, from behavior history information stored in a reference score/behavior history storage unit ( Keim, paragraph 30, “ new items of information are reviewed in an initial lesson and test or an initial test to obtain data to calculate a first lag time ”; paragraph 31, “ determination of a user’s knowledge of information occurs when an item ‘A’ of information is tested … and the received data from a user is recorded ”; paragraph 34, “ the data from the user’s response, such as, but not limited to, number of correct answers, number of incorrect answers, speed of responding, the number of times the user has seen the item, or the like, are used to calculate and employ a lag time ”; paragraph 44, “ a lag-based learning system … creates and updates a user model, tracking a user’s performance and/or history on selected information over time … a user model includes different types of information, such as … a number of correct and incorrect responses, a number of times a user has encountered the content and in what context, the speed of response, patterns of correct or incorrect responses, a user’s desired learning objective, i.e., desired level of knowledge ”; paragraph 45, “ Delay factors 1201 represent how long it takes a user to answer a question … the amount of time the answer takes … and correctness … Each of the three items is graded with a number higher or lower than 1 ”; Keim teaches extracting/using stored user-model information corresponding to a first behavior history. The claimed “plurality of behaviors” reads on Keim’s repeated user review/testing events, including the initial test and subsequent review/testing events. The claimed “time of each behavior” reads on Keim’s recorded/calculated lag time, previous testing time, subsequent review/testing time, delay, elapsed response time, and time when/between items are presented. The claimed “numerical value indicating a state of the person after each behavior” reads on Keim’s numerical grades/scores, correctness values, response-time grades, and desired/actual knowledge-level information stored in the user model after user responses. Thus, Keim teaches stored behavior-history information including repeated user behaviors, time information associated with those behaviors, and numerical user-state/performance values resulting from the behaviors. ) identifying, by the processor, a specific time interval from a first point in time at which the numerical value exceeds a previous threshold to a second point in time at which the numerical value exceeds a current threshold different from the previous threshold, based on the first behavior history of the person ( Keim, paragraph 33, “ the equation … calculates the lag time, i.e. minimum amount of time between a previous testing and a subsequent review and testing of an item of information ”; paragraph 44, “ the customized path may include parameters, such as … the time when or between when the items are presented … the result is a customized user experience that updates to keep the information at or near the threshold of the user’s capabilities, always advancing the user’s level of knowledge until the user reaches the desired level of knowledge ”; paragraph 50, “ When a user can produce an item beyond a certain pre-selected threshold or level of knowledge … the user is considered fluent … learning a phase may require progression through at least another phase … a user may be required to pass at least one phase before an item of information from the phase becomes available for another phase ”; paragraph 51, “ a transitional test is employed to help a user transition from one phase to another phase ”; paragraph 53, “ an average user takes a given number of repetitions to learn something in a particular phase then a given amount of time before the average user is ready for a test at another phas e”; Keim teaches identifying a time interval between staged threshold/phase events. The claimed “previous threshold” reads on a lower pre-selected knowledge threshold/phase that the user must pass, and the claimed “current threshold different from the previous threshold” reads on a subsequent higher knowledge threshold/phase. The “first point in time” is the time associated with passing/exceeding the lower phase/threshold, and the “second point in time” is the later time associated with passing/exceeding the next phase/threshold. Keim expressly teaches calculating lag/elapsed time between previous and subsequent testing and teaches a given amount of time before the user is ready for testing at another phase; therefore, Keim teaches identifying the claimed specific time interval based on the user’s stored performance/history. ) evaluating, by the processor, a first feature indicating a first amount of effort the person makes until the numerical value exceeds the current threshold, based on the first behavior history of the person ( Keim, paragraph 11, “ a method assesses a test of a user’s knowledge of an item with multi-level variables, such as … time delay, pronunciation, number of guesses, choices guessed, number of times user has seen an item before, whether the answer is correct or incorrect, elapsed time of response, writing, grading ”; paragraph 34, “ the data from the user’s response, such as … number of correct answers, number of incorrect answers, speed of responding, the number of times the user has seen the item … are used to calculate and employ a lag time ”; paragraph 44, “ a user model includes … a number of correct and incorrect responses, a number of times a user has encountered the content … the speed of response ”; paragraph 50, “ learning a phase may require progression through at least another phase ”; paragraph 53, “ an average user takes a given number of repetitions to learn something in a particular phase ”; Keim’s user-response variables teach the claimed first feature indicating an amount of effort. Specifically, the number of guesses, number of times the user has seen an item, number of correct/incorrect responses, response speed, elapsed response time, and number of repetitions quantify how much review/testing work the user performs. The phrase “until the numerical value exceeds the current threshold” is taught because Keim uses these variables during repeated testing/review until the user passes or progresses to the next phase/current pre-selected knowledge threshold. Thus, the first feature is evaluated from the user’s behavior history and indicates the effort expended up to the point where the user’s knowledge/performance value exceeds the current threshold. ) due to a change in rating of the person caused by the behaviors of the person, based on the first behavior history ( Keim, paragraph 40, “ In accordance with at least one embodiment for a non-binary test using active and passive assessment, a test can passively test N items while actively testing 1 item. N is a random integer, such as, but not limited to, 1, 4, 6, or the like. Now referring to FIG. 4, where N is 3, a user is presented with an image of an apple and asked to select the correct label for the fruit out of four options: “Pear”, “Apple”, “Pumpkin”, and “Celery”. If a user picks choice 1 for “Pear” when choice 2 for “Apple” is the answer, the test actively assesses that the user does not know choice 2 for “Apple” and passively assesses that the user does not know choice 1 for “Pear”. Because the user is confused with one answer over another, the user provides more information about the level of knowledge for the tested items. User model for items 1 and 2, “Pear” and “Apple, respectively, receives a small change in the negative direction indicating that the user requires more review and testing for items 1 and 2. Additionally, because the user did not pick choices 3 or 4, i.e., “Pumpkin” or “Celery”, respectively, the user is not confused that those items resemble an apple. As a result, user model for items 3 and 4 receives a small change in the positive direction indicating that the user requires less review and testing for items 3 and 4. ”; Keim, paragraph 44, “ a lag-based learning system … creates and updates a user model, tracking a user’s performance and/or history on selected information over time … A user model includes different types of information, such as, but not limited to, a number of correct and incorrect responses, a number of times a user has encountered the content and in what context, the speed of response, patterns of correct or incorrect responses, a user’s desired learning objective, i.e., desired level of knowledge, or the like. ”; Keim teaches that the user model is updated positively or negatively based on the user’s testing/answer behavior and that the user model tracks the user’s performance/history over time. The user model value is interpreted as the claimed rating of the person, the correct/incorrect testing responses are interpreted as the claimed behaviors of the person, and the positive/negative update to the user model teaches the claimed change in rating caused by the behaviors based on the first behavior history. ) Gerber, in the same field of prediction model implementation, teaches the following limitations which Keim fails to teach: evaluating, by the processor, a second feature indicating a degree of habituation of the person determined by how many times in the past the person has exceeded the current threshold in a positive direction as of the second point in time ( Gerber, paragraph 25, “ Generally, the systems and methods described herein can calculate a statistic quantifying the Gerber relationship between variables. This statistic representing the Gerber relationship, as described herein, is known as the ‘Gerber statistic.’ In some embodiments, the Gerber statistic can be a positive or negative number reflecting the relative direction and strength of the relationship. Calculating the Gerber statistic may include counting the number of instances when values of both variables changed beyond a threshold and considering whether those changes were both in the same direction or in opposite directions. Instances when both variables move beyond the threshold and in the same direction (i.e., have a positive relation) are referred to herein as ‘positive unions,’ while instances when both variables move beyond the threshold and in opposite directions (i.e., have a negative relation) are referred to herein as ‘negative unions.’ Only periods in which both variables have movements beyond the threshold may be considered when calculating the Gerber statistic .”; Gerber, paragraph 27, “ In another example of calculating a Gerber statistic, there are 5 of the 10 days where the value of the first variable moved more than the threshold value. During those 5 days, the second variable only moved more than the threshold value 4 times. Therefore, the number of total unions is 4. During 3 of those 4 days, the first and second variables moved in the same direction (i.e., both positive or both negative), so the number of positive unions is 3. ”; Gerber, claim 8, “ determining, by a computer, how many instances in the historical data the performance of both the first and second variables exceed the threshold value; for each instance in which the performance of both the first and second variables exceeds the threshold value, determining, by a computer, whether the first and second variables have a positive union or a negative union ”; Gerber teaches a processor/computer determining how many historical instances exceed a threshold and determining whether those threshold-exceeding instances are positive-direction/same-direction events. Gerber’s “positive unions” teach the claimed positive-direction threshold exceedance events, and Gerber’s counted number of positive unions teaches “how many times in the past” the threshold was exceeded in the positive direction. The count of repeated positive-direction threshold exceedance events reasonably reads on the claimed degree of habituation, because the count indicates the frequency with which the relevant behavior/value has repeatedly attained or exceeded the current threshold in the positive direction. ) and outputting a result of the predicting to a display connected to the processor ( Gerber, paragraph 29, “ the systems and methods can comprise a computer program embodied on a computer-readable medium that can automatically perform the functions described herein, retrieve information to perform these functions, and display or output the results on a graphical user interface or provide the results to another system for further processing ”; Gerber, paragraph 55, “ the system 200 can present information for display on computer 210 … regarding data points associated with a variable record”; “the system 200 can output this information for each variable ”; Gerber teaches outputting computer-calculated analysis results to a graphical user interface/display. Gerber teaches the display/output of computer-generated analytical results. Therefore, it would have been obvious to output predicted next-behavior time to a display connected to the processor. ) Keim teaches maintaining a time-based user performance history, multi-stage knowledge thresholds/phases, and effort-like user-history variables for determining review/testing timing. Gerber teaches counting historical threshold-exceeding events in a direction-sensitive manner. It would have been obvious to incorporate Gerber’s known threshold-exceedance counting technique into Keim’s staged user-performance history, because each reference uses historical time-series/performance data to produce computer-calculated metrics for prediction, scheduling, or modeling, and the combination would predict when a user will next reach or perform a behavior at a desired level with greater personalization. Du, in the same field of prediction model implementation, teaches the following limitations which Settles fails to teach: training a prediction model ( Du, page 1555, col. 1, paragraph 2, “ We develop an efficient stochastic gradient algorithm for learning the model parameters which can readily scale up to millions of events. ”, Du teaches training a prediction model for event sequences. The trained RMTPP model learns parameters from historical event sequences so that the model can predict future event timing and event type. This teaches the claimed “training a prediction model.” ) in which the first feature and the second feature are used as explanatory variables ( Du, page 1556, col. 1, paragraph 5, “ We point out that the proposed Recurrent Marked Temporal Point Process establishes a previously unexplored connection between recurrent neural networks and point processes, which has implications beyond temporal-spatial settings by incorporating more rich contextual information and features. ”, Du, page 1559, section 5.2, paragraph 2, “ In our algorithm framework 1,we need both sparse (the marker yj) and dense features at time tj. Meanwhile, the output is also mixed of discrete markers and real-value time, which is then fed into different loss functions including the cross-entropy of the next predicted marker and the negative log-likelihood of the next predicted event timing. ”, Du teaches using event-history-derived information and covariates/contextual features in a predictive temporal model. In the combination, an effort feature and a habituation feature are supplied as known user-history/contextual features. Thus, Du teaches the model-training framework in which such features would be used as explanatory variables. ) and the specific time interval is used as an explained variable, based on the first feature, the second feature, and the specific time interval ( Du, page 1556, col. 1, paragraph 4, “ We propose a novel marked point process to jointly model the time and the marker information by learning a general representation of the nonlinear dependency over the history based on recurrent neural networks. Using our model, event history is embedded into a compact vector representation which can be used for predicting the next event time and marker type. ”; Du, page 1558, col. 2, paragraph 2, “ Since now hj represents the influence of the history up to the j-th event, the conditional density for the next event timing can be naturally represented as f ∗ (tj+1) = f(tj+1|Ht) = f(tj+1|hj) = f(dj+1|hj), (8) ”, Du teaches that temporal information is a model output/target in an event-sequence prediction model. The claimed “specific time interval” is a temporal target value. In the combination, the threshold-to-threshold interval identified from Adaptive Recall’s phased user history is used as the target/explained variable in Du’s temporal prediction model, while the effort and habituation features are explanatory variables. ) predicting a time of a next behavior of the person in a second behavior history by using the stored prediction model ( Du, page 1556, col. 1, paragraph 4, “ We propose a novel marked point process to jointly model the time and the marker information by learning a general representation of the nonlinear dependency over the history based on recurrent neural networks. Using our model, event history is embedded into a compact vector representation which can be used for predicting the next event time and marker type. ”, Du expressly teaches next-event timing prediction from historical event sequences. The claimed “next behavior” corresponds to a next event in Du’s event-history framework. Therefore, Du teaches using the trained model to predict a time of a next behavior based on behavior history. ) storing the trained prediction model in a prediction model storage unit ( Du, abstract, “ We develop an efficient stochastic gradient algorithm for learning the model parameters ”; Du, page 1559, section 5.2, “ we can learn the model by maximizing the joint log-likelihood … we take b consecutive samples … apply the feed-forward operation through the network, and update the parameters with respect to the loss function ”; Du teaches that the prediction model is an RMTPP/RNN model having learned parameters. The claimed “trained prediction model” is interpreted as Du’s model after its parameters have been learned/updated through training. Because Du’s learned parameters are required for later next-event timing prediction, it would have been obvious to store the trained model parameters in memory/storage of the computer system so that the stored trained prediction model can subsequently be used to predict a next behavior time. ) It would have been obvious to a person of ordinary skill in the art, prior to the effective filing date of the invention, to incorporate Du’s RMTPP recurrent neural network temporal point process model with the features taught by Keim and Gerber as additional contextual/user-profile inputs, because Du expressly teaches a recurrent (including LSTM) time-series model for predicting future event timing that is readily generalized to use contextual information beyond inter-event timing alone. Accordingly, combining Du’s trained timing model with Keim/Gerber’s effort/habituation features would have been a routine and sensible design choice to improve the accuracy and personalization of predicting the time interval until the next user practice behavior. A motivation of which would have been to personalize and improve next-behavior timing prediction by incorporating user-history contextual features into the recurrent temporal point process model: “Besides, in addition to the inter-event temporal features, our model can be readily generalized to incorporate other contextual information. For instance… we can also take the potential user-profile features into account for personalization.” (Du, page 1563, section 7, paragraph 1). Claim 2 : Keim, Gerber, and Du teaches the limitations of claim 1, Keim further teaches: The behavior prediction method according to claim 1, wherein the previous threshold is one stage lower than the current threshold ( Keim, paragraph 50, “ When a user can produce an item beyond a certain pre-selected threshold or level of knowledge, e.g., without strong context, the user is considered fluent with a particular item. When a user moves successfully to phase 4 , i.e., production without context, the user may have achieved a goal for long term memory development among other goals depending on the user's desired level of knowledge. In some embodiments, learning a phase may require progression through at least another phase. In at least one embodiment, a user may be required to pass at least one phase before an item of information from the phase becomes available for another phase. ”, Keim teaches staged knowledge levels/phases. Because a user progresses from a lower phase to a higher phase, the prior phase/level is one stage lower than the current phase/level. Thus, Keim teaches a previous threshold that is one stage lower than a current threshold. ) wherein the first feature further indicates a second amount of effort the person makes from the first point in time to the second point in time ( Keim, paragraph 11, “ In another embodiment, a method assesses a test of a user's knowledge of an item with multi-level variables, such as, but not limited to, time delay, pronunciation, number of guesses, choices guessed, number of times user has seen an item before, whether the answer is correct or incorrect, elapsed time of response, writing, grading, whether path level assessment is involved, or the like. For example, an adjustment of lag and dependency selection may be based upon whether a user knows the item at least to an expert level. In one embodiment, if a user answers the questions well, then lag time would increase. ”, Keim’s stored response variables quantify the user’s effort during learning/testing. For the interval between a lower phase/threshold and the next higher phase/threshold, the number of attempts, number of times seen, correctness, response speed, and elapsed response time indicate the amount of user effort made during that interval. Therefore, Keim teaches a first feature further indicating the effort made from the first threshold-crossing time to the second threshold-crossing time. ) Gerber further teaches: wherein the second feature further indicates a degree of habituation of the person to the current threshold (Gerber, paragraph 25, “ Calculating the Gerber statistic may include counting the number of instances when values of both variables changed beyond a threshold and considering whether those changes were both in the same direction or in opposite directions. Instances when both variables move beyond the threshold and in the same direction (i.e., have a positive relation) are referred to herein as ‘positive unions,’ while instances when both variables move beyond the threshold and in opposite directions (i.e., have a negative relation) are referred to herein as ‘negative unions.’ Only periods in which both variables have movements beyond the threshold may be considered when calculating the Gerber statistic. ”; Gerber, paragraph 27, “ During 3 of those 4 days, the first and second variables moved in the same direction (i.e., both positive or both negative), so the number of positive unions is 3. ”; Gerber, claim 8, “ determining, by a computer, how many instances in the historical data the performance of both the first and second variables exceed the threshold value; for each instance in which the performance of both the first and second variables exceeds the threshold value, determining, by a computer, whether the first and second variables have a positive union or a negative union ”; Gerber, claim 10, “ wherein the first and second variables represent at least one of: behavior variables; sports variables; and market variables. ”; Gerber teaches counting historical threshold-exceeding instances for behavior variables and determining whether those instances are positive-direction/same-direction threshold events. The number of positive unions indicates the frequency of repeated positive threshold attainment. Therefore, Gerber’s counted positive-direction threshold exceedance events teach a feature indicating a degree of habituation to the current threshold, because the feature reflects how repeatedly the relevant behavior variable has reached or exceeded that threshold in the positive direction.) Claims 4-5 recite limitations substantially similar to claims 1-2 as such a similar analysis applies. Claim 4 also recites the following additional limitations for consideration: a processor and a memory storing executable instructions which, when executed by the processor, cause the processor to ( Du, page 1559, section 5.2, “ The backend is supported via CUDA and MKL for GPU and CPU platform, ”, it is interpreted by the examiner that a CPU and GPU implemented by Du encompasses the processor and memory storing executable instructions. ) Claim 7 recites limitations substantially similar to claims 1, as such a similar analysis applies. Claim 7 also recites the following additional limitations for consideration: A non-transitory computer-readable recording medium storing a program that causes the processor to execute the behavior prediction method according to claim 1. ( Du, page 1559, section 5.2, “ The backend is supported via CUDA and MKL for GPU and CPU platform, ”, it is interpreted by the examiner that a CPU and GPU implemented by Du encompasses the non-transitory computer-readable medium storing an executable program of the behavior prediction model. ) Conclusion 07-96 AIA The prior art made of record and not relied upon is considered pertinent to applicant's disclosure : Zhu, Y., Li, H., Liao, Y., Wang, B., Guan, Z., Liu, H., & Cai, D. (2017, August). What to do next: Modeling user behaviors by time-LSTM. In IJCAI (Vol. 17, pp. 3602-3608). Du, N., Wang, Y., He, N., Sun, J., & Song, L. (2015). Time-sensitive recommendation from recurrent user activities. Advances in neural information processing systems, 28. Mei, H., & Eisner, J. M. (2017). The neural hawkes process: A neurally self-modulating multivariate point process. Advances in neural information processing systems, 30. Yang, G., Cai, Y., & Reddy, C. K. (2018, October). Recurrent spatio-temporal point process for check-in time prediction. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management (pp. 2203-2211). Piech, C., Bassen, J., Huang, J., Ganguli, S., Sahami, M., Guibas, L. J., & Sohl-Dickstein, J. (2015). Deep knowledge tracing. Advances in neural information processing systems , 28 . US 2017/0246540 A1 US 2020/0250555 A1 US 2020/0387797 A1 THIS ACTION IS MADE FINAL. 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 HYUNGJUN B YI whose telephone number is (703)756-4799. The examiner can normally be reached M-F 9-5. 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, Usmaan Saeed can be reached on (571) 272-4046. 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. /H.B.Y./Examiner, Art Unit 2146 /USMAAN SAEED/Supervisory Patent Examiner, Art Unit 2146 Application/Control Number: 18/250,089 Page 2 Art Unit: 2146 Application/Control Number: 18/250,089 Page 3 Art Unit: 2146 Application/Control Number: 18/250,089 Page 4 Art Unit: 2146 Application/Control Number: 18/250,089 Page 5 Art Unit: 2146 Application/Control Number: 18/250,089 Page 6 Art Unit: 2146 Application/Control Number: 18/250,089 Page 7 Art Unit: 2146 Application/Control Number: 18/250,089 Page 8 Art Unit: 2146 Application/Control Number: 18/250,089 Page 9 Art Unit: 2146 Application/Control Number: 18/250,089 Page 10 Art Unit: 2146 Application/Control Number: 18/250,089 Page 11 Art Unit: 2146 Application/Control Number: 18/250,089 Page 12 Art Unit: 2146 Application/Control Number: 18/250,089 Page 13 Art Unit: 2146 Application/Control Number: 18/250,089 Page 14 Art Unit: 2146 Application/Control Number: 18/250,089 Page 15 Art Unit: 2146 Application/Control Number: 18/250,089 Page 16 Art Unit: 2146 Application/Control Number: 18/250,089 Page 17 Art Unit: 2146 Application/Control Number: 18/250,089 Page 18 Art Unit: 2146 Application/Control Number: 18/250,089 Page 19 Art Unit: 2146 Application/Control Number: 18/250,089 Page 20 Art Unit: 2146 Application/Control Number: 18/250,089 Page 21 Art Unit: 2146 Application/Control Number: 18/250,089 Page 22 Art Unit: 2146 Application/Control Number: 18/250,089 Page 23 Art Unit: 2146 Application/Control Number: 18/250,089 Page 24 Art Unit: 2146 Application/Control Number: 18/250,089 Page 25 Art Unit: 2146 Application/Control Number: 18/250,089 Page 26 Art Unit: 2146
Prosecution Insights