DETAILED ACTION
This Office Action is in response to Applicant's Communication received on 09/01/2026 for the above identified application.
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 .
Response to Amendment
The amendment filed on 09/01/2026 has been entered.
Claims 1, 8, and 15 are amended. Claims 2, 7, 9, 14, 18, and 23 are canceled. Claim 24 is new. Claims 1, 3, 5, 8, 10, 12, 15, 20-22, and 24 are pending in the application.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim 24 is rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention.
Dependent Claim 24 recites “wherein the computing device comprises a display device, and wherein providing the predicted interval value comprises displaying, on a graphical user interface of the display device, a visual forecast dashboard comprising: (i) a visual representation of the selected motif in the time series data; and (ii) the predicted interval value for the future time interval”. The specification [0065] recites "The computing system 500 may also include an I/O device interface 510 connecting I/O devices 512 (e.g., keyboard, display and mouse devices) to the computing system 500"; [0050] describes "The forecasting component 115 may provide the predicted interval value to the entity computing device 120 as the forecast 122." The specification does not disclose a graphical user interface, a visual forecast dashboard, any visual representation of a motif in the time series data, or displaying a predicted interval value. None of the cited paragraphs in the remarks describes the claimed dashboard or the visual representation of the selected motif. At best, the specification suggests sending the predicted interval value to the entity's computing device. Therefore, the language - wherein the computing device comprises a display device, and wherein providing the predicted interval value comprises displaying, on a graphical user interface of the display device, a visual forecast dashboard comprising: (i) a visual representation of the selected motif in the time series data; and (ii) the predicted interval value for the future time interval - constitutes new matter. (See also 37 C.F.R. 1.121(f), MPEP 608.04, 706.03(o)). For the purposes of examination, this limitation will be interpreted as: providing the predicted interval value to the computing device.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 3, 5, 8, 10, 12, 15, 20-22, and 24 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
Claims 1, 3, 5, 21-22, and 24 are directed to a method, Claims 8,10, and 12 are directed to a system, and Claims 15 and 20 are directed to medium. Thus, the claims fall within one of the statutory categories (process, machine, articles of manufacture) and are eligible under Step 1.
Step 2A Prong 1
Independent Claims
Claims 1, 8, and 15 recite:
calculating a matrix profile for the received time series data based on the received window size - these limitations encompass mathematical concepts, such as mathematical calculations to compute a matrix profile based on window length.
determining a plurality of subsequences of the time series data using the calculated matrix profile; determining a set of motifs from the plurality of subsequences using the calculated matrix profile, wherein each motif of the set of motifs is associated with a plurality of the subsequences and each motif of the set of motifs has a size equal to the received window size - these limitations encompass mathematical concepts and mental processes, such as a person making an observation/ evaluation/ judgement to identify patterns/ motifs having specific size/ duration from time series data using mathematical calculations.
for each motif of the set of motifs, selecting a forecasting algorithm based on an associated forecast error when predicting the interval value for time intervals from the plurality of subsequences of the time series data associated with the motif, wherein selecting the forecasting algorithm based on the associated forecast error comprises: (i) for each forecasting algorithm of the plurality of forecasting algorithms: sequentially processing each subsequence of the plurality of subsequences of the time series data associated with the motif in a chronological order of occurrence in the time series data by: (B) for each time interval in the subsequence: (1) using the forecasting algorithm to predict the interval value for the time interval; (2) determining a difference between the interval value predicted for the time interval and the interval value associated with the time interval in the time series data; and (3) accumulating the determined difference into a cumulative forecast error for the forecasting algorithm; and (ii) selecting the forecasting algorithm for the motif based on the forecasting algorithm having a lowest cumulative forecast error - these limitations encompass mathematical concepts and mental processes, such as a person making an observation/ evaluation/ judgement to select a forecasting algorithm for a motif based on forecasting error (mathematical relationship), perform mathematical calculations to compute value, calculate difference, sum the differences, compare the resulting sums and choose the smallest by comparing numbers. Requiring that the subsequences be processes sequentially… in a chronological order specifies only the order in which these calculations are carried out.
generating a mapping that associates each motif of the set of motifs with a corresponding selected forecasting algorithm - these limitations encompass mental processes, such as a person writing down the pairing of motif with the algorithm selected for it.
determining a default forecasting algorithm that performed the best for time intervals that are not associated with any motif of the set of motifs - these limitations encompass mental processes, such as a person making an evaluation/ judgement over forecast-error values (mathematical results) for the non-motif time intervals.
determining whether the future time interval is associated with a motif of the set of motifs; in response to determining the future time interval is associated with a motif of the set of motifs, using the mapping to select the forecasting algorithm for the motif associated with the future time interval; in response to determining the future time interval is not associated with any motif of the set of motifs, selecting the default forecasting algorithm - these limitations encompass mental processes, such as a person deciding a motif associated with a future time interval by checking whether the requested time falls within the recurring window, then looking up the algorithm in the table, or else using the default.
using the selected forecasting algorithm to predict the interval value for the future time interval - these limitations encompass mathematical concepts, such as use the selected forecasting algorithm (mathematical calculations) associated with the motif to predict/ calculate interval value.
Accordingly, these claims recite an abstract idea that falls under the “mathematical concepts” and “mental process” grouping.
Step 2A Prong 2
Independent Claims
Additional elements
Claims 1, 8, and 15:
receiving time series data by a computing device, wherein the time series data comprises a plurality of time intervals and each time interval is associated with an interval value; receiving a plurality of forecasting algorithms by the computing device; receiving a window size by the computing device; receiving a request to forecast the interval value at a future time interval by the computing device - these imitations amount to insignificant extra-solution activity of mere data gathering (see MPEP § 2106.05(g)).
by a computing device - this limitation is recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
(A) training the forecasting algorithm to predict the interval value using a portion of the time series data that is chronologically before the subsequence in the time series data - these limitations recite training a generic forecasting algorithm at a high level of generality, such that it amounts to no more than an instruction to apply the abstract idea using a generic machine learning tool (see MPEP § 2106.05(f)) and generally linking the use of a judicial exception to the field of machine learning (see MPEP § 2106.05(h)).
providing the predicted interval value for the future time interval by the computing device - these imitations amount to insignificant extra-solution activity of mere data gathering/ outputting (see MPEP § 2106.05(g)).
scheduling one or more workers to work during the future time interval based on the predicted interval value - this limitation is recited at a high-level of generality such that it amounts to no more than merely indicating the field of use/ workforce planning in which to apply the judicial exception (see MPEP § 2106.05(h)).
Claim 8:
a system comprising: one or more processors; and a memory storing instructions that when executed by the one or more processors cause the system to perform method - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 15:
a non-transitory computer-readable medium storing instructions that when executed by one or more processors of a system cause the system to perform method - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Accordingly, these additional elements do not integrate the judicial exception into a practical application because they do not impose any meaningful limits on practicing the abstract idea. These claims are directed to the abstract idea.
Step 2B
Independent Claims
Additional elements
receiving time series data by a computing device, wherein the time series data comprises a plurality of time intervals and each time interval is associated with an interval value; receiving a plurality of forecasting algorithms by the computing device; receiving a window size by the computing device; receiving a request to forecast the interval value at a future time interval by the computing device - these imitations amount to insignificant extra-solution activity of mere data gathering, which is a well-understood, routine, and conventional activity (see MPEP § 2106.05(d), “receiving/ transmitting data”).
by a computing device - this limitation is recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
(A) training the forecasting algorithm to predict the interval value using a portion of the time series data that is chronologically before the subsequence in the time series data - these limitations recite training a generic forecasting algorithm at a high level of generality, such that it amounts to no more than an instruction to apply the abstract idea using a generic machine learning tool (see MPEP § 2106.05(f)) and generally linking the use of a judicial exception to the field of machine learning (see MPEP § 2106.05(h)).
providing the predicted interval value for the future time interval by the computing device - these imitations amount to insignificant extra-solution activity of mere data gathering/ outputting, which is a well-understood, routine, and conventional activity (see MPEP § 2106.05(d), “receiving/ transmitting data”, “presenting offers”).
scheduling one or more workers to work during the future time interval based on the predicted interval value - this limitation is recited at a high-level of generality such that it amounts to no more than merely indicating the field of use/ workforce planning in which to apply the judicial exception (see MPEP § 2106.05(h)).
Claim 8:
a system comprising: one or more processors; and a memory storing instructions that when executed by the one or more processors cause the system to perform method - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 15:
a non-transitory computer-readable medium storing instructions that when executed by one or more processors of a system cause the system to perform method - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Accordingly, these additional elements do not amount to significantly more than the judicial exception. As such, these claims are patent ineligible.
Step 2A Prong 1
Dependent Claims
Claims 3, 10:
the interval value is one of a communication volume, an average handling time, or a shrinkage – these limitations merely furthers the mathematical concepts by specifying the predicted interval value.
Claims 5, 12, 20:
selecting the forecasting algorithm based on the associated forecast error comprises selecting the forecasting algorithm with a minimum forecast error - these limitations encompass a mental process (which is observing, evaluating, and judging that can be practically performed in the human mind or by a human using a pen and paper) of selecting forecasting algorithm from the plurality of forecasting algorithms based on the minimum associated error (mathematical relationship).
Claim 21:
selecting the forecasting algorithm for a motif comprises calculating a forecast error using absolute differences between predicted interval values and observed interval values averaged across subsequences associated with the motif - these limitations merely furthers the mathematical concepts by specifying the mathematical calculations used in selecting the forecasting algorithm for the motif.
Claim 22:
the forecast error comprises a mean absolute percentage error - these limitations merely furthers the mathematical concepts by specifying the forecast error in selecting the forecasting algorithm for the motif.
Thus, the claims recite the abstract idea.
Step 2A Prong 2
Dependent Claims
Additional elements
Claims 10 and 12:
the system - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 20:
the non-transitory computer-readable medium - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 24:
the computing device comprises a display device, and wherein providing the predicted interval value comprises displaying, on a graphical user interface of the display device, a visual forecast dashboard comprising: (i) a visual representation of the selected motif in the time series data; and (ii) the predicted interval value for the future time interval - based on the 112(a) discussion above, this limitation is interpreted as: providing the predicted interval value to the computing device, which amounts to insignificant extra-solution activity of mere data gathering/ outputting (see MPEP § 2106.05(g)).
Accordingly, these additional elements do not integrate the judicial exception into a practical application because they do not impose any meaningful limits on practicing the abstract idea. These claims are directed to the abstract idea.
Step 2B
Dependent Claims
Additional elements
Claims 10 and 12:
the system - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 20:
the non-transitory computer-readable medium - these limitations are recited at a high-level of generality such that it amounts to no more than merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)). This limitation can also be viewed as generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)).
Claim 24:
the computing device comprises a display device, and wherein providing the predicted interval value comprises displaying, on a graphical user interface of the display device, a visual forecast dashboard comprising: (i) a visual representation of the selected motif in the time series data; and (ii) the predicted interval value for the future time interval - based on the 112(a) discussion above, this limitation is interpreted as: providing the predicted interval value to the computing device, which amounts to insignificant extra-solution activity of mere data gathering/ outputting, which is a well-understood, routine, and conventional activity (see MPEP § 2106.05(d), “receiving/ transmitting data”, “presenting offers”).
Accordingly, these additional elements do not amount to significantly more than the judicial exception. As such, these claims are patent ineligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3, 5, 8, 10, 12, 15, 20-22, and 24 are rejected under 35 U.S.C. 103 as being unpatentable over Yocum et al. (US 2021/0271925 A1 hereinafter Yocum) in view of Law (US 2020/0258157 A1), further in view of Bledsoe et al (US 2018/0300737 A1 hereafter Bledsoe).
Regarding Claim 1, Yocum teaches a method for selecting a forecasting algorithm for a motif ([0023] for a given schedule of individual days/ points in time for which predictions are to be made, the system may select the best model by forecast group, prediction horizon, and model class for a particular day (i.e., motif)) comprising:
receiving time series data by a computing device, wherein the time series data comprises a plurality of time intervals and each time interval is associated with an interval value ([0013] models are trained to predict time series data, which may include the average handle time and call volume; [0015] the call volume identifies the number of phone calls that are expected to be received during a specified time interval; [0023] for a given schedule of individual days/ points in time for which predictions are to be made, the system may select the best model by forecast group, prediction horizon, and model class for a particular day; predictions are then made with a set of models to generate a piecewise forecast using the best model for a particular day; [0032] the repository (150) is a computing system that may include multiple computing devices in accordance with the computing system (500) and the nodes (522) and (524); the data in the repository (150) include multiple variables, parameters, historical time series data (i.e., time series data comprising plurality of time intervals), and forecasts that are generated (i.e., interval value) and used by the components of the system - thus, comprising plurality of time intervals and each time interval associated with forecasts/ interval value; [0075] the nodes may include functionality to receive requests from the client device and transmit responses to the client device; the client device may be a computing system; [0076] the computing system or group of computing systems include functionality to perform a variety of operations disclosed);
receiving a plurality of forecasting algorithms by the computing device ([0002] obtaining, by a model discovery service, a plurality of models; [0037] the process generates predictions used to generate estimated headcounts from models with multiple horizons and classes; in Step 202, models are obtained corresponding to model horizons; [0043] multiple models may be trained having multiple model horizons and classes; a model discovery service may have a model space of different models that are trained; the model space may include time series models that have different classes, different lags, and be of different types; the different classes of models may include in-season models, out-of-season models, and all-season models; [0076] the computing system or group of computing systems include functionality to perform a variety of operations disclosed);
for each motif of the set of motifs, selecting a forecasting algorithm based on an associated forecast error when predicting the interval value for time intervals from the plurality of subsequences of the time series data associated with the motif by the computing device ([0020] a model class identifies, for a model, the class of the model; the classes (i.e., set of motifs) may relate to seasonality, which is how the model performs with respect to a particular period of time; [0023] for a given schedule of individual days/ points in time for which predictions are to be made (i.e., predicting the interval value), the system may select the best model by forecast group, prediction horizon, and model class for a particular day (i.e., time intervals from the plurality of subsequences of the time series data/ period of time associated with the motif/ class); [0025] select the best scoring models according to the future time range to forecast; [0037] the models may include lags from target time series; a target time series is a time series for which a model is used to generate predictions; [0039] as a part of generating the values and for each day to be predicted, the model having the model horizon with a lowest value that is greater than or equal to an ordinal value of the day to be predicted may be selected for that day to generate a piecewise forecast for the time series variable being predicted - thus, each class/ motif associated with plurality of subsequences of the time series data/ period of time comprising plurality of time intervals/ individual days; [0046] cross validation metrics are generated; different classes of models may use different data for testing and generating cross validation metrics; in-season models may be tested against in-season data, out-of-season model may be tested against out-of-season data, and all-season models may be tested against all-season data; the cross validation metrics may include a mean absolute scaled error (MASE), a weighted mean absolute percentage error (wMAPE) for the models, mean absolute percentage error (MAPE), root mean square error (RMSE), explained variance (R2), and mean absolute error (MAE); [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error (i.e., minimum associated error); [0076] the computing system or group of computing systems include functionality to perform a variety of operations disclosed);
(ii) selecting the forecasting algorithm for the motif based on the forecasting algorithm having a lowest cumulative forecast error ([0020] a model class identifies, for a model, the class of the model; the classes (i.e., set of motifs) may relate to seasonality, which is how the model performs with respect to a particular period of time; [0023] for a given schedule of individual days/ points in time for which predictions are to be made (i.e., predicting the interval value), the system may select the best model by forecast group, prediction horizon, and model class for a particular day ; [0025] select the best scoring models according to the future time range to forecast; [0046] cross validation metrics are generated; different classes of models may use different data for testing and generating cross validation metrics; in-season models may be tested against in-season data, out-of-season model may be tested against out-of-season data, and all-season models may be tested against all-season data; the cross validation metrics may include a mean absolute scaled error (MASE), a weighted mean absolute percentage error (wMAPE) for the models, mean absolute percentage error (MAPE), root mean square error (RMSE), explained variance (R2), and mean absolute error (MAE); [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error (i.e., lowest cumulative forecast error));
generating a mapping that associates each motif of the set of motifs with a corresponding selected forecasting algorithm ([0048] trains and uses models to predict time series data by a model discovery service; [0049] after training the list may include the best models selected based on the cross validation metrics of the trained models; the list of best models may reference the models by the horizon and class of the model so that “M(H,C)” identifies model “M” for the horizon “H” and the class “C” (i.e., mapping from each class/ motif to its selected model));
determining a default forecasting algorithm that performed the best for time intervals that are not associated with any motif of the set of motifs ([0020] the “out-of-season” class of models may be trained and/or scored using data that is “out of season”, e.g., during the months outside of February through April; [0046] out-of-season model may be tested against out-of-season data; [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error - thus, the out-of-season class covers the time intervals not associated with motif and it's best scoring model is the default forecasting algorithm);
receiving a request to forecast the interval value at a future time interval by the computing device ([0025] select the best scoring models according to the future time range to forecast, and provide this selection of models to the demand prediction service; [0058] FIG. 4 illustrates how the piecewise model selection displayed on the graphical user interface (400); [0063] the row (410) displays the date for the predicted days; the dates range from January 5 to January 13 (i.e., future time interval); [0064] the row (412) displays a piecewise prediction forecast generated based on multiple models; [0037] the process generates predictions used to generate estimated headcounts (i.e., forecast interval value) from models with multiple horizons and classes; [0076] the computing system or group of computing systems include functionality to perform a variety of operations disclosed);
determining whether the future time interval is associated with a motif of the set of motifs by the computing device ([0050] classes are determined for each date; [0051] for each of the identified dates, the class that applies to that date is identified; the classes may include in-season, out-of-season, and all-season classes, and one of the classes may be assigned to each date for which a prediction will be generated - thus, determining whether the date falls within a motif-associated period or an out-of-season period);
in response to determining the future time interval is associated with a motif of the set of motifs, using the mapping to select the forecasting algorithm for the motif associated with the future time interval ([0052] the models may be returned by date in the form of a list of tuples; the model “M(h,c)” is the best model from the model space for the horizon “h” and the class “c” selected by the model discovery service based on cross validation metrics for the models; [0064] the model “T(1,A)” is used to generate the predicted value of “8” for the predicted day “0” that corresponds to the date of January 5 using a horizon of 1 and a class of A; the class A identify the model as an in-season model);
in response to determining the future time interval is not associated with any motif of the set of motifs, selecting the default forecasting algorithm ([0052] the models may be returned by date in the form of a list of tuples; the model “M(h,c)” is the best model from the model space for the horizon “h” and the class “c” selected by the model discovery service based on cross validation metrics for the models - thus, a date to which the out-of-season class applies is assigned the best out-of-season model M(h,c); [0059] model T(3,B), T(5,B); [0060] class B may be an out-of-season class);
using the selected forecasting algorithm to predict the interval value for the future time interval by the computing device ([0023] for a given schedule of individual days/ points in time for which predictions are to be made, the system may select the best model by forecast group, prediction horizon, and model class for a particular day; [0025] select the best scoring models according to the future time range to forecast; [0037] the process generates predictions used to generate estimated headcounts (i.e., forecast interval value) from models with multiple horizons and classes; [0056] predictions are made for each date of the prediction schedule using the trained models identified in Step 306 from the model schedule; [0058] FIG. 4 illustrates how the piecewise model selection displayed on the graphical user interface (400); [0060] the row (404) displays the classes of the models used to generate the predictions of the row (412); [0064] the row (412) displays a piecewise prediction forecast generated based on multiple models; the model “T(1,A)” is used to generate the predicted value of “8” for the predicted day “0” that corresponds to the date of January 5 using a horizon of 1 and a class of A; [0042] the estimated headcount generated using the Erlang-C traffic modeling formula with the values from the piecewise forecast and displayed on a client computing device; [0076] the computing system or group of computing systems include functionality to perform a variety of operations disclosed);
providing the predicted interval value for the future time interval by the computing device ([0037] the process generates predictions used to generate estimated headcounts (i.e., forecast interval value) from models with multiple horizons and classes; [0058] FIG. 4 illustrates how the piecewise model selection displayed on the graphical user interface (400); [0064] the row (412) displays a piecewise prediction forecast generated based on multiple models; the model “T(1,A)” is used to generate the predicted value of “8” for the predicted day “0” that corresponds to the date of January 5 using a horizon of 1 and a class of A; [0042] the estimated headcount generated using the Erlang-C traffic modeling formula with the values from the piecewise forecast and displayed on a client computing device); and
scheduling one or more workers to work during the future time interval based on the predicted interval value by the computing device ([0037] the process generates predictions used to generate estimated headcounts from models with multiple horizons and classes; [0001] call centers need to have an appropriate headcount of staff available to handle the amount of calls received; [0017] the headcount identifies how many representatives are needed to be able to handle the expected amount of calls; the headcount identifies the number of representatives needed over a period of time).
However, Yocum fails to expressly teach wherein receiving a window size by the computing device; calculating a matrix profile for the received time series data based on the received window size; determining a plurality of subsequences of the time series data using the calculated matrix profile; determining a set of motifs from the plurality of subsequences using the calculated matrix profile, wherein each motif of the set of motifs is associated with a plurality of the subsequences and each motif of the set of motifs has a size equal to the received window size.
In the same field of endeavor, Law teaches wherein receiving a window size by the computing device ([0172] computing the matrix profile allows quickly finding motifs and identifies the nearest neighbor for subsequences within the time series; [0173] the stump function receives two parameters: a time series and a window size, m; [0109] a UI element allows a user to specify a pattern length); calculating a matrix profile for the received time series data based on the received window size ([0172] computing the matrix profile allows quickly finding motifs and identifies the nearest neighbor for subsequences within the time series; [0173] the stump function receives two parameters: a time series and a window size, m; [0200] calculates the matrix profile with a window size of m=20; [0109] based on that pattern length, a matrix profile and a matrix profile index are calculated); determining a plurality of subsequences of the time series data using the calculated matrix profile ([0173] the output of stump is an array that contains all of the matrix profile values, such as, z-normalized Euclidean distance to the nearest neighbor and matrix profile indices; the following code can be used to plot the matrix profile below the raw data, as shown in FIG. 15; [0174] the global minima (vertical dashed lines) of the matrix profile correspond to the locations of the two subsequences (i.e., plurality of subsequences of the time series data using the matrix profile) that make up the motif pair; [0069] the matrix profile 320 at line 324 is a measure of the similarity between the subsequence 332 and the most similar subsequence from among all the other possible subsequences of length m in the time-series data 300); determining a set of motifs from the plurality of subsequences using the calculated matrix profile, wherein each motif of the set of motifs is associated with a plurality of the subsequences and each motif of the set of motifs has a size equal to the received window size ([0173] the output of stump is an array that contains all of the matrix profile values, such as, z-normalized Euclidean distance to the nearest neighbor and matrix profile indices; the following code can be used to plot the matrix profile below the raw data, as shown in FIG. 15; [0174] the global minima (vertical dashed lines) of the matrix profile correspond to the locations of the two subsequences that make up the motif pair; [0066] an indication 304 of length m demonstrates the subsequence length (also called pattern length) that will be studied in the time-series data; [0173] the stump function receives two parameters: a time series and a window size, m - thus, the pattern length/ size of motifs equal to the received window size, m).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have incorporated wherein receiving a window size by the computing device; calculating a matrix profile for the received time series data based on the received window size; determining a plurality of subsequences of the time series data using the calculated matrix profile; determining a set of motifs from the plurality of subsequences using the calculated matrix profile, wherein each motif of the set of motifs is associated with a plurality of the subsequences and each motif of the set of motifs has a size equal to the received window size, as taught by Law into Yocum. Doing so would be desirable because it would allow for identifying favorable trends and making favorable predictions (Law [0004]), thereby enhancing user experience.
However, Yocum and Law fail to expressly teach wherein selecting the forecasting algorithm based on the associated forecast error comprises: (i) for each forecasting algorithm of the plurality of forecasting algorithms: sequentially processing each subsequence of the plurality of subsequences of the time series data associated with the motif in a chronological order of occurrence in the time series data by: (A) training the forecasting algorithm to predict the interval value using a portion of the time series data that is chronologically before the subsequence in the time series data; and (B) for each time interval in the subsequence: (1) using the forecasting algorithm to predict the interval value for the time interval; (2) determining a difference between the interval value predicted for the time interval and the interval value associated with the time interval in the time series data; and (3) accumulating the determined difference into a cumulative forecast error for the forecasting algorithm; and (ii) selecting the forecasting algorithm based on a lowest cumulative forecast error.
In the same field of endeavor, Bledsoe teaches wherein selecting the forecasting algorithm based on the associated forecast error comprises: (i) for each forecasting algorithm of the plurality of forecasting algorithms ([0082] each of the entrant forecasting models is trained and tested to produce estimates or forecast of time series data points for a given testing period(s) of time; [0087]each entrant forecasting model from the set of entrant forecasting model are tested over more than one time period; [0096] the process continues until the current entrant forecasting model is not the last entrant forecasting model): sequentially processing each subsequence of the plurality of subsequences of the time series data associated with the motif in a chronological order of occurrence in the time series data ([0083] entrant forecasting models are tested during multiple test periods i.e., cross-validation; [0089] the training sets are made of data points that were sampled during a time period earlier than the time period when data points of the testing sets were sampled; [0101] training datasets 701A, 701B, and 701C have overlapping data points, that is, training dataset 701B includes all the data points of training dataset 701C, and training dataset 701A includes all the data points of training dataset 701B and hence, 701C (i.e., training datasets are nested and expanding training windows ending at successively later points); [0103] testing periods A, B, and C respectively labeled as 703A, 703B, and 703C in FIG. 7; an entrant forecasting model trained with training dataset 701C can be tested during testing period 703C; estimation failure or imprecision can be overcome by subsequently training the entrant forecasting model with training sets 701A and 701B - thus, each successive training sets ends at a later point in the time series and each testing period is forecast using only data preceding it, so the testing periods are evaluated in time order (i.e., sequentially processing subsequences of time series data in a chronological order of occurrence in time series data)); by: (A) training the forecasting algorithm to predict the interval value using a portion of the time series data that is chronologically before the subsequence in the time series data ([0089] training and testing data sets can be defined in multiple ways as long as the training sets are made of data points that were sampled during a time period earlier than the time period when data points of the testing sets were sampled; [0090] TSF server 101 trains an entrant forecasting model using data points of the time series included in the first set/ training set; [0101] training datasets 701A, 701B, and 701C have overlapping data points, that is, training dataset 701B includes all the data points of training dataset 701C, and training dataset 701A includes all the data points of training dataset 701B and hence, 701C (i.e., nested training datasets); [0103] testing periods A, B, and C respectively labeled as 703A, 703B, and 703C in FIG. 7; an entrant forecasting model trained with training dataset 701C can be tested during testing period 703C; estimation failure or imprecision can be overcome by subsequently training the entrant forecasting model with training sets 701A and 701B - thus, each successive training sets ends at a later point in the time series and each testing period is forecast using only data preceding it, so the testing periods are evaluated in time order (i.e., processing in a chronological order of occurrence in time series data)); and (B) for each time interval in the subsequence: (1) using the forecasting algorithm to predict the interval value for the time interval ([0082] each of the entrant forecasting models is trained and tested to produce estimates or forecast of time series data points for a given testing period(s) of time; [0090] TSF server 101 trains an entrant forecasting model using data points of the time series included in the first set/ training set; [0104] forecasting accuracy scores are generally computed as a function of a comparison between a forecasted data point for time i with a time series data point sampled or observed at time i); (2) determining a difference between the interval value predicted for the time interval and the interval value associated with the time interval in the time series data ([0091]-[0092] TSF server 101 calculates forecast accuracy scores for the entrant forecasting models based on a comparison of the set of forecasted data points with data points included in the second set/ testing set); a forecast error can be defined as: e i =y i −ŷ i where yi is the ith observed or sampled data point of a testing set and ŷi is a forecasted data point); and (3) accumulating the determined difference into a cumulative forecast error for the forecasting algorithm ([0092]-[0093] calculating a Mean Absolute Error (MAE) where n is the number of pairs (yi, ŷi) used during a testing phase of an entrant forecasting model; MAE measures the average magnitude of a set of errors |ei| in a set of forecasted data points |ŷi| as the average of the absolute difference between forecasted data points and actual observation (i.e., per-point differences are summed); [0104] TSF server 101 computes an overall forecasting accuracy score for a given entrant forecasting model by computing descriptive statistics (e.g., mean, mode or other suitable descriptive statistics) on a set of forecasting accuracy scores computed for an entrant forecasting model. (i.e., errors from each testing period are accumulated into one overall error for the model) and (ii) selecting the forecasting algorithm based on a lowest cumulative forecast error ([0084 ]TSF server 101 selects an entrant forecasting model having superior forecasting scores than the rest of the models I; [0099] TSF server 101, stores each of the trained entrant forecasting models and their computed forecast accuracy scores in memory and uses the forecast accuracy scores computed for each of the entrant forecasting models to select entrant forecast model).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have incorporated wherein selecting the forecasting algorithm based on the associated forecast error comprises: (i) for each forecasting algorithm of the plurality of forecasting algorithms: sequentially processing each subsequence of the plurality of subsequences of the time series data associated with the motif in a chronological order of occurrence in the time series data by: (A) training the forecasting algorithm to predict the interval value using a portion of the time series data that is chronologically before the subsequence in the time series data; and (B) for each time interval in the subsequence: (1) using the forecasting algorithm to predict the interval value for the time interval; (2) determining a difference between the interval value predicted for the time interval and the interval value associated with the time interval in the time series data; and (3) accumulating the determined difference into a cumulative forecast error for the forecasting algorithm; and (ii) selecting the forecasting algorithm based on a lowest cumulative forecast error, as taught by Bledsoe into Yocum and Law. Doing so would be desirable because it would provide self-adaptive forecasting systems that can identify significant data affecting future values of a time-varying phenomena, and able to adapt and optimize forecasting techniques demanded by highly volatile environments (Bledsoe [0003]).
As to dependent Claim 3, Yocum and Law teach all the limitations of Claim 1. Yocum further teaches wherein the interval value is one of a communication volume, an average handling time, or a shrinkage ([0037] the process generates predictions used to generate estimated headcounts (i.e., interval value) from models with multiple horizons and classes; [0017] the headcount identifies how many representatives are needed to be able to handle the expected amount of calls).
As to dependent Claim 5, Yocum and Law teach all the limitations of Claim 1. Yocum further teaches wherein selecting the forecasting algorithm based on the associated forecast error comprises selecting the forecasting algorithm with a minimum forecast error ([0045] the training data used to train the models may include historical time series data for at least two years; [0046] cross validation metrics are generated; different classes of models may use different data for testing and generating cross validation metrics; in-season models may be tested against in-season data, out-of-season model may be tested against out-of-season data, and all-season models may be tested against all-season data; the cross validation metrics may include a mean absolute scaled error (MASE), a weighted mean absolute percentage error (wMAPE) for the models, mean absolute percentage error (MAPE), root mean square error (RMSE), explained variance (R2), and mean absolute error (MAE); [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error (i.e., minimum forecast error)).
As to dependent Claim 21, Yocum and Law teach all the limitations of Claim 1. Yocum further teaches wherein selecting the forecasting algorithm for a motif comprises calculating a forecast error using absolute differences between predicted interval values and observed interval values averaged across subsequences associated with the motif ([0045] the training data used to train the models may include historical time series data for at least two years; [0046] cross validation metrics are generated; different classes of models may use different data for testing and generating cross validation metrics; in-season models may be tested against in-season data, out-of-season model may be tested against out-of-season data, and all-season models may be tested against all-season data (i.e., models based on the subsequences of the time series data associated with the class/ days associated with the season/ motif); the cross validation metrics may include a mean absolute scaled error (MASE), a weighted mean absolute percentage error (wMAPE) for the models, mean absolute percentage error (MAPE), root mean square error (RMSE), explained variance (R2), and mean absolute error (MAE) - these averages absolute differences between predicted and observed values across the data associated with each class model/ subsequences associated with the motif; [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error).
As to dependent Claim 22, Yocum and Law teach all the limitations of Claim 1. Yocum further teaches wherein the forecast error comprises a mean absolute percentage error ([0045] the training data used to train the models may include historical time series data for at least two years; [0046] cross validation metrics are generated; different classes of models may use different data for testing and generating cross validation metrics; in-season models may be tested against in-season data, out-of-season model may be tested against out-of-season data, and all-season models may be tested against all-season data; the cross validation metrics may include a mean absolute scaled error (MASE), a weighted mean absolute percentage error (wMAPE) for the models, mean absolute percentage error (MAPE), root mean square error (RMSE), explained variance (R2), and mean absolute error (MAE); [0047] models are selected based on cross validation metrics; the model discovery service may select the model with the lowest mean absolute scaled error or the lowest weighted mean absolute percentage error).
As to dependent Claim 24, Yocum, Law, and Bledsoe teach all the limitations of claim 1. Yocum further teaches wherein the computing device comprises a display device, and wherein providing the predicted interval value comprises displaying, on a graphical user interface of the display device, a visual forecast dashboard comprising: (i) a visual representation of the selected motif in the time series data and (ii) the predicted interval value for the future time interval ([0070] output device, such as a screen (e.g., a liquid crystal display (LCD); [0058] the piecewise model selection and stitching work and may be displayed on the graphical user interface (400); [0059] displays the model schedule; the model schedule includes multiple trained models for multiple horizons and multiple classes; [0064] row (412) displays a piecewise prediction forecast generated based on multiple models; [0065] selection of the button (422) may display a menu for selecting a forecast group for which to display predictions).
Claims 8, 10, and 12 are system claims that are corresponding to the method claims 1, 3, 5 respectively and therefore, rejected for the same reasons. Yocum further teaches a system ([0067] computing system) comprising: one or more processors ([0067] computing system include one or more computer processors); and a memory storing instructions that when executed by the one or more processors cause the system to perform steps ([0067] computing system include one or more computer processors, non-persistent storage (e.g., volatile memory, such as random access memory (RAM), cache memory), persistent storage (e.g., a hard disk, an optical drive such as a compact disk (CD) drive or digital versatile disk (DVD) drive, a flash memory, etc.); [0071] software instructions in the form of computer readable program code to perform embodiments of the invention may be stored on a non-transitory computer readable medium; the software instructions may correspond to computer readable program code that, when executed by a processor(s), is configured to perform one or more embodiments of the invention).
Claims 15 and 20 are medium claims that are corresponding to the method claims 1 and 5 respectively and therefore, rejected for the same reasons. Yocum further teaches a non-transitory computer-readable medium storing instructions that when executed by one or more processors of a system cause the system to perform steps ([0067] computing system include one or more computer processors, non-persistent storage (e.g., volatile memory, such as random access memory (RAM), cache memory), persistent storage (e.g., a hard disk, an optical drive such as a compact disk (CD) drive or digital versatile disk (DVD) drive, a flash memory, etc.); [0071] software instructions in the form of computer readable program code to perform embodiments of the invention may be stored on a non-transitory computer readable medium; the software instructions may correspond to computer readable program code that, when executed by a processor(s), is configured to perform one or more embodiments of the invention).
Response to Arguments
35 U.S.C. §101: In the remarks, Applicant argues that:
(a) the claims recite a specific technological process for "sequentially processing each subsequence" in a "chronological order", "training the forecasting algorithm" using historical data occurring "chronologically before the subsequence", and "accumulating the determined difference into a cumulative forecast error". This provides a specialized rolling window evaluation loop to improve the accuracy of the selected machine learning models.
(b) the amended claims are directed to a technological improvement to a forecasting system – “… the accuracy of the forecasts will be improved which is an improvement to any technological field that relies on forecasting.”
(c) the claims recite significantly more than an abstract idea. The cited prior art fails to show that this specific combination of chronological retraining, cumulative error calculation, and branched fallback logic is well understood, routine, or conventional in the art.
Examiner respectfully disagrees with Applicant’s arguments.
As to point (a), the recited steps of calculating a matrix profile; determining a plurality of subsequences of the time series data using the calculated matrix profile; determining a set of motifs from the plurality of subsequences using the calculated matrix profile; selecting a forecasting algorithm based on an associated forecast error when predicting the interval value for time intervals from the plurality of subsequences of the time series data associated with the motif, etc. encompasses a person to perform these steps mentally, such as a person making an evaluation/ judgement to calculate matrix profile, identify patterns/ motifs using the calculated matrix profile (mathematical calculations), select a forecasting algorithm for a motif based on forecasting error (mathematical relationship), decide a motif associated with a future time interval and use the forecasting algorithm (mathematical calculations) associated with the motif to predict/ calculate interval value. For each subsequence, the claim predicts a value with a forecasting function, subtracts it from the observed value, and adds the difference to a running total. It then chooses the algorithm with the smallest total. Processing subsequences “sequentially… in a chronological order” specifies only the order in which the calculations are performed. The training step (A) is analyzed as an additional element at Step 2A, Prong 2 and Step 2B. Accordingly, the claims recite an abstract idea that falls under the “mathematical concepts” and “mental process” grouping. See the detailed analysis under 35 U.S.C. 101 rejections above. Regarding applicant’s remarks, the “rolling window evaluation loop to improve the accuracy of the selected machine learning models”, any purported improvement is based on the recited steps of mathematical analysis and mental process, as analyzed in detail under Step 2A, Prong 1 portion of the 35 U.S.C. 101 rejections above. Per MPEP § 2106.05 (a), “It is important to note, the judicial exception alone cannot provide the improvement”. Further, the recited claim merely involves, at most, an improvement to the abstract idea itself with the aid of a generic machine learning tools. Examiner notes that “It is important to keep in mind that an improvement in the abstract idea itself is not an improvement in technology” (see MPEP 2106.05(a)(II)).
As to point (b), any purported improvement is based on the recited steps of mathematical analysis and mental process, as analyzed in detail under Step 2A, Prong 1 portion of the 35 U.S.C. 101 rejections above. Per MPEP § 2106.05 (a), “It is important to note, the judicial exception alone cannot provide the improvement”. Further, the recited claim merely involves, at most, an improvement to the abstract idea itself with the aid of a generic machine learning tools. Examiner notes that “It is important to keep in mind that an improvement in the abstract idea itself is not an improvement in technology” (see MPEP 2106.05(a)(II)). See the detailed analysis under 35 U.S.C. 101 rejections above.
As to point (c), the WURC inquiry at Step 2B applies to the additional elements, not to the judicial exception. Generating a mapping/ lookup table, determining whether a time falls within a motif, choosing between the mapped algorithm and a default and accumulating errors are part of the abstract idea identified at Step 2A, Prong 1. The additional elements are the receiving and providing steps, the generic computing device, the generic training step, and scheduling, which amount to insignificant extra-solution activity of mere data gathering that is a well-understood, routine, and conventional activity (see MPEP § 2106.05(d), “receiving/ transmitting data”, “presenting offers”); merely using computer as a tool, by using it in its ordinary capacity to perform tasks (see MPEP § 2106.05(f)) and generally linking the use of a judicial exception to the field of generic computer (see MPEP § 2106.05(h)); an instruction to apply the abstract idea using a generic machine learning tool (see MPEP § 2106.05(f)) and generally linking the use of a judicial exception to the field of machine learning (see MPEP § 2106.05(h)); and merely indicating the field of use/ workforce planning in which to apply the judicial exception (see MPEP § 2106.05(h)). Examiner asserts that the additional elements in the pending claims, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. Thus, the claims are not patent eligible in Step 2B as further detailed in the 35 U.S.C. 101 rejections above.
Accordingly, Applicant’s arguments concerning the §101 rejections are not persuasive.
35 U.S.C. §103: In the remarks, applicant argues that the cited references do not teach the features recited in amended independent claims 1, 8, and 15.
Applicant's arguments with respect to the 103 rejections have been considered, but are moot in view of new ground of rejection made under 35 U.S.C. § 103.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 CFR § 1.111(c) to consider these references fully when responding to this action.
Shahand et al. (US 2018/0322400 A1) teaches: a plurality of test forecasts are generated with respective forecasting mechanisms of a plurality of forecasting mechanisms using a first subset of the plurality of data points; errors are then determined for the respective forecasting mechanisms, such as based on comparisons of corresponding ones of the plurality of test forecasts and a second subset of the plurality of data points; one of the plurality of forecasting mechanisms is selected based on the errors; an output forecast is then generated with the selected forecasting mechanism using the first and second subsets of the plurality of data points (see Abstract).
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 extension fee 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 REJI KARTHOLY whose telephone number is (571)272-3432. The examiner can normally be reached on Monday - Thursday 7:30 am - 3:30 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jennifer Welch, can be reached at telephone number (571)272-7212. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/REJI KARTHOLY/Primary Examiner, Art Unit 2143
