DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Notice to Applicant
The following is a Final Office action. In response to Examiner’s Non-Final Rejection of 08/28/2025, Applicant, on 01/28/2026, did not amend or add any claims, rather only arguments are presented. Claims 1-21 are pending in this application and have been rejected below.
Response to Arguments
Applicant's arguments filed 01/28/2026 have been fully considered, but they are not fully persuasive. The 35 USC § 103 rejection of claims 1-21 is maintained.
The Applicant argues that Taylor fails to teach the decision to process locally or send to a difference computing device and fails to a predicted change to a concurrency requirement. In response, the Examiner disagrees, and does not find the arguments persuasive. Firstly, the limitation of deciding, by the computing device, on the basis of the at least one change in the at least one metric associated with processing the first computer data stream, whether to process the second computer data stream by the computing device, is taught b Appleby, not Taylor, thus the Applicant is refuting the wrong reference. The citations from Appleby below: Appleby 0006: “a method and system dynamically allocate servers among a plurality of connected server compute-resources and a free-pool of servers. Each server compute-resource comprises a plurality of servers. Each server allocated to a compute-resource is monitored for one metric…0083-0086: “Step 212: A check is made to see if allocations are enabled. Steps 300-324: Servers are reallocated if thresholds are violated. It is important to note that “double exponential smoothing” or some other kind of data smoothing should always be used to remove temporary metric peaks and valleys. Smoothing can be performed at one or more steps in the method. For example, time smoothing can be performed when metrics are originally measured (before calculation of P values), on P values after the P values are calculated, and/or on G values after G values are calculated.” Attacking Taylor alone or a limitation mapped to Appleby is improper where the rejection is a combination, see MPEP 2145(IV). Additionally, Taylor teaches concurrent processing of partitioned streams. See Taylor citations below: 0022: “Two or more pipelines of computations can then operate in parallel on two or more independent data streams…0107: a single input data stream is partitioned into multiple data streams. For example, a financial market data feed may transmit updates for one thousand financial instruments on a single UDP multicast channel. This feed may be partitioned into multiple data streams based on the financial instruments to which the update messages apply. For example, the feed may be partitioned into eight data streams where each data stream has an update rate of approximately one eighth of the feed. If the multiple input data streams of FIG. 7 have the same data rate, then this embodiment achieves a throughput that is N times higher than the embodiment of FIG. 6.” A throughput requirement driving stream partitioning across pipelines, which is the claimed concurrency requirement. Taylor also teaches the forked execution path, matching locally versus routing away to a difference market; 0109: “must either match or improve those prices or route the orders away to a market with a superior price. Similarly, arbitrage trading strategies must be able to identify superior and inferior prices across multiple markets in order to identify profitable trading opportunities.” Lastly, Lei supplies the prediction driving the decision. See Li 024: “a real time threshold prediction module 242 that, based upon real time operational information, predicts if and when a performance measure for a work item queue will cross a threshold.” Li’s prediction feeds Appleby’s threshold-based reallocation decision, applied in Taylors partitioned parallel architecture. Thus, taking the rejection as a whole, one of ordinary skill in the art would agree that the combination of the references teaches all the limitations of claim 1.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3-5, 7-9, 12, 14-16, and 18-21 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent 8175254 (hereinafter “Li”) et al., in view of U.S. PGPub 20070233866 to (hereinafter “Appleby”) et al., in further view of U.S. PGPub 20200364791 to (hereinafter “Taylor”) et al.
As per claim 1, Li teaches A computer implemented method for, upon receipt of a second computer data stream, predicting a change in processing a first computer data stream, the method comprising:
receiving, at a computing device, the first computer data stream; generating, by the computing device, a first data sequence comprising a time of receipt of the first computer data stream;
receiving, at the computing device, the second computer data stream; generating, by the computing device, a second data sequence comprising a time of receipt of the second computer data stream; sending, by the computing device, the first and second data sequences to a prediction model;
Li 010-012: “The switch 130 and/or server 110 can be any architecture for directing contacts to one or more communication devices. In some embodiments, the switch 130 may perform load-balancing functions by allocating incoming or outgoing contacts among a plurality of logically and/or geographically distinct contact centers… in the server 110 is a work item routing processwork item routing process 216. Contacts incoming to the contact center are assigned by work item routing process 216 to different contact queues 208a-n based upon a number of predetermined criteria, including customer identity, customer needs, contact center needs, current contact center queue lengths, customer value, and the agent skill that is required for the proper handling of the contact. Agents who are available for handling contacts are assigned to agent queues 212a-n based upon the skills that they possess. An agent may have multiple skills, and hence may be assigned to multiple agent queues 212a-n simultaneously. Furthermore, an agent may have different levels of skill expertise (e.g., skill levels 1-N in one configuration or merely primary skill levels and secondary skill levels in another configuration), and hence may be assigned to different agent queues 212a-n at different expertise levels…045: The threshold exception is calculated from the current real time operational information. Threshold growth rate is calculated from the number of calls in queue for the threshold. Current threshold processing rate is calculated from the current average agent handle time, the number of agents staffed, and the abandon rate. Based on the incoming threshold growth rate and the threshold processing rate, it is estimated that the first queue will exceed its KPI of 10 second ASA in three minutes and face a $20,000 penalty.” Examiner Note: The art teaches the ability to receive and route a plurality (one or two) incoming packets (data streams), and route them to a plurality of contact center agents.
predicting, by the prediction model, at least one change in at least one metric associated with processing the first computer data stream, the predicted change based at least in part on the first data sequence and the second data sequence,; and sending, by the prediction model, to the computing device, the at least one change in the at least one metric associated with processing the first computer data stream…;
Li 024: The performance reporting module 238 includes a real time threshold prediction module 242 that, based upon real time operational information, predicts if and when a performance measure for a work item queue will cross a threshold, determines the penalty (e.g., financial impact) for crossing the threshold and/or provides a set of recommendations or corrective actions to avoid violating the threshold and/or reduce the penalty associated with threshold violation. The performance reporting module may be incorporated into the operational contact center reporting module (not shown)… the module 242 presents the subset of recommendations to an administrator or automatically selects the "best" recommendation…the module 242 provides the subset of recommendations to the work item routing process 216 and/or resource contact selection module 232 for implementation. The routing vector settings would be adjusted based on the recommendation(s).”Examiner Note: The art teaches a prediction model that analyzes the data streams and the processing computers, and sends/routes them based on the prediction model analysis.
Li may not explicitly teach the following. However, Appleby teaches:
deciding, by the computing device, on the basis of the at least one change in the at least one metric associated with processing the first computer data stream, whether to process the second computer data stream by the computing device; Appleby 0006: “a method and system dynamically allocate servers among a plurality of connected server compute-resources and a free-pool of servers. Each server compute-resource comprises a plurality of servers. Each server allocated to a compute-resource is monitored for one metric…0083-0086: “Step 212: A check is made to see if allocations are enabled. Steps 300-324: Servers are reallocated if thresholds are violated. It is important to note that “double exponential smoothing” or some other kind of data smoothing should always be used to remove temporary metric peaks and valleys. Smoothing can be performed at one or more steps in the method. For example, time smoothing can be performed when metrics are originally measured (before calculation of P values), on P values after the P values are calculated, and/or on G values after G values are calculated.”
Li and Appleby are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li with the aforementioned teachings from Appleby with a reasonable expectation of success, by adding steps that allow the software to utilize multiple resources with the motivation to more efficiently and accurately organize and analyze data [Appleby 0083].
Li and Appleby may not explicitly teach the following. However, Taylor teaches:
sequence, the predicted change being due to a requirement for processing the second computer data stream by the computing device concurrently with the first computer data stream… concurrently with the first computer data stream, or to send the second computer data stream to be processed by the different computing device… and processing the second computer data stream by the computing device or sending the second computer data stream to be processed by a different computing device based on the decision;Taylor 0022: “Computations can be sub-divided into two or more pipeline stages, where each pipeline stage operates in parallel to the other pipeline stages. Two or more pipelines of computations can then operate in parallel on two or more independent data streams. In some embodiments, the output of multiple pipelines can be combined to produce aggregated trading signals and estimators. As noted, these solutions can be implemented with a variety of parallel processing technologies that include, general purpose processors that contain multiple Central Processing Unit (CPU) cores (e.g., CMPs), application-specific firmware logic in FPGAs, and GPUs that contain numerous compute cores…0031-0039: “Such derivative trading signals are predicated on the ability to predict or detect market dynamics with each market data update event. Example embodiments disclosed herein include the ability to generate derivative trading signals for Tier 2 and 3 market participants by consuming and summarizing the output of trading signals computed at low latency for Tier 1 market participants. The efficiency of this approach allows for a wide range of time intervals for delivery of derivative signals…FIG. 7 shows multiple example independent signal generation pipelines operating in parallel, processing independent data streams. The transmit stage delivers messages from each pipeline to downstream consumers, where a given consumer may receive messages from one or more pipelines. FIG. 8 shows an example data aggregate stage consuming the output of multiple independent pipelines. A signal generate stage is optionally interposed between processing steps by the data aggregate stage. Another signal generate stage is interposed between the data aggregate stage and a transmit stage and it operates in parallel to transmitting messages output from the data aggregate stage to downstream consumers.”Examiner note: Matching, concurrently with parallel processing.
Li, Appleby, and Taylor are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li and Appleby with the aforementioned teachings from Taylor with a reasonable expectation of success, by adding steps that allow the software to utilize processing information with the motivation to more efficiently and accurately organize and analyze data [Taylor 0039].
As per claim 3, Li, Appleby, and Taylor teach all the limitations of claim 1.
In addition, Li teaches:
wherein deciding, by the computing device, whether to process the second computer data stream concurrently with the first computer data stream, or send the second computer data stream to be processed by the different computing device, is further based on a concurrency threshold; Li 012: in the server 110 is a work item routing processwork item routing process 216. Contacts incoming to the contact center are assigned by work item routing process 216 to different contact queues 208a-n based upon a number of predetermined criteria, including customer identity, customer needs, contact center needs, current contact center queue lengths, customer value, and the agent skill that is required for the proper handling of the contact. Agents who are available for handling contacts are assigned to agent queues 212a-n based upon the skills that they possess. An agent may have multiple skills, and hence may be assigned to multiple agent queues 212a-n simultaneously. Furthermore, an agent may have different levels of skill expertise (e.g., skill levels 1-N in one configuration or merely primary skill levels and secondary skill levels in another configuration), and hence may be assigned to different agent queues 212a-n at different expertise levels…045: The threshold exception is calculated from the current real time operational information. Threshold growth rate is calculated from the number of calls in queue for the threshold. Current threshold processing rate is calculated from the current average agent handle time, the number of agents staffed, and the abandon rate. Based on the incoming threshold growth rate and the threshold processing rate, it is estimated that the first queue will exceed its KPI of 10 second ASA in three minutes and face a $20,000 penalty.”
As per claim 4, Li, Appleby, and Taylor teach all the limitations of claim 1.
In addition, Li teaches:
predicting by the prediction model, a change in at least one metric associated with processing the second computer data stream upon initiating processing of the second computer data stream concurrently with processing the first computer data stream, by the computing device, the predicted change based at least in part on the first data sequence and the second data sequence; Li 012: “(15) (a) work item queues; (16) (b) agent queues and/or agents servicing work items from the work item queues; and (17) (c) a real time threshold prediction module operable to: (C1) determine that a selected performance measure will, during a future time interval, likely cross a selected threshold; (C2) in response, perform at least one of the following operations: (i) determine a consequence (e.g., financial impact) of crossing the selected threshold; and (ii) determine a set of corrective actions to reduce a likelihood that the selected performance measure will cross the selected threshold. The performance measure is commonly one or more of service level, expected wait time, predicted wait time, actual wait time, number of contacts waiting… Contacts Finished Waiting, Contacts Waiting, Cost per contact (per media), Customer defection rate, Customer Satisfaction, Critical Duration, Curr. Max. Wait Duration, Dequeues, Direct Agent Contacts Waiting, Discussion from On Hold, Error or rework percentage or volume, Exp. Wait, First Call Resolution (First resolution rate),”Note: The art teaches tracking and analyzing wait time (a change in a metric), and performing actions based on said change.
As per claim 5, Li, Appleby, and Taylor teach all the limitations of claim 1.
In addition, Li teaches:
wherein the at least one metric associated with processing the first computer data stream comprises a duration for processing the first computer data stream, and wherein predicting, by the prediction model, comprises predicting a change in a first duration for the computing device to process the first computer data stream, the predicted change based on the first data sequence, the second data sequence, and a second duration for the computing device to process the second computer data stream; Li 012-017: “(15) (a) work item queues; (16) (b) agent queues and/or agents servicing work items from the work item queues; and (17) (c) a real time threshold prediction module operable to: (C1) determine that a selected performance measure will, during a future time interval, likely cross a selected threshold; (C2) in response, perform at least one of the following operations: (i) determine a consequence (e.g., financial impact) of crossing the selected threshold; and (ii) determine a set of corrective actions to reduce a likelihood that the selected performance measure will cross the selected threshold. The performance measure is commonly one or more of service level, expected wait time, predicted wait time, actual wait time, number of contacts waiting… Contacts Finished Waiting, Contacts Waiting, Cost per contact (per media), Customer defection rate, Customer Satisfaction, Critical Duration, Curr. Max. Wait Duration, Dequeues, Direct Agent Contacts Waiting, Discussion from On Hold, Error or rework percentage or volume, Exp. Wait, First Call Resolution (First resolution rate).. (c) a real time threshold prediction module operable to: (C1) determine that a selected performance measure will, during a future time interval, likely cross a selected threshold; (C2) in response, perform at least one of the following operations… First Call Resolution (First resolution rate), Handle Duration, Handle Duration.”Note: The art teaches the ability to predict changes based on changing data (resolution rates).
As per claim 7, Li, Appleby, and Taylor teach all the limitations of claim 1.
In addition, Li teaches:
wherein the first computer data stream represents a plurality of computer data streams being processed by the computing device, the method steps repeated for each computer data stream of the plurality of computer data streams; Li 010-012: “The switch 130 and/or server 110 can be any architecture for directing contacts to one or more communication devices. In some embodiments, the switch 130 may perform load-balancing functions by allocating incoming or outgoing contacts among a plurality of logically and/or geographically distinct contact centers… in the server 110 is a work item routing processwork item routing process 216. Contacts incoming to the contact center are assigned by work item routing process 216 to different contact queues 208a-n based upon a number of predetermined criteria, including customer identity, customer needs, contact center needs, current contact center queue lengths, customer value, and the agent skill that is required for the proper handling of the contact. Agents who are available for handling contacts are assigned to agent queues 212a-n based upon the skills that they possess. An agent may have multiple skills, and hence may be assigned to multiple agent queues 212a-n simultaneously. Furthermore, an agent may have different levels of skill expertise (e.g., skill levels 1-N in one configuration or merely primary skill levels and secondary skill levels in another configuration), and hence may be assigned to different agent queues 212a-n at different expertise levels…045: The threshold exception is calculated from the current real time operational information. Threshold growth rate is calculated from the number of calls in queue for the threshold. Current threshold processing rate is calculated from the current average agent handle time, the number of agents staffed, and the abandon rate. Based on the incoming threshold growth rate and the threshold processing rate, it is estimated that the first queue will exceed its KPI of 10 second ASA in three minutes and face a $20,000 penalty.” Examiner Note: The art teaches the ability to receive and route a plurality (one or two) incoming calls (data streams), and repeat this process.
As per claim 8, Li, Appleby, and Taylor teach all the limitations of claim 1.
In addition, Li teaches:
wherein the first computer data stream and the second computer data stream represent communications being handled in a contact centre; Li 010-012: “The switch 130 and/or server 110 can be any architecture for directing contacts to one or more communication devices. In some embodiments, the switch 130 may perform load-balancing functions by allocating incoming or outgoing contacts among a plurality of logically and/or geographically distinct contact centers… in the server 110 is a work item routing processwork item routing process 216. Contacts incoming to the contact center are assigned by work item routing process 216 to different contact queues 208a-n based upon a number of predetermined criteria, including customer identity, customer needs, contact center needs, current contact center queue lengths, customer value, and the agent skill that is required for the proper handling of the contact. Agents who are available for handling contacts are assigned to agent queues 212a-n based upon the skills that they possess. An agent may have multiple skills, and hence may be assigned to multiple agent queues 212a-n simultaneously. Furthermore, an agent may have different levels of skill expertise (e.g., skill levels 1-N in one configuration or merely primary skill levels and secondary skill levels in another configuration), and hence may be assigned to different agent queues 212a-n at different expertise levels…045: The threshold exception is calculated from the current real time operational information. Threshold growth rate is calculated from the number of calls in queue for the threshold. Current threshold processing rate is calculated from the current average agent handle time, the number of agents staffed, and the abandon rate. Based on the incoming threshold growth rate and the threshold processing rate, it is estimated that the first queue will exceed its KPI of 10 second ASA in three minutes and face a $20,000 penalty.”
Claim 9 is directed to a method for performing nearly identical steps of the method of claims 1 and 3 above. Since Li and Appleby has been shown to teach all the elements of claim 9 (in claim 1 and 3), the same art and rationale apply.
Claims 12-16 and 18-19 are directed to the system for performing the method of claims 1-5 and 7-8 above. Since Li, Appleby, and Taylor teaches the system, the same art and rationale apply.
As per claim 20, Li, Appleby, and Taylor teach all the limitations of claim 12.
In addition, Li teaches:
wherein the first computing device is configured to execute the prediction model; Li 09-022: “Referring to FIG. 2, one possible configuration of the server 110 is depicted. The server 110 is in communication with a plurality of customer communication lines 200a-y (which can be one or more trunks, phone lines, etc.) and agent communication line 204 (which can be a voice-and-data transmission line such as LAN 142 and/or a circuit switched voice line). The server 110 can include a operational contact center reporting module (not shown), such as Avaya IQ.TM., CMS.TM., Basic Call Management System.TM., Operational Analyst.TM., and Customer Call Routing or CCR.TM. by Avaya, Inc., gathers call records and contact-center statistics for use in generating contact-center reports… The resource contact selection module 232 and performance reporting module 238 are stored either in the main memory or in a peripheral memory (e.g., disk, CD ROM, etc.) or some other computer-readable medium of the center 100.”
As per claim 21, Li, Appleby, and Taylor teach all the limitations of claim 9.
In addition, Appleby teaches:
wherein the central server is to decide to assign the incoming computer data stream to be processed by the first computing device if the at least one change in the at least one metric associated with the one or more computer data streams currently being processed by the first computing device is below a predefined threshold, or else to be processed by a second computing device; Appleby 0006: “a method and system dynamically allocate servers among a plurality of connected server compute-resources and a free-pool of servers. Each server compute-resource comprises a plurality of servers. Each server allocated to a compute-resource is monitored for one metric…0083-0086: “Step 212: A check is made to see if allocations are enabled. Steps 300-324: Servers are reallocated if thresholds are violated. It is important to note that “double exponential smoothing” or some other kind of data smoothing should always be used to remove temporary metric peaks and valleys. Smoothing can be performed at one or more steps in the method. For example, time smoothing can be performed when metrics are originally measured (before calculation of P values), on P values after the P values are calculated, and/or on G values after G values are calculated.”
Li and Appleby are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li with the aforementioned teachings from Appleby with a reasonable expectation of success, by adding steps that allow the software to utilize multiple resources with the motivation to more efficiently and accurately organize and analyze data [Appleby 0083].
Claims 6, 10-11, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent 8175254 (hereinafter “Li”) et al., in view of U.S. PGPub 20220180276 to (hereinafter “Silverman”) et al., in further view of U.S. PGPub 20200364791 to (hereinafter “Taylor”) et al., and in further view of U.S. PGPub 20070233866 to (hereinafter “Appleby”) et al.
As per claim 6, Li, Appleby, and Taylor teach all the limitations of claim 1.
Li, Appleby, and Taylor may not explicitly teach the following. However, Silverman teaches:
wherein the prediction model comprises one or more of: a machine learning algorithm; a regression algorithm; a deep learning algorithm; a neural network; a long short term memory neural network; a fully connected neural network; and a convolutional neural network;Silverman 0028-0037: “ the forecasting module 125 may generate a plurality of forecasting models 127 using a variety of methods including machine learning and statistical methods. Each model 127 may be trained with a different set of historical demand data 116 or using different weights and/or heuristics. Any type of prediction model may be used… the training module 120 may analyze the historical demand data 116 for a plurality of past intervals, and the event data 117 for those intervals, to determine which particular events are relevant for the entity associated with the call center. For example, an event such as an awards show on television may not significantly change the call demand for an entity such as a bank, but an event such as a change in interest rates by the federal reserve bank may. The training module 120 may determine those events that have an impact on demand for an entity using machine learning, for example.”
Li, Appleby, Taylor, and Silverman are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li, Appleby, and Taylor with the aforementioned teachings from Silverman with a reasonable expectation of success, by adding steps that allow the software to utilize learning models with the motivation to more efficiently and accurately organize and analyze data [Silverman 0037].
As per claim 10, Li, Appleby, and Taylor teach all the limitations of claim 9.
Li, Appleby, and Taylor may not explicitly teach the following. However, Silverman teaches:
wherein deciding by the central is further based on comparing the number of the one or more computer data streams currently being processed by the first computing device Silverman 0054-0063: “ At 325, whether the difference satisfied a threshold is determined. The determination may be made by the forecasting module 125. The difference may satisfy the threshold when it is less than, or greater than a certain amount. The amount may be set by a user or administrator. If the difference satisfies the threshold, the method 300 may continue to 330. Else the method 300 may return to 310 where the demand is predicted using a different future interval... The graphical user interface 400 further includes a window 450 through which the administrator is made aware of upcoming events that are predicted to affect demand. In particular, the window 450 indicates that call volume is predicted to increase by 10% while chat volume is predicted to drop 5% on Monday evening due to a basketball game and stormy weather. As noted above, when one or more upcoming events are predicted to effect demand by more than a threshold percentage they may be displayed to the administrator.”
Li, Appleby, Taylor, and Silverman are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li, Appleby, and Taylor with the aforementioned teachings from Silverman with a reasonable expectation of success, by adding steps that allow the software to utilize learning models with the motivation to more efficiently and accurately organize and analyze data [Silverman 0037].
As per claim 11, Li, Appleby, and Taylor teaches all the limitations of claim 9.
Li, Appleby, and Taylor may not explicitly teach the following. However, Silverman teaches:
wherein the central server comprises a prediction model, the prediction model comprising one or more of: a machine learning algorithm; a regression algorithm; a deep learning algorithm; a neural network; a long short term memory neural network; a fully connected neural network; and a convolutional neural networkx;Silverman 0028-0037: “ the forecasting module 125 may generate a plurality of forecasting models 127 using a variety of methods including machine learning and statistical methods. Each model 127 may be trained with a different set of historical demand data 116 or using different weights and/or heuristics. Any type of prediction model may be used… the training module 120 may analyze the historical demand data 116 for a plurality of past intervals, and the event data 117 for those intervals, to determine which particular events are relevant for the entity associated with the call center. For example, an event such as an awards show on television may not significantly change the call demand for an entity such as a bank, but an event such as a change in interest rates by the federal reserve bank may. The training module 120 may determine those events that have an impact on demand for an entity using machine learning, for example.”
Li, Appleby, Taylor, and Silverman are deemed to be analogous references as they are reasonably pertinent to each other and directed towards measuring, collecting, and analyzing information with a series of inputs to solve similar problems in the similar environments. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Li, Appleby, and Taylor with the aforementioned teachings from Silverman with a reasonable expectation of success, by adding steps that allow the software to utilize learning models with the motivation to more efficiently and accurately organize and analyze data [Silverman 0037].
Claim 17 is directed to the system for performing the method of claim 6 above. Since Li, Appleby, Taylor, and Silverman teach the system, the same art and rationale apply.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure:
Guerrero; Jose Luis Beltran. CALL CENTER RESOURCE ALLOCATION, .U.S. PGPub 20120087486 A method for determining call center resource allocation can include modeling call center performance over an operations time period using a computer. A number of replicas of the modeled call center performance are simulated, using the computer, over a planning time period, each replica having random contact arrivals and contact service times following a stochastic arrival and service process according to a probability distributions of inter-arrival time and service time. Multiple iterations of each simulation are run on the computer to optimize call center resource allocation.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. 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 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Arif Ullah, whose telephone number is (571) 270-0161. The examiner can normally be reached from Monday to Friday between 9 AM and 5:30 PM.
If any attempt to reach the examiner by telephone is unsuccessful, the examiner’s supervisor, Beth Boswell, can be reached at (571) 272-6737. The fax telephone numbers for this group are either (571) 273-8300 or (703) 872-9326 (for official communications including After Final communications labeled “Box AF”).
/Arif Ullah/
Primary Examiner, Art Unit 3625
Prosecution Insights