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 .
Continued Examination Under 37 CFR 1.114
1. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 08/05/2026 has been entered.
Response to Amendment
2. This in response to an RCE amendment filed 08/05/2026. No claims have been added. Claims 4-6, 10 and 17-19 have been canceled. Claims 1-3, 7-9, 11-16 and 20 have been amended. Claim 1-3, 7-9, 11-16 and 20 are now pending in this application.
Claim Rejections - 35 USC § 103
3. 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-3, 7-9, 11-16 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over McCourt (US PAT # 12/230,253 B2) in view of Jarah et al. (Pub.No.: 2025/0030611 A1) in view of Kolar et al. (2021/0304061 A1) and further in view of Consul et al. (US PAT # 8,620,869 B2).
Regarding claims 1 and 14, McCourt teaches method and system of managing interaction transcripts based on rules (reads on receiving and classifying call transcript using trained classification model, see Figs. 1-3 and corresponding text), the method comprising:
identifying one or more decision parameters in an interaction transcript, the one or more decision parameters comprising interaction metadata (reads on receiving a call transcript that may include timestamps for blocks of text, total call length, party or party-type identifiers, labels, annotation metadata and other call information. McCourt also teaches extracting representations of features from the call transcript for use in classifying the call. The transcript metadata, topics, labels, and extracted representations correspond to the claimed decision parameters and interaction metadata, see Figs. 1-3 and corresponding text description, also discussion of call-transcript metadata and classification at col. 4. Lines 54-67 in addition to col. 5, lines 1-38, see also step 202 call transcripts received, wherein feature model extracts topics/word distributions from the transcript, lines 11-29 of col. 8);
submitting parameters derived from the interaction script to a machine learning model to at least determine one decision category (reads on supplying the transcript-derived features to classifier A, which determines one or more outcomes or class labels for the call, see Fig. 2 and Fig. 3 and corresponding texts); and
calculating probabilities for the at least one determined decision one or category using the more decision parameters (McCourt ‘s classifier uses the probability-distribution features and produces a probability distribution of class labels F. see Fig. 3, col. 7, lines 25-33)
McCourt does not specifically teach “normalizing and encoding the one or more identified decision parameters into a numerical format, wherein the encoding comprises converting each categorical decision parameter into a binary vector”, “calculating probabilities for the at least one determined decision category for the rule using the one or more normalized and encoded decision parameters, wherein a probability for each of the at least one determined decision category is calculated using a one-vs-rest multi- class classification in which a separate binary classifier is trained for each decision category” and “selected action category comprising one of a retention action, a deletion action, and a copy action, and calculating a time period for performing each selected action category”.
However, Kolar teaches transforming categorical attributes into numerical machine-learning features using one-hot encoding. Specially, Kolar teaches that when an input attribute is categorical, a one-hot encoding of the categorical attribute may be used as the features supplied to a multiclass classifier. One-hot encoding of the categorical converts a categorical value into a vector having binary-valued components. Kolar also teaches normalizing numerical values between zero and one, see [0079-0083].
Thus, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify McCourt’s preprocessing to normalize the identified decision parameters and one-hot-encode each categorical decision parameter as taught by Kolar. McCourt already transforms transcripts information into numerical features readable by its classification model. Kolar ‘s preprocessing would have provided a known and predictable numerical representation of McCourt ‘s categorical transcript metadata, avoided assigning an artificial ordinal relationship to categorical values, and placed the numerical values on a consistent scale for machine-learning processing.
McCourt and Kolar features have been addressed in the above rejection. Note that the combination of McCourt and Kolar still does not specifically teach “calculated using a one-vs-rest multi- class classification in which a separate binary classifier is trained for each decision category”.
Jarah teaches training multiple machine-learning models using features derived from communication information. Jarah teaches that machine-learning models produce probabilities for respective categories. Jarah further teaches a “one versus the rest” classification technique in which each machine-learning model is implemented as a binary classifier that determines whether the input belongs to one specific category or to the other categories (see [0056]- [0059] and [0063]-[0068]).
Thus, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to modify McCourt-Kolar classification model to use Jarah ‘s one-versus-rest technique. McCourt and Jarah both classify numerical features derived form communications into multiple possible categories and calculate category probabilities. Using Jarah ‘s separately trained binary classifiers would have been predictably provided an individual probability for each decision category and facilitated the selection of one or more applicable categories based on their respective probabilities.
The combination of McCourt, Kolar and Jarah, still does not specifically teach “applying the rule by selecting one or more action categories for the interaction transcript based on the calculated probabilities”, “each selected action category comprising one of a retention action, a deletion action, and a copy action”, “calculating a time period for performing each selected action category” and “executing the rule to manage the interaction transcript by performing each selected action category on the interaction transcript in accordance with the calculated time period”.
Yet, Consul teaches retention-policy rules for automatically managing communications and other data items. Consul specifically teaches retention policy tags maybe used to automatically retain, copy or delete data items. Consul further teaches that an action is associated with each retention policy tag and that such actions include retaining item, deleting an item, journaling or forwarding the item, and moving the item to an archive, see col. 2, lines 44-56, col. 6 lines 39-48 and Fig.5 with corresponding text). Consul also teaches associating the selected policy action with expiration information comprising a date or duration. Consul teaches retention periods specifying how long an item us retained, calculating an expiration data based on a start time and policy length, deleting an item on its expiration date and performing journaling or archival actions according to the applicable policy settings (see Fig. 5 and corresponding text).
Thus, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to use the decisions category selected by the McCourt-Kolar-Jarah classifier, based on its calculated probability, to select the corresponding Consul retention-policy rule and associated action. The modification would have predictably automated lifecycle management of the classified interaction transcript. The machine-learning result would identify the applicable decision category, and the policy associated with that category would specify whether the transcript is retained, deleted, or copied and the time at which the selected action is performed.
In McCourt, the claimed “a computing device” as recited in independent claim 14 reads on computing device 120 (as shown in Fig. 1), “a memory” as recited in independent claim 14 reads on memory 506 (as shown in Fig.5) and “a processor” as recited in independent claim 14 reads on reads on processor 504 (as shown in Fig. 5).
Regarding claims 2 and 15, the combination of McCourt, Kolar, Jarah and Consul teaches method according to claim 1, wherein when the selection of the one or more action categories is based on a single-level selection, the rule is based on the decision category with the highest probability (McCourt teaches identifying decision parameters in an interaction script, and determine one decision category from those parameters and calculating probabilities for the categories (see col.1, line 63 through col. 2, line 3 and col. 4, lines 54-67 and Fig. 4A). McCourt classifier outputs a probability distribution for multiple categories. Thus, it would have been obvious to a one of an ordinary skill in the art to select the single category with the highest probability as the operative decision. This being the well-known maximum likelihood principle used in classification models to simplify rule application)).
Regarding claims 3 and 16, the combination of McCourt, Kolar, Jarah and Consul teaches wherein when the selection of the one or more action categories is based on a multi-level selection, the rule is based on the at least one decision category whose calculated probabilities lie above a pre-set threshold value (McCourt teaches computing probabilities for multiple decision categories (see col. 7, lines 25-33. Also, Fig. 3 and corresponding text). Consul teaches using rules that apply to multiple messages tagged for retention or deletion when criteria exceed a threshold (see col. 5, lines 4-17). Thus, it would have been obvious to extend McCourt probabilistic classification by selecting all categories whose probabilities exceed thresholds, similar to Consul rule trigger threshold, in order to handle multiple outcomes per transcript).
Regarding claim 7, the combination of McCourt, Kolar, Jarah and Consul teaches wherein when the selected action category is a retention action, calculating a retention period and storing the interaction transcript for the retention period (Consul teaches a retention policy that when a retention action is applied, calculate retention period and stores data for that duration, see col. 2, lines 44-56) .Note that it would have been obvious to incorporate timed retention mechanism as taught by Consul into the teaching of action logic as taught by McCourt, so that selected retention action causes the transcript to be stored for a computed period.
Regarding claim 8, the combination of McCourt, Kolar, Jarah and Consul teaches wherein when the selected action category is a deletion action, calculating a deletion period and storing the interaction transcript until expiry of the deletion period (McCourt teaches applying rules selecting an action category, see col. 7, lines 25-33. Also, Fig. 3 and corresponding text. Now, Consul teaches a deletion policy specifies an expiration or deletion period after which the data is automatically removed, see col.2 and lines 44-56. Thus, it would have been obvious to extend McCourt framework such that when a deletion action is selected, the transcript is held until expiry of the deletion period, as taught by Consul.
Regarding claim 9, the combination of McCourt, Kolar, Jarah and Consul teaches wherein when the selected action category is a copy action, calculating a copy period and copying the interaction transcript after expiry of the copy period (McCourt teaches applying an action to a transcript, see col. 7, lines 25-33. Also, Fig. 3 and corresponding text. Consul teaches that items subject to a copy or archive policy may be duplicated after a retention interval, see col.6, lines 39-48. Thus, it would have been obvious to configure McCourt system so that when a copy action is selected, the transcript is copied after a defined copy period as taught by Consul).
Regarding claim 11, the combination of McCourt, Kolar, Jarah and Consul teaches wherein applying the rule comprises automatically generating rule recommendations using machine learning for managing the interaction transcript (McCourt teaches applying ML-based rules to interaction transcripts, see col. 1, line 40 through col. 2, line 3. Consul teaches automatic policy recommendation and updates based on historical data usage patterns, see col. 9, lines 40-49. Thus, it would have been obvious to generate rule recommendation using ML within McCourt framework to automatically manage transcripts according to learned patterns of data retention and action application, consistent with Consul adaptive policy learning).
Regarding claim 12, the combination of McCourt, Kolar, Jarah and Consul teaches wherein the generated rule recommendations for managing the interaction transcript are automatically applied to the interaction transcript (reads on automatic routing/action based on classification, see McCourt, see col. 1, lines 40-50).
Regarding claim 13, the combination of McCourt, Kolar, Jarah and Consul teaches wherein the selected one or more selected action categories within the applied rule are periodically updated using machine learning (reads on retraining the classification model using feedback data to improve accuracy, see McCourt, col. 4, lines 28-42).
Independent claim 20 is rejected for the same reasons addressed in independent claims 1 and 14.
Conclusion
4. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Rasha S. AL-Aubaidi whose telephone number is (571) 272-7481. The examiner can normally be reached on Monday-Friday from 8:30 am to 5:30 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Ahmad Matar, can be reached on (571) 272-7488.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/RASHA S AL AUBAIDI/Primary Examiner, Art Unit 2693
Prosecution Insights