Jump to content

Patent Application 17501813 - SYSTEMS AND METHODS FOR APPLICATION OF - Rejection

From WikiPatents

Patent Application 17501813 - SYSTEMS AND METHODS FOR APPLICATION OF

Title: SYSTEMS AND METHODS FOR APPLICATION OF CONTEXT-BASED POLICIES TO VIDEO COMMUNICATION CONTENT

Application Information

  • Invention Title: SYSTEMS AND METHODS FOR APPLICATION OF CONTEXT-BASED POLICIES TO VIDEO COMMUNICATION CONTENT
  • Application Number: 17501813
  • Submission Date: 2025-05-20T00:00:00.000Z
  • Effective Filing Date: 2021-10-14T00:00:00.000Z
  • Filing Date: 2021-10-14T00:00:00.000Z
  • National Class: 709
  • National Sub-Class: 204000
  • Examiner Employee Number: 84701
  • Art Unit: 2459
  • Tech Center: 2400

Rejection Summary

  • 102 Rejections: 0
  • 103 Rejections: 9

Cited Patents

The following patents were cited in the rejection:

Office Action Text


    DETAILED ACTION
	
Introduction
Claims 31-33 and 35-45 are pending. Claims 1-30 and 34 are previously cancelled. Claims 31-33, 41, and 42 are amended. Claim 45 is new. This Office action is in response to Applicant’s request for continued examination (RCE) filed on 9/20/2024. 

Allowable Subject Matter
Claim 45 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.

Response to Arguments
The arguments of Applicant’s representative are discussed below.
Rejection of claim 31 under 35 U.S.C. 103
Applicant’s representative has amended claim 31 to recite several features related to a workflow user interface, and now argues that neither the previously cited combination of Claudatos, Jou, and NLPPA, nor the previously cited combination of Wetjen, Thomsen, and NLPPA, teaches the system of claim 31, as amended. Examiner agrees. Nonetheless, Examiner rejects claim 31 based on two new combinations of prior art in the alternative rejections found below. 

Claim Rejections: 35 U.S.C. 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 31, 32, 37-40, 43, and 44 are rejected under 35 U.S.C. 103 because they are unpatentable over Claudatos (US 2006/0004819) in view of Nash (US 8,572,013) and Natural Language Processing Prior Art (hereinafter, NLPPA).1 
Regarding claim 31, Claudatos teaches a method performed by at least one processor for processing data from video communication applications within an organization, the method comprising: extracting at least one of information or metadata from a video communication originating from the video communication applications (The system extracts information from a video file. See par. 34-35. The video file may be generated using any known manner of generating a video file, including generating the video file using a video communication application, such as a video communication application), information or metadata comprising at least one of extracted video, voice, chat, instant messaging, digital content, or file exchange content contained in the video communication (The extracted information comprises video, audio, text, and/or files. See par. 34-35); evaluating, by the at least one processor, a risk score for the information or metadata by: classifying the information or metadata based on (i) applying a plurality of rules to the information or metadata (The system uses a set of analytic policies for evaluating the information. See par. 105. Moreover, the information is evaluated using various classification rules, such as classification rules identifying keywords and/or orderings of keywords. See par. 30-31, 36-45, 84), and (ii) at least one of: a regulatory risk, a privacy risk, a risk of loss of trade secret protection, a corporate policy compliance risk, an operational compliance policy, a non-public information risk, an acceptable use risk, or a risk to electronic communication security (The policies and rules relate to various risks such as a regulatory risk of compliance with HIPAA regulations. See par. par. 101); and calculating and assigning, based on the classification of the information or metadata, the risk score to the information or metadata (A classification level is automatically assigned to the video file based on the result of the evaluating. See par. 92); and applying, based on (i) the assigned risk score or the adjusted risk score and (ii) the classification of the information or metadata, workflow policies to the classified information or metadata (A user may override the automatically assigned classification level by manually assigning a new classification level to the video using a “right-click” menu of a user interface. See par. 91. The system applies one or more information lifecycle management (ILM) policies to the video file based on the automatically assigned classification level or the new classification level. See par. 92).
However, Claudatos does not teach causing display, based on the risk score assigned to the information or metadata, of (i) a workflow user interface comprising a plurality of workflow control options for the video communication and (ii) a video player for selection and review of the video communication; and receiving, by the workflow user interface, an input comprising a selection of (i) the assigned risk score or (ii) an adjusted risk score. Nonetheless, Nash teaches causing display, based on an existing classification that is automatically assigned to a video and a confidence score associated with the existing classification, a user interface of a classification review application that (i) displays a plurality of workflow control options comprising at least a first option for a reviewer to accept the existing classification as correct, and a second option for the reviewer to reject the existing classification as incorrect (thereby triggering a reclassification of the video to generate a new classification), and (ii) a web page for presenting the video to the reviewer. See col. 8, ln. 7-16. Nash further teaches receiving, via the user interface of the classification review application, a selection of the existing classification by accepting the existing classification as correct, or a selection of a new classification by rejecting the existing classification as incorrect. Id. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos so that the system displays, based on the existing classification level automatically assigned to the video file, a user interface of a classification review application that displays (i) a plurality of workflow control options for reviewing the video file and (ii) a video player for viewing the video file, and so that the system receives, via the user interface, a selection of the existing classification level or the new classification level, because doing so allows for manual review of the existing classification level when a confidence score associated with the existing classification level is below a threshold. 
Lastly, Claudatos and Nash do not teach wherein at least one rule of the plurality of rules is derived from first machine learning-based techniques. However, using machine learning to derive rules for processing (i.e., classifying) natural language content was ubiquitous in the art before the effective filing date of the claimed invention. See Bangalore, col. 1, ln. 18-29; Huang, par. 24; Kakirwar, par. 2.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos and Nash so that the analytic policies are derived from machine learning techniques because doing so allows the system to take advantage of the well-known benefits of using machine learning to perform natural language processing. 
Regarding claim 32, Claudatos, Nash, and NLPPA teach the method of claim 31, wherein the information or metadata extracted from the video communication further comprises content selected from the group consisting of audio, text, graphics, still images of objects, still images of people, file-based content, or document-based content exchanged during the video communication (Claudatos teaches that the extracted information may include video, audio, text, and graphics. See par. 34-35).
Regarding claim 37, Claudatos, Nash, and NLPPA teach the method of claim 31, further comprising compressing, based on the risk score and compression workflow policies, the information or metadata (Claudatos teaches that the ILM policy corresponding to the classification level assigned to the video file may specify a compression policy. See claim 4).
Regarding claim 38, Claudatos, Nash, and NLPPA teach the method of claim 38, wherein compressing the information or metadata comprises: selecting a level of encryption (Claudatos teaches that the ILM policy corresponding to the classification level assigned to the video file may specify both compression parameters and level of encryption. See par. 26; claim 4).
Regarding claim 39, Claudatos, Nash, and NLPPA teach the method of claim 31, further comprising storing, based on the risk score and storage workflow policies, the information or metadata (Claudatos teaches that the system stores the video file based on a retention policy that corresponds to the classification level assigned to the video file. See par. 26). 
Regarding claim 40, Claudatos, Nash, and NLPPA teach the method of claim wherein storing the information or metadata comprises: determining a storage location, a storage format, and a storage duration (Claudatos teaches that the ILM polices that correspond to the classification level assigned to the video file may dictate the physical location where the video file is stored, compression and encryption parameters (i.e., storage formats), and the length of time that the video file is to be stored before deletion. See par. 26, 46). 
Regarding claims 43 and 44, Claudatos, Nash, and NLPPA teach the method of claim 31, wherein the at least one rule of the plurality of rules is derived by applying the first machine learning-based techniques to external research data (user-provided data) (NLPPA teaches using one or more sets of training data, which may be characterized as both user-provided data and external research data. See Bangalore, col. 2, ln. 55-65, col. 5, ln. 7-25; Huang, par. 28, 39, 44; Kakirwar, par. 2, 21, 25. Thus, NLPPA suggests further modifying the system of Claudatos, Nash, and NLPPA so that one or more of the analytic policies are derived by applying machine learning techniques to a set of user-provided and or external research training data, because doing so is useful for the reasons provided above with respect to claim 31).
Claim 33 is rejected under 35 U.S.C. 103 because it is unpatentable over Claudatos, Nash, and NLPPA, as applied to claim 31 above, in further view of either Watson (US 2014/0241519) or Iannone (US 2015/0066504).
Regarding claim 33, Claudatos, Nash, and NLPPA teach the method of claim 31, wherein the information or metadata comprises data extracted from the audio portion of the video communication (Claudatos teaches that the system may extract audio data from the video file. See par. 34-35), the method further comprising: preparing a transcript of the data extracted from the audio portion of the video communication (Claudatos further teaches that the system can generate a transcript of any speech occurring in the video file. See par. 72). However, Claudatos, Nash, and NLPPA do not teach comparing the transcript to a library comprising a plurality of pre-approved scripts; applying second machine learning-based techniques to at least one of (i) analyze and correct errors in the transcript; and based at least in part on the comparison, at least one of: identifying non-matching data; or calculating and assigning a risk score to the information or metadata. Nonetheless, Watson and Iannone both teach: comparing the transcript to a library comprising a plurality of pre-approved scripts (Watson teaches comparing a transcript of speech by a customer service agent to a database of pre-approved script text. See par. 25. Iannone teaches comparing a transcript of speech to a model transcript of a desired script. See par. 28); applying second machine learning-based techniques to at least one of (i) analyze and correct errors in the transcript (Watson teaches using machine learning to analyze differences between the transcript and the pre-approved script text. See par. 26, 44. Iannone also teaches using machine learning to analyze differences between the transcript and the model transcript. See par. 28, 32); and based at least in part on the comparison, at least one of: identifying non-matching data; or assigning a risk score to the information or metadata (Watson teaches computing a confidence score indicating a degree of confidence that the customer service agent adhered to the pre-approved script text based on the comparison. See par. 44. Iannone teaches determining a script accuracy indicative of a degree to which the transcript matches the model transcript based on the comparison. See par. 32). 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos, Nash, and NLPPA to incorporate the above features of either Watson or Iannone because doing so allows the system to evaluate a risk of speech that does not comply with a pre-approved script.  
Claims 35 and 36 are rejected under 35 U.S.C. 103 because they are unpatentable over Claudatos, Nash, and NLPPA, as applied to claim 31 above, in further view of Stolikj (US 2019/0073562).
Regarding claim 35, Claudatos, Nash, and NLPPA do not teach the method of claim 31, wherein the information or metadata comprises a distinctive image, the method further comprising: comparing the distinctive image to a library of known object images; applying second machine learning-based techniques to analyze the distinctive image; and based at least in part on the comparison, calculating and assigning the risk score to the information or metadata. However, Stolikj teaches a system for detecting unauthorized usage of video or image content whereby the system compares a logo extracted from video or image content to a library of predetermined logos using a convolutional neural network, and determines whether a usage of the video or image content is unauthorized based on the comparison. See par. 25, 104. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos, Nash, and NLPPA so that the system compares extracted from the content of the call to a library of predetermined logos using machine learning, and calculates the risk score based on the comparison, because doing so allows the system to determine compliance violations associated with unauthorized usage of trademarked content. 
Regarding claim 36, Claudatos, Nash, NLPPA, and Stolikj teach the method of claim 35, further comprising, based at least in part on the comparison, labeling the distinctive image (Stolikj teaches labeling a log with an identifier of the detected logo (See par. 105), which suggests modifying the system of Claudatos, Nash, and NLPPA so that the transcript of the call is labelled with the identifier of a detected logo, because doing so facilitates manual review of any compliance violation that may be associated with the detected logo).
Claims 41 and 42 are rejected under 35 U.S.C. 103 because it is unpatentable over Claudatos, Nash, and NLPPA, as applied to claim 31 above, in further view of Weber (US 9,224,386).
Regarding claims 41 and 42, Claudatos, Nash, and NLPPA do not teach the method of claim 31,  further comprising redacting and remediating the non-matching data. However, Weber teaches a speech recognition system that detects non-matching data based on comparison of a hypothetical transcript with a model transcript and deletes the non-matching data or otherwise remediates the non-matching data. See col. 2, ln. 50-61. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos, Nash, and NLPPA so that the system redacts and remediates non-matching data in the transcript because doing so allows the system to fix transcription errors. 
Alternatively, claims 31, 32, 43, and 44 are rejected under 35 U.S.C. 103 because they are unpatentable over Wetjen (US 2015/0106091) in view of Nash and NLPPA. 
Regarding claim 31, Wetjen teaches a method performed by at least one processor for processing data from video communication applications within an organization, the method comprising: extracting at least one of information or metadata from a video communication originating from the video communication applications (The system extracts information from a recorded call. See par. 111), the information or metadata comprising at least one of extracted video, voice, chat, instant messaging, digital content, or file exchange content (The extracted information may comprise video, audio, and/or text information. See par. 27); evaluating, by the at least one processor, a risk score for the information or metadata by: classifying the information or metadata based on (i) applying a plurality of rules to the information or metadata (The system classifies the audio information using words and/or phrases contained in a set of policies and/or legal rules. See par. 111) and (ii) at least one of: a regulatory risk, a privacy risk, a risk of loss of trade secret protection, a corporate policy compliance risk, an operational compliance policy, a non-public information risk, an acceptable use risk, or a risk to electronic communication security (The policies and legal rules relate to risks associated with violating corporate policies and legal regulations. See par. 50, 154); and calculating and assigning, based on the classification of the information or metadata, the risk score to the information or metadata (The system computes a percentage probability, confidence score, or other rating associated with the classification (hereinafter referred to as a “risk score.” See par. 112); and applying, based on (i) the assigned risk score and (ii) the classification of the information or metadata, workflow policies to the classified information or metadata (The system applies one or more policies to the content of the call based on the classification and the risk score, such as logging a compliance violation in association with the content of the call, correlating a compliance violation with a transcript of the call, and/or notifying a supervisor, legal group, or compliance group of a policy violation. See par. 111-112).
However, Wetjen does not teach causing display, based on the risk score assigned to the information or metadata, of (i) a workflow user interface comprising a plurality of workflow control options for the video communication and (ii) a video player for selection and review of the video communication; receiving, by the workflow user interface, an input comprising a selection of (i) the assigned risk score or (ii) an adjusted risk score. Lastly, Wetjen does not teach that the applying step may be alternatively based on the adjusted risk score. Nonetheless, Nonetheless, Nash teaches causing display, based on an existing classification that is automatically assigned to a video and a confidence score associated with the existing classification, a user interface of a classification review application that (i) displays a plurality of workflow control options comprising at least a first option for a reviewer to accept the existing classification as correct, and a second option for the reviewer to reject the existing classification as incorrect (thereby triggering a reclassification of the video to generate a new classification), and (ii) a web page for presenting the video to the reviewer. See col. 8, ln. 7-16. Nash further teaches receiving, via the user interface of the classification review application, a selection of the existing classification by accepting the existing classification as correct, or a selection of a new classification by rejecting the existing classification as incorrect. Id. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Claudatos so that the system displays, based on the existing classification level automatically assigned to the video file, a user interface of a classification review application that displays (i) a plurality of workflow control options for reviewing the video file and (ii) a video player for viewing the video file, and so that the system receives, via the user interface, a selection of the existing classification level or the new classification level for use in selecting the one or more policies that are to be applied to the video file, because doing so allows for manual review of the existing classification level when a confidence score associated with the existing classification level is below a threshold. 
Lastly, Wetjen and Nash do not teach wherein at least one rule of the plurality of rules is derived from first machine learning-based techniques. However, using machine learning to derive rules for processing (i.e., classifying) natural language content was ubiquitous in the art before the effective filing date of the claimed invention. See Bangalore, col. 1, ln. 18-29; Huang, par. 24; Kakirwar, par. 2.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Wetjen and Nash so that the analytic policies are derived from machine learning techniques because doing so allows the system to take advantage of the well-known benefits of using machine learning to perform natural language processing. 
Regarding claim 32, Wetjen, Nash, and NLPPA teach the method of claim 31, wherein the information or metadata extracted from the video communication further comprises content selected from the group consisting of audio, text, graphics, still images of objects, still images of people, file-based content, or document-based content exchanged during the video communication (Wetjen teaches that the extracted information includes audio information. See par. 111).
Regarding claims 43 and 44, Wetjen, Nash, and NLPPA teach the method of claim 31, wherein the at least one rule of the plurality of rules is derived by applying the first machine learning-based techniques to external research data (user-provided data) (NLPPA teaches using one or more sets of training data, which may be characterized as both user-provided data and external research data. See Bangalore, col. 2, ln. 55-65, col. 5, ln. 7-25; Huang, par. 28, 39, 44; Kakirwar, par. 2, 21, 25. Thus, NLPPA suggests further modifying the system of Wetjen and Nash so that one or more of the analytic policies are derived by applying machine learning techniques to a set of user-provided and or external research training data, because doing so is useful for the reasons provided above with respect to claim 31).
Claim 33 is rejected under 35 U.S.C. 103 because it is unpatentable over Wetjen, Nash, and NLPPA, as applied to claim 31 above, in further view of either Watson or Iannone.
Regarding claim 33, Wetjen, Nash, and NLPPA teach wherein the information or metadata comprises data extracted from the audio portion of the video communication application (The extracted information includes audio information. See par. 111), the method further comprising: preparing a transcript of the data extracted from the audio portion of the video communication application (Wetjen teaches that the system generates a transcript of the audio information. See par. 111). However, Wetjen, Nash, and NLPPA do not teach comparing the transcript to a library comprising a plurality of pre-approved scripts; applying second machine learning-based techniques to at least one of (i) analyze and correct errors in the transcript; and based at least in part on the comparison, at least one of: identifying non-matching data; or calculating and assigning the risk score to the information or metadata. Nonetheless, Watson and Iannone both teach: comparing the transcript to a library comprising a plurality of pre-approved scripts (Watson teaches comparing a transcript of speech by a customer service agent to a database of pre-approved script text. See par. 25. Iannone teaches comparing a transcript of speech to a model transcript of a desired script. See par. 28); applying second machine learning-based techniques to at least one of (i) analyze and correct errors in the transcript (Watson teaches using machine learning to analyze differences between the transcript and the pre-approved script text. See par. 26, 44. Iannone also teaches using machine learning to analyze differences between the transcript and the model transcript. See par. 28, 32); and based at least in part on the comparison, at least one of: identifying non-matching data; or assigning a risk score to the information or metadata (Watson teaches computing a confidence score indicating a degree of confidence that the customer service agent adhered to the pre-approved script text based on the comparison. See par. 44. Iannone teaches determining a script accuracy indicative of a degree to which the transcript matches the model transcript based on the comparison. See par. 32). 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Wetjen, Nash, and NLPPA to incorporate the above features of either Watson or Iannone because doing so allows the system to evaluate a risk of speech that does not comply with a pre-approved script.  
Claims 35 and 36 are rejected under 35 U.S.C. 103 because they are unpatentable over Wetjen, Nash, and NLPPA, as applied to claim 31 above, in further view of Stolikj.
Regarding claim 35, Wetjen, Nash, and NLPPA do not teach the method of claim 31, wherein the information or metadata comprises a distinctive image, the method further comprising: comparing the distinctive image to a library of known object images; applying second machine learning-based techniques to analyze the distinctive image; and based at least in part on the comparison, calculating and assigning the risk score to the information or metadata. However, Stolikj teaches a system for detecting unauthorized usage of video or image content whereby the system compares a logo extracted from video or image content to a library of predetermined logos using a convolutional neural network, and determines whether a usage of the video or image content is unauthorized based on the comparison. See par. 25, 104. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Wetjen, Nash, and NLPPA so that the system compares extracted from the content of the call to a library of predetermined logos using machine learning, and calculates the risk score based on the comparison, because doing so allows the system to determine compliance violations associated with unauthorized usage of trademarked content. 
Regarding claim 36, Wetjen, Nash, NLPPA, and Stolikj teach further comprising, based at least in part on the comparison, labeling the distinctive image (Stolikj teaches labeling a log with an identifier of the detected logo (See par. 105), which suggests modifying the system of Wetjen, Nash, and NLPPA so that the transcript of the call is labelled with the identifier of a detected logo, because doing so facilitates manual review of any compliance violation that may be associated with the detected logo).
Claims 37-40 are rejected under 35 U.S.C. 103 because they are unpatentable over Wetjen, Nash, and NLPPA, as applied to claim 31 above, in further view of Claudatos. 
Regarding claim 37, Wetjen, Nash, and NLPPA do not teach the method of claim 31, further comprising compressing, based on the risk score and compression workflow policies, the information or metadata. However, Claudatos teaches a system for selecting one or more information lifecycle management (ILM) policies to apply to content of a call based on a classification level of the call, whereby an ILM policy may specify a compression policy. See claim 4.  
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Wetjen, Nash, and NLPPA so that the system compresses content of the call based on the risk score, because doing so facilitates archiving of the content of the call when a compliance violation is detected. 
Regarding claim 38, Wetjen, Nash, NLPPA, and Claudatos teach the method of claim 37, wherein compressing the information or metadata comprises: selecting a level of encryption (Claudatos teaches that the ILM policy corresponding to the classification level assigned to the video file may specify both compression parameters and level of encryption. See par. 26; claim 4. Thus, Claudatos suggests modifying the system of Wetjen, Nash, and NLPPA so that the system archives content of a call based on a selected level of encryption because doing so is beneficial for the reasons provided above with respect to claim 37).
Regarding claim 39, Wetjen, Nash, NLPPA, and Claudatos teach the method of claim 31, further comprising storing, based on the risk score and storage workflow policies, the information or metadata (Claudatos teaches storing a video file based on a retention policy that corresponds to the classification level assigned to the video file (See par. 26), which suggests modifying the system of Wetjen, Nash, and NLPPA so that the system stores content of the call based on a retention policy that corresponds to the risk score, because doing so is beneficial for the reasons provided above with respect to claim 37). 
Regarding claim 40, Wetjen, Nash, NLPPA, and Claudatos teach the method of claim 39, wherein storing the information or metadata comprises: determining a storage location, a storage format, and a storage duration (Claudatos teaches that the ILM polices that correspond to the classification level assigned to the video file may dictate the physical location where the video file is stored, compression and encryption parameters (i.e., storage formats), and the length of time that the video file is to be stored before deletion. See par. 26, 46. Thus, Claudatos suggests modifying the system of Wetjen, Nash, and NLPPA so that archiving content of the call comprises determining a storage location, storage format, and storage duration, because doing so is beneficial for the reasons provided above with respect to claim 37). 
Claims 41 and 42 are rejected under 35 U.S.C. 103 because it is unpatentable over Wetjen, Nash, and NLPPA, as applied to claim 31 above, in further view of Weber.
Regarding claims 41 and 42, Wetjen, Nash, and NLPPA do not teach the method of claim 31, further comprising redacting and remediating the non-matching data. However, Weber teaches a speech recognition system that detects non-matching data based on comparison of a hypothetical transcript with a model transcript and deletes the non-matching data or otherwise remediates the non-matching data. See col. 2, ln. 50-61. 
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Wetjen, Nash, and NLPPA so that the system redacts and remediates non-matching data in the transcript because doing so allows the system to fix transcription errors. 

Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew Georgandellis whose telephone number is 571-270-3991.  The examiner can normally be reached on Monday through Friday, 7:30-5:00 PM EST. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tonia Dollinger, can be reached on 571-272-4170.  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 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). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.


/ANDREW C GEORGANDELLIS/Primary Examiner, Art Unit 2459                                                                                                                                                                                                        






    
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
        
            
    

    
        1 Examiner collectively refers to Bangalore (US 7,451,125), Huang (US 2019/0057145), and Kakirwar (US 2019/0042561) as Natural Language Processing Prior Art (NLPPA). 
    


Cookies help us deliver our services. By using our services, you agree to our use of cookies.