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 .
DETAILED ACTION
Claims 1-20 are pending in this application.
Priority
Acknowledgment is made of applicant's claim for foreign priority based on an application filed in INDIA on 12/22/2023. It is noted, however, that applicant has not filed a certified copy of the IN202411087999 application as required by 37 CFR 1.55.
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 (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 and 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over Madavarapu et al (“Madavarapu,” US 20200302017) and further in view of Ramanathan et al (“Ramanathan,” US 20230164095).
Regarding claim 1, Madavarapu discloses a system comprising:
a processor; and (Madavarapu, FIG 3, [0046]-[0057] describes a processor)
a non-transitory computer readable medium stored thereon instructions that are executable by the processor to cause the system to perform operations comprising: (Madavarapu, FIG 3, [0046]-[0057] describe a non-transitory computer readable medium stored thereon instructions that are executable by the processor to cause the system to perform operations comprising)
receive a conversation log between a first user and a second user; (Madavarapu discloses FIG 1A, [0014]-[0017] describes receive a conversation log between a first user and a second user)
process, via a second machine learning model, the at least one text chunk, (Madavarapu in FIG 1B, FIG 4; [0021], [0034], [0058]-[0061] discloses process, via a second machine learning model, the at least one text chunk)
the second machine learning model trained using previous conversation logs to determine whether the at least one text chunk indicates a vulnerability; (Madavarapu in FIG 1B, FIG 4; [0021], [0034], [0058]-[0061] discloses the second machine learning model trained using previous conversation logs to determine whether the at least one text chunk indicates a vulnerability)
in response to the at least one text chunk indicating the vulnerability, classify a type of the indicated vulnerability; and (Madavarapu, FIG 1B, [0021], [0028]-[0030] describes in response to the at least one text chunk indicating the vulnerability, classify a type of the indicated vulnerability)
automatically execute a remedial action based on the classified type (Madavarapu, FIG 1C, FIG 5, [0011], [0030], [0043], also see [0070]-[0075] describes automatically execute a remedial action based on the classified type)
Madavarapu fails to explicitly disclose derive, via a first machine learning model, at least one text chunk from the conversation log;
However, in an analogous art, Ramanathan discloses derive, via a first machine learning model, at least one text chunk from the conversation log (Ramanathan, Figures 1C-1F; [0041]-[0045], also see [0050]-[0059] describe derive, via a first machine learning model, at least one text chunk from the conversation log)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Ramanathan with the method and system of Madavarapu to include derive, via a first machine learning model, at least one text chunk from the conversation log. One would have been motivated to analyze chatbot communication sessions to reduce escalation (Ramanathan, [0013]-[0014]).
Regarding claim 6, Madavarapu and Ramanthan disclose the system of claim 1.
Madavarapu further discloses wherein training the second machine learning model comprises: retrieving a plurality of previous conversation logs; (Madavarapu, [0010]-[0011], [0021], [0034] describes wherein training the second machine learning model comprises: retrieving a plurality of previous conversation logs)
labelling each of the plurality of previous conversation logs based on whether the conversation log indicates a vulnerability; and (Madavarapu discloses [0021], [0030], [0034] describes labelling each of the plurality of previous conversation logs based on whether the conversation log indicates a vulnerability; and)
generating a set of training data comprising the plurality of previous conversation logs and the respective labels, (Madavarapu, FIG 4; [0034], [0058]-[0061] describes generating a set of training data comprising the plurality of previous conversation logs and the respective labels)
Regarding claim 7, Madavarapu and Ramanthan disclose the system of claim 1.
Madavarapu further discloses wherein the instructions are further configured to cause the system to perform operations comprising: prior to deriving the at least one text chunk, pre-processing the conversation log to remove noise (Madavarapu, [0020] describes wherein the instructions are further configured to cause the system to perform operations comprising: prior to deriving the at least one text chunk, pre-processing the conversation log to remove noise)
Claims 2-3 are rejected under 35 U.S.C. 103 as being unpatentable over Madavarapu et al (“Madavarapu,” US 20200302017) in view of Ramanathan et al (“Ramanathan,” US 20230164095) and further in view of Sorensen et al (“Sorensen,” US 20200004765).
Regarding claim 2, Madavarapu and Ramanathan disclose the system of claim 1.
Ramanathan further discloses wherein deriving the at least one text chunk from the conversation log comprises: (Ramanathan, Figures 1C-1F; [0041]-[0045], also see [0050]-[0059] describe derive, via a first machine learning model, at least one text chunk from the conversation log)
divide the conversation log into initial chunks; (Ramanathan, Figures 1C-1F, FIG 5; [0031]-[0045], also see [0050]-[0055] describe divide the conversation log into initial chunks)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Ramanathan with the method and system of Madavarapu to include wherein deriving the at least one text chunk from the conversation log comprises: divide the conversation log into initial chunks. One would have been motivated to analyze chatbot communication sessions to reduce escalation (Ramanathan, [0013]-[0014]).
Madavarapu and Ramanathan fail to explicitly disclose process each of the initial chunks to identify at least one part of speech; and determine one or more of the initial chunks as text chunks based on the identified parts of speech.
However, in an analogous art, Sorenson discloses process each of the initial chunks to identify at least one part of speech; (Sorenson, FIG 2, [0028]-[0032] describes process each of the initial chunks to identify at least one part of speech)
and determine one or more of the initial chunks as text chunks based on the identified parts of speech, (Sorensen disclose [0029]-[0032], [0045] describe and determine one or more of the initial chunks as text chunks based on the identified parts of speech)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Sorensen with the method and system of Madavarapu and Ramanathan to include process each of the initial chunks to identify at least one part of speech; and determine one or more of the initial chunks as text chunks based on the identified parts of speech. One would have been motivated to provide methods, systems, and computer media that allow for converting unstructured text to structured text, (Sorensen, [0003]).
Regarding claim 3, Madavarapu and Ramanthan disclose the system of claim 1.
Madavarapu further discloses in FIG 1B, [0028]-[0030] category confidence levels for chat logs; and [0030] describes threshold/confidence category assignment.
Madavarapu and Ramanthan fail to explicitly disclose wherein determining whether the at least one text chunk indicates the vulnerability comprises: generate, by the second machine learning model, a vulnerability score for each of the at least one text chunk; and label each at least one text chunk as indicating the vulnerability in response to a respective vulnerability score being greater than a threshold value.
However, in an analogous art, Sorensen discloses wherein determining whether the at least one text chunk indicates the vulnerability comprises: generate, by the second machine learning model, a vulnerability score for each of the at least one text chunk; (Sorensen, [0012], [0045] describe wherein determining whether the at least one text chunk indicates the vulnerability comprises: generate, by the second machine learning model, a vulnerability score for each of the at least one text chunk)
and label each at least one text chunk as indicating the vulnerability in response to a respective vulnerability score being greater than a threshold value, (Sorensen, [0046] describes and label each at least one text chunk as indicating the vulnerability in response to a respective vulnerability score being greater than a threshold value)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Sorensen with the method and system of Madavarapu and Ramanathan to include wherein determining whether the at least one text chunk indicates the vulnerability comprises: generate, by the second machine learning model, a vulnerability score for each of the at least one text chunk; and label each at least one text chunk as indicating the vulnerability in response to a respective vulnerability score being greater than a threshold value. One would have been motivated to provide methods, systems, and computer media that allow for converting unstructured text to structured text (Sorensen, [0003]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Madavarapu et al (“Madavarapu,” US 20200302017), Ramanathan et al (“Ramanathan,” US 20230164095) in view of Sorensen et al (“Sorensen,” US 20200004765) and further in view of Weston et al (“Weston,” US 20190332617).
Regarding claim 4, Madavarapu, Ramanthan and Sorensen disclose the system of claim 3.
Madavarapu, Ramanthan and Sorensen fail to explicitly disclose wherein generating the vulnerability score comprises: convert each of the at least one text chunk into a corresponding embeddings vector; and compute the vulnerability score for each text chunk based on the corresponding vector.
However, in an analogous art, Weston discloses wherein generating the vulnerability score comprises: convert each of the at least one text chunk into a corresponding embeddings vector; (Weston, Figures 5 & 7; [0041], [0051]-[0058] describes wherein generating the vulnerability score comprises: convert each of the at least one text chunk into a corresponding embeddings vector)
and compute the vulnerability score for each text chunk based on the corresponding vector, (Weston discloses Figures 4 & 7, [0041]-[0048]; [0057]-[0059] describes and compute the vulnerability score for each text chunk based on the corresponding vector)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Madavarapu, Ramanathan and Sorensen to include wherein generating the vulnerability score comprises: convert each of the at least one text chunk into a corresponding embeddings vector; and compute the vulnerability score for each text chunk based on the corresponding vector. One would have been motivated to provide a deep-learning model to predict relevant labels for a text query (Weston, [0006]).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Madavarapu et al (“Madavarapu,” US 20200302017) in view of Ramanathan et al (“Ramanathan,” US 20230164095) and further in view of Weston et al (“Weston,” US 20190332617).
Regarding claim 5, Madavarapu and Ramanthan disclose the system of claim 1.
Madavarapu and Ramanthan fail to explicitly disclose wherein the classifying the type of the indicated vulnerability comprises: retrieve a set of reference text chunks that correspond to a set of vulnerability types, wherein each of the set of reference text chunks represent one of the set of vulnerability types; compare the at least one text chunk indicative of the vulnerability to the set of reference text chunks; and
determine the type of the indicated vulnerability based on the reference text chunk most similar to the at least one text chunk.
However, in an analogous art, Weston discloses wherein the classifying the type of the indicated vulnerability comprises: retrieve a set of reference text chunks that correspond to a set of vulnerability types, wherein each of the set of reference text chunks represent one of the set of vulnerability types; (Weston, Figures 3B, 4; [0039]-[0042] describes wherein the classifying the type of the indicated vulnerability comprises: retrieve a set of reference text chunks that correspond to a set of vulnerability types, wherein each of the set of reference text chunks represent one of the set of vulnerability types)
compare the at least one text chunk indicative of the vulnerability to the set of reference text chunks; and (Weston, Figure 4; [0041]-[0048] describes compare the at least one text chunk indicative of the vulnerability to the set of reference text chunks)
determine the type of the indicated vulnerability based on the reference text chunk most similar to the at least one text chunk (Weston, Figure 7; [0057]-[0059] describes determine the type of the indicated vulnerability based on the reference text chunk most similar to the at least one text chunk)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Madavarapu and Ramanathan to include wherein generating the vulnerability score comprises: convert each of the at least one text chunk into a corresponding embeddings vector; and compute the vulnerability score for each text chunk based on the corresponding vector. One would have been motivated to provide a deep-learning model to predict relevant labels for a text query (Weston, [0006]).
Claims 8 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Ramanathan et al (“Ramanathan,” US 20230164095) in view of Madavarapu et al (“Madavarapu,” US 20200302017) and further in view of Jeyakumar et al (“Jeyakumar,” US 20200344251).
Regarding claim 8, Ramanathan discloses a method comprising:
dividing a chat log into a plurality of text portions; (Ramanathan, Figures 1C-1F; Figure 5; [0031]-[0045]; also see [0050]-[0055] describe dividing a chat log into a plurality of text portions)
processing, by a first machine learning model, the plurality of text portions to identify a first text portion of the plurality of text portions that indicates a vulnerability action; (Ramanathan, FIG 2, [0057]-[0059] describes processing, by a first machine learning model, the plurality of text portions to identify a first text portion of the plurality of text portions that indicates a vulnerability action)
determine a type of the vulnerability action indicated by the first text portion or of the vulnerability state indicated by the second text portion; and (Ramanathan, [0028]-[0030], [0039]-[0043] describes determine a type of the vulnerability action indicated by the first text portion or of the vulnerability state indicated by the second text portion; and)
autonomously execute a remedial action based on the determined type, (Ramanathan in FIG 5; [0048]-[0050], [0056] describes autonomously execute a remedial action based on the determined type)
Ramanathan fails to explicitly disclose processing, by a second machine learning model, the plurality of text portions to identify a second text portion of the plurality of text portions that indicate a vulnerability state;
However, in an analogous art, Madavarapu discloses processing, by a second machine learning model, the plurality of text portions to identify a second text portion of the plurality of text portions that indicate a vulnerability state; (Madavarapu, [0028]-[0030], [0039]-[0043] describes processing, by a second machine learning model, the plurality of text portions to identify a second text portion of the plurality of text portions that indicate a vulnerability state)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Madavarapu with the method and system of Ramanathan to include derive, via a first machine learning model, at least one text chunk from the conversation log. One would have been motivated to provide chat analysis using machine learning (Madavarapu, [0061]).
Ramanathan and Madavarapu fail to explicitly disclose autonomously execute a remedial action based on the determined type.
However, in an analogous art, Jeyakumar discloses autonomously execute a remedial action based on the determined type, (Jeyakumar, [0107]-[0108] describes autonomously execute a remedial action based on the determined type)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Jeyakumar with the method and system of Ramanathan and Madavarapu to include autonomously execute a remedial action based on the determined type. One would have been motivated to provide techniques for detecting email based threats in the security field (Jeyakumar, [0002]).
Regarding claim 13, Ramanathan, Madavarapu and Jeyakumar disclose the method of claim 8.
Madavarapu further discloses further comprising: generating, by the first machine learning model, a first confidence score for the first text portion; (Madavarapu, [0030] describes further comprising: generating, by the first machine learning model, a first confidence score for the first text portion)
and generating, by the second machine learning model, a second confidence score for the second text portion, (Madavarapu, [0030] discloses and generating, by the second machine learning model, a second confidence score for the second text portion)
wherein the autonomous execution of the remedial action is in response to the first confidence score and the second confidence score exceeding a threshold value, (Madavarapu, [0030], [0043], [0070]-[0075] discloses wherein the autonomous execution of the remedial action is in response to the first confidence score and the second confidence score exceeding a threshold value)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Madavarapu with the method and system of Ramanathan to include further comprising: generating, by the first machine learning model, a first confidence score for the first text portion; and generating, by the second machine learning model, a second confidence score for the second text portion, wherein the autonomous execution of the remedial action is in response to the first confidence score and the second confidence score exceeding a threshold value. One would have been motivated to provide chat analysis using machine learning (Madavarapu, [0061]).
Claims 9-12 are rejected under 35 U.S.C. 103 as being unpatentable over Ramanathan et al (“Ramanathan,” US 20230164095), Madavarapu et al (“Madavarapu,” US 20200302017), in view of Jeyakumar et al (“Jeyakumar,” US 20200344251) and further in view of Weston et al (“Weston,” US 20190332617).
Regarding claim 9, Ramanathan, Madavarapu and Jeyakumar disclose the method of claim 8.
Ramanathan, Madavarapu and Jeyakumar fail to explicitly disclose wherein identifying the first text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors; determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability action; and determining the first text portion as the plurality of text portions having a respective vulnerability score above a threshold.
However, in an analogous art, Weston discloses wherein identifying the first text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors; (Weston, FIG 5, FIG 7; [0051]-[0058] describe wherein identifying the first text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors)
determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability action; (Weston FIG 4, FIG 7, [0041]-[0048], [0057]-[0059] describe determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability action)
and determining the first text portion as the plurality of text portions having a respective vulnerability score above a threshold, (Weston, Figure 7; [0057]-[0059] describes and determining the first text portion as the plurality of text portions having a respective vulnerability score above a threshold)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Ramanathan, Madavarapu and Jeyakumar to include wherein identifying the first text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors; determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability action; and determining the first text portion as the plurality of text portions having a respective vulnerability score above a threshold. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Regarding claim 10, Ramanathan, Madavarapu, Jeyakumar and Weston discloses the method of claim 9.
Ramanathan further discloses wherein: the type of the vulnerability action comprises a plurality of action types, (Ramanathan, [0048]-[0050] describe wherein: the type of the vulnerability action comprises a plurality of action types)
Weston further discloses the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of action types, (Weston, Figure 3B; [0039]-[0042] describes the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of action types)
and determining the type of the vulnerability action indicated by the first text portion further comprises: establishing the type of the vulnerability action as the action type represented by the reference vector closest to the embeddings vector corresponding to the first text portion (Weston, FIG 7; [0057]-[0059] describes and determining the type of the vulnerability action indicated by the first text portion further comprises: establishing the type of the vulnerability action as the action type represented by the reference vector closest to the embeddings vector corresponding to the first text portion) .
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Ramanathan, Madavarapu and Jeyakumar to include the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of action types, and determining the type of the vulnerability action indicated by the first text portion further comprises: establishing the type of the vulnerability action as the action type represented by the reference vector closest to the embeddings vector corresponding to the first text portion. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Regarding claim 11, Ramanathan, Madavarapu. Jeyakumar, and Weston disclose the method of claim 8.
Weston further discloses wherein identifying the second text portion comprises:
converting the plurality of text portions into a corresponding plurality of embeddings vectors; (Weston, Figures 5 & 7; [0051]-[0058] describe wherein identifying the second text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors)
determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability state; (Weston, FIG 3B, FIG 4; [0039]-[0042] describe determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability state)
and determining the second text portion as the plurality of text portions having a respective vulnerability score above a threshold, (Weston, FIG 7; [0057]-[0059] describes and determining the second text portion as the plurality of text portions having a respective vulnerability score above a threshold)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Ramanathan, Madavarapu and Jeyakumar to include wherein identifying the second text portion comprises: converting the plurality of text portions into a corresponding plurality of embeddings vectors; determining a vulnerability score for each of the plurality of text portions based on a distance between each of the corresponding plurality of embeddings vectors and a reference vector representative of the vulnerability state; and determining the second text portion as the plurality of text portions having a respective vulnerability score above a threshold. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Regarding claim 12, Ramanathan, Madavarapu, Jeyakumar and Weston disclose the method of claim 11.
Madavarapu further discloses wherein: the type of the vulnerability state comprises a plurality of state types, (Madavarapu, [0021], [0028]-[0030] describes wherein: the type of the vulnerability state comprises a plurality of state types)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Madavarapu with the method and system of Ramanathan to include wherein: the type of the vulnerability state comprises a plurality of state types. One would have been motivated to provide chat analysis using machine learning (Madavarapu, [0061]).
Weston further discloses the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of state types, (Weston, Figures 3B & 4; [0039]-[0042] describes the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of state types)
and determining the type of the vulnerability action indicated by the second text portion further comprises: establishing the type of the vulnerability state as the state type represented by the reference vector closest to the embeddings vector corresponding to the second text portion, (Weston, Figure 7, [0057]-[0059] describes and determining the type of the vulnerability action indicated by the second text portion further comprises: establishing the type of the vulnerability state as the state type represented by the reference vector closest to the embeddings vector corresponding to the second text portion)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Ramanathan, Madavarapu and Jeyakumar to include the reference vector comprises a plurality of reference vectors, each of the plurality of reference vectors being representative of one of the plurality of state types, and determining the type of the vulnerability action indicated by the second text portion further comprises: establishing the type of the vulnerability state as the state type represented by the reference vector closest to the embeddings vector corresponding to the second text portion. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Claims 14-15 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Garg et al (“Garg,” US 20190114360) in view of Sorensen et al (“Sorensen” US 20200004765) and further in view of Jeyakumar et al (“Jeyakumar,” US 20200344251).
Regarding claim 14, Garg discloses a method comprising:
receiving an unstructured text file; (Garg, FIG 1, [0002], [0005]-[0006], [0018]-[0021] describes receiving an unstructured text file)
converting the unstructured text file into a standardized format file; (Garg, [0006], [0020]-[0021] describes converting the unstructured text file into a standardized format file)
processing the standardized format file to remove noise; (Garg, [0017]-[0021] describes processing the standardized format file to remove noise)
extracting a plurality of text features from the processed file; (Garg, [0020]-[0021] describes extracting a plurality of text features from the processed file)
Garg fails to explicitly disclose determining, by a machine learning model, a text feature indicative of a vulnerable feature, the machine learning model taking, as input, the plurality of text features and generating, as output, a likelihood that each of the plurality of text features includes the vulnerable feature;
However, in an analogous art, Sorensen discloses determining, by a machine learning model, a text feature indicative of a vulnerable feature, (Sorensen, FIG 2, [0044]-[0045])
the machine learning model taking, as input, the plurality of text features and generating, as output, a likelihood that each of the plurality of text features includes the vulnerable feature; (Sorensen, [0044]-[0045] describe the machine learning model taking, as input, the plurality of text features and generating, as output, a likelihood that each of the plurality of text features includes the vulnerable feature)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Sorensen with the method and system of Garg to include determining, by a machine learning model, a text feature indicative of a vulnerable feature, the machine learning model taking, as input, the plurality of text features and generating, as output, a likelihood that each of the plurality of text features includes the vulnerable feature. One would have been motivated to provide methods, systems, and computer media that allow for converting unstructured text to structured text (Sorensen, [0003]).
Garg and Sorensen fails to explicitly disclose generating a notification comprising an indication of the determined text feature and a recommended action based on the indicated vulnerable feature; and executing the recommended action.
However, in an analogous art, Jeyakumar discloses generating a notification comprising an indication of the determined text feature and a recommended action based on the indicated vulnerable feature; (Jeyakumar, Figures 3-4; [0105], [0107]-[0108] describe generating a notification comprising an indication of the determined text feature and a recommended action based on the indicated vulnerable feature)
and executing the recommended action, (Jeyakumar, Figures 4 & 6; [0107]-[0108]; [0133]-[0145] describe and executing the recommended action)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Jeyakumar with the method and system of Garg and Sorensen to include generating a notification comprising an indication of the determined text feature and a recommended action based on the indicated vulnerable feature; and executing the recommended action. One would have been motivated to provide techniques for detecting email based threats in the security field (Jeyakumar, [0002]).
Regarding claim 15, Garg, Sorensen and Jeyakumar disclose the method of claim 14.
Jeyakumar further disclose further comprising determining a type of the indicated vulnerable feature, (Jeyakumar, Figures 5 & 16; [0088]-[0094]; [0133]-[0145] describe further comprising determining a type of the indicated vulnerable feature)
wherein the recommended action corresponds to the determined type, (Jeyakumar, Figures 4 & 6; [0095]-[0104]; [0107]-[0108] describes wherein the recommended action corresponds to the determined type)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Jeyakumar with the method and system of Garg and Sorensen to include further comprising determining a type of the indicated vulnerable feature, wherein the recommended action corresponds to the determined type. One would have been motivated to provide techniques for detecting email based threats in the security field (Jeyakumar, [0002]).
Regarding claim 17, Garg, Sorensen and Jeyakumar disclose the method of claim 14.
Sorensen further discloses further comprising generating a confidence score indicative of the output likelihood from the machine learning model, (Sorensen, [0012], [0045] describe further comprising generating a confidence score indicative of the output likelihood from the machine learning model)
wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value, (Sorensen, [0046] describes wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Sorensen with the method and system of Garg to include further comprising generating a confidence score indicative of the output likelihood from the machine learning model, wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value. One would have been motivated to provide methods, systems, and computer media that allow for converting unstructured text to structured text (Sorensen, [0003]).
Jeyakumar also discloses wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value (Jeyakumar, [0107]-[0108] describes wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Jeyakumar with the method and system of Garg and Sorensen to include wherein the execution of the recommended action is in response to the confidence score exceeding a threshold value. One would have been motivated to provide techniques for detecting email based threats in the security field (Jeyakumar, [0002]).
Claims 16 and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Garg et al (“Garg,” US 20190114360), Sorensen et al (“Sorensen” US 20200004765) in view of Jeyakumar et al (“Jeyakumar,” US 20200344251) and further in view of Weston et al (“Weston,” US 20190332617).
Regarding claim 16, Garg, Sorensen and Jeyakumar disclose the method of claim 15.
Garg, Sorensen and Jeyakumar fail to explicitly disclose wherein the determining the type of the indicated vulnerable feature comprises: retrieving a set of reference text features that correspond to a set of vulnerability types, wherein each of the set of reference text features represent one of the set of vulnerability types; comparing the text feature indicative of the vulnerability to the set of reference text features; and determining the type of the indicated vulnerable feature based on a reference text feature of the set of reference text features most similar to the at least one text feature.
However, in an analogous art, Weston discloses wherein the determining the type of the indicated vulnerable feature comprises: retrieving a set of reference text features that correspond to a set of vulnerability types, wherein each of the set of reference text features represent one of the set of vulnerability types; (Weston, Figures 3B & 4; [0039]-[0042] describe wherein the determining the type of the indicated vulnerable feature comprises: retrieving a set of reference text features that correspond to a set of vulnerability types, wherein each of the set of reference text features represent one of the set of vulnerability types)
comparing the text feature indicative of the vulnerability to the set of reference text features; (Weston, FIG 4, [0041]-[0048] describes comparing the text feature indicative of the vulnerability to the set of reference text features)
and determining the type of the indicated vulnerable feature based on a reference text feature of the set of reference text features most similar to the at least one text feature, (Weston, FIG 7; [0057]-[0059] describes and determining the type of the indicated vulnerable feature based on a reference text feature of the set of reference text features most similar to the at least one text feature)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Garg, Sorensen and Jeyakumar to include wherein the determining the type of the indicated vulnerable feature comprises: retrieving a set of reference text features that correspond to a set of vulnerability types, wherein each of the set of reference text features represent one of the set of vulnerability types; comparing the text feature indicative of the vulnerability to the set of reference text features; and determining the type of the indicated vulnerable feature based on a reference text feature of the set of reference text features most similar to the at least one text feature. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Regarding claim 18, Garg, Sorensen and Jeyakumar disclose the method of claim 14.
Garg further discloses wherein the extraction of the plurality of text features comprises: dividing the processed file into a plurality of text chunks; and (Garg, FIG 4, [0020]-[0021] describe wherein the extraction of the plurality of text features comprises: dividing the processed file into a plurality of text chunks)
Weston further discloses generating a plurality of embeddings vectors representative of the plurality of text chunks, (Weston, FIG 5; [0051]-[0058] describes generating a plurality of embeddings vectors representative of the plurality of text chunks)
wherein the plurality of text features comprises the generated plurality of embeddings vectors, (Weston, Figures 5 & 7; [0051]-[0059] describe wherein the plurality of text features comprises the generated plurality of embeddings vectors)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Garg, Sorensen and Jeyakumar to include generating a plurality of embeddings vectors representative of the plurality of text chunks, wherein the plurality of text features comprises the generated plurality of embeddings vectors. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Regarding claim 19, Garg, Sorensen and Jeyakumar disclose the method of claim 18.
Weston further discloses wherein the determination that the text feature is indicative of the vulnerable feature comprises: retrieving a reference embeddings vector representative of a reference text feature, the reference text feature indicative of the vulnerable feature; (Weston, Figures 3B & 4; [0039]-[0042] describe wherein the determination that the text feature is indicative of the vulnerable feature comprises: retrieving a reference embeddings vector representative of a reference text feature, the reference text feature indicative of the vulnerable feature)
determining a distance between each of the generated plurality of embeddings vectors; (Weston, Figures 4 & 7; [0041]-[0048], [0057]-[0059] describes determining a distance between each of the generated plurality of embeddings vectors)
and determining the text feature indicative of the vulnerable feature as the text feature corresponding to a closest one of the generated plurality of embeddings vectors to the reference embeddings vector, (Weston, Figure 7; [0057]-[0059] describes and determining the text feature indicative of the vulnerable feature as the text feature corresponding to a closest one of the generated plurality of embeddings vectors to the reference embeddings vector)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Weston with the method and system of Garg, Sorensen and Jeyakumar to include wherein the determination that the text feature is indicative of the vulnerable feature comprises: retrieving a reference embeddings vector representative of a reference text feature, the reference text feature indicative of the vulnerable feature; determining a distance between each of the generated plurality of embeddings vectors; and determining the text feature indicative of the vulnerable feature as the text feature corresponding to a closest one of the generated plurality of embeddings vectors to the reference embeddings vector. One would have been motivated to provide a deep-learning model to predict relevant labels for text query (Weston, [0006]).
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Garg et al (“Garg,” US 20190114360), Sorensen et al (“Sorensen” US 20200004765) in view of Jeyakumar et al (“Jeyakumar,” US 20200344251) and further in view of Madavarapu et al (“Madavarapu,” US 20200302017)
Regarding claim 20, Garg, Sorensen and Jeyakumar disclose the method of claim 14.
Garg, Sorensen and Jeyakumar fail to explicitly disclose wherein the unstructured text file comprises a transcript of an online chat.
However, in an analogous art, Madavarapu discloses wherein the unstructured text file comprises a transcript of an online chat (Madavarapu, FIG 1, [0014]-[0017] describes wherein the unstructured text file comprises a transcript of an online chat)
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Madavarapu with the method and system of Garg, Sorensen and Jeyakumar to include wherein the unstructured text file comprises a transcript of an online chat. One would have been motivated to provide chat analysis using machine learning (Madavarapu, [0061]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAMES J WILCOX whose telephone number is (571)270-3774. The examiner can normally be reached M-F: 8 A.M. to 5 P.M..
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Luu T. Pham can be reached at (571)270-5002. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JAMES J WILCOX/Examiner, Art Unit 2439
/LUU T PHAM/Supervisory Patent Examiner, Art Unit 2439
Prosecution Insights