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 .
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
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, 5, 8-15, 17, 23, 25-27, and 29-31 are rejected under 35 U.S.C. 103 as being unpatentable over US 11019015 B1 to Shapira et al (“Shapira”) in view of US 11784948 B2 to Zong et al. (“Zong”) and in further view of US 10917374 B2 to Sarafa et al. (“Sarafa”) and in further view of Mukherjee et al. (“Mukherjee”).
Regarding claim 1, Shapira taught a system (“computing device”) comprising: memory comprising machine readable instructions; a processor to execute the machine readable instructions to cause the processor to (consider column 7, lines 15-23 and column 18, lines 4-14):
output on a display a first graphical user interface (GUI) comprising a message thread having a plurality of messages between a user and a recipient and a compose message field to receive information from the user (“editable text input field”), wherein the message thread is one of a plurality of message threads between the user and a respective plurality of recipients (consider Fig. 3C and column 20, lines 16-59); (consider at least Fig. 4 and column 23, lines 24-54, specifically lines 24-35) and
output on the display a second GUI with one or more graphical elements for specifying an attribute, wherein the attribute is specified by the user in response to the interacting with the one or more graphical elements. (consider column 8, line 63-column 9, line 29 regarding a “machine-learned model” which “analyze” “a direct message received from the user”) (consider column 20, lines 7-15, “In some cases, the message 216 may include a control 218 that is selectable by the first user (in this case, the “let us know” portion of the message 216), that, when selected by the first user, may provide feedback to the social networking system that the first user does not interpret the comment 202 to be offensive. Such feedback may be used to refine the parameters of the machine-learned model to make more accurate predictions on offensive content in the future.”)
Shapira may be interpreted as not expressly teaching wherein the attribute is a relationship attribute for the recipient, learning behavior of the user by tracking message in the plurality of message threads between the user and the plurality of recipients and the relationship attribute being used to evaluate the information within the compose message field wherein the processor evaluates the information within the compose message field, using a natural language processing (NLP) module trained on different types of relationship attributes, to determine the likelihood that the information within the compose message field is intended for the recipient based on behavior of the user, however, Shapira did teach that attributes including those for a recipient may be used to evaluate the information with the compose message field to reduce the likelihood that the information is not sent to a recipient (consider column 23, line 36-column 24, line 14, “The user interface 400 may include a selectable control 404 for sharing the direct message 402 with the second user. In examples, the social networking system may detect the potentially offensive direct message 402 prior to the direct message being shared with the second user, and provide the first user with the ability to revise and/or withdraw the direct message from being shared. For instance, when the first user selects the control 404, the direct message may be input into a machine-learned model trained to detect offensive content, such as described in relation to FIG. 1. The machine-learned model may output an offensiveness score of the direct message, and the score may be compared to a threshold corresponding to an offensiveness level of the direct message. If the score is greater than (or less than) the threshold score, the social networking system 106 of FIG. 1 may output an instruction to the computing device of the first user to display a notification 406 associated with the direct message 402 in the user interface 400, prior to the direct message being shared with the second user. The machine-learned model may take a variety of forms, as discussed in relation to the machine-learned model 114 of FIG. 1. For instance, the machine-learned model may be an image classifier trained to identify potentially offensive images or video in the direct message 402, an audio classifier trained to identify potentially offensive sounds or speech in the direct message 402, and/or a text classifier trained to identify potentially offensive text (including text detected using OCR) in the direct message 402, to name a few examples. In cases where the direct message 402 includes different types of content (e.g., an image, a sound, a text overlay, a text-based message, and so forth), the offensiveness score may be based on combining individual scores associated with an offensiveness of each of the content types included in the direct message 402 (e.g., by summation, weighted average, etc.). In some cases, the potentially offensive content identifiable by the machine-learned model may include different content categories, such as bullying content, hate speech content, nudity content, or others. The machine-learned model may be able to determine that the content included in the direct message 402 is a content category of the known offensive content categories. Based on determining that the direct message 402 is a content category of the known offensive content categories, the social networking system may customize the notification 406 accordingly, such as described in relation to FIGS. 3A-3C.”)
In an analogous art relating to reducing the likelihood that information within a compose message field is sent to an unintended recipient (consider column 2, line 65-column 3, line 11), Zong taught that a relationship attribute may be specified for a recipient and further taught learning behavior of the user by tracking message in a plurality of message threads between a user and a plurality of recipients and the specified relationship attribute being used to evaluate information within a compose message field and wherein the information within a compose message field within a GUI may be evaluated using a natural language processing (NLP) module trained on different types of relationship attributes to determine the likelihood that the information within the compose message field is intended for the recipient based on the behavior of the user. (consider column 3, lines 19-48, specifically “In certain embodiments described herein, a machine learning model for determining message unsuitability is trained using example documents created by combining prior messages. The prior messages comprise text strings, which can be collected from text-based messages as well as from speech converted to text by a speech-to-text processor…A system, in accordance with some embodiments, can extract one or more tokens from a message prior to transmitting the message over a communications network. The system can determine from the tokens one or more intended recipients of the message. At least one machine learning model corresponding to the one or more recipients of the message can be selected by the system, the machine learning model(s) trained with tokens extracted from prior messages that were combined to create a plurality training documents. The system can classify the one or more tokens extracted from the message using the machine learning model(s). The system can generate an alert message in response to determining based on the classifying that the chat message is unsuited for sending.”) (consider further column 6, lines 7-27, “Token processor 202 is automatically invoked by system 200 in response to a user creating a message using a device (e.g., smartphone, computer system) that can communicatively couple to a communications network. Prior to delivery of the message to one or more recipients over the communications network, token processor 202 extracts one or more tokens (words) from the message and, based on the extracted token(s), determines one or more intended recipients of the message. System 200 invokes token classifier 206. If token classifier 206, based on the classification model generated by classifier modeler 204, determines from the tokens extracted from the message that the message is unsuitable for sending, then alarm processor 208 is invoked. Alarm processor 208 generates a message warning the user (message sender) that the message is unsuited for sending. As defined herein, “unsuited for sending” refers to a message that is determined based on the machine learning classifier model to include language (identified by tokens representing words) that is likely (statistically) to offend a recipient of the message, create misunderstanding, and/or disclose privileged or confidential information to unauthorized persons.”) (consider further column 7, lines 9-50 regarding wherein “Classifier modeler 204 constructs a probabilistic topic model using machine learning and based on tokens extracted from prior messages” such that “The tokenized representations of prior messages are combined to create a plurality documents corresponding to message text strings, which can be stored in documents database 212 and used for training one or more probabilistic topic models using machine learning” and an “aggregated message document” “) (consider further column 8, line 48-column 9, line 12, specifically “The aLDA is augmented by assigning weights to extracted tokens, the weights influencing the respective probabilities and thus making topics associated with more heavily weighted tokens more likely to be “discovered.” A topic is more likely discovered or identified the greater the number of tokens (words) corresponding to the topic drawn from the documents. A token (word) that appears more frequently and/or that is more heavily weighted than other tokens thus has a greater likelihood that a topic corresponding to that token will be identified. The weights can be based on a group relationship graph. One or more group relationship graphs are stored in graph relationship graph database 214. A group relationship graph can be based on relationships between different individuals that belong to a specific group of individuals who exchange messages with one another over a specified time period. Different weights can be assigned to members of the group based on specific relationships.”) (consider further column 12, lines 18-22, “A group relationship graph can be constructed from explicit user input. Based on the documents one or more group models is generated at block 418. Based on a group model and individualized documents, individual models are generated at block 420.”) (consider further column 13, line 61-column 14, line 4, “The system at block 602 creates a plurality of aggregated messages documents (AMDs). The AMDs can be based on messages exchanged among a specified group of users during a specific span of time. Optionally, a block 604, the system can create, in response to user input, a group relationship graph based on a set of predefined relationships among the specified group of users. Tokens (words) contained in the AMDS can optionally be weighted at block 606. The weights can be determined based on the optionally created group relationship graph.”)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to modify the teachings of Shapira to include the taught features of Zong such that the modification includes every element as claimed. Given Shapira’s disclosure of attributes attributable to particular users to reduce the likelihood that information is not sent to a recipient, Zong specifically taught that relationship attributes may be specified for particular recipients of messages in order to readily determine the likelihood that information within a compose message field is intended for the recipient based on the behavior of the user (again, consider column 6, lines 7-27 and column 8, line 48-column 9, line 12). Given this specific advantage in Zong, one skilled in the art would have been motivated to modify the teachings of Shapira with the teachings of Zong such that the processor as taught in Shapira may be further enhanced by using information within a compose message field within a GUI to evaluate using a natural language processing (NLP) module trained on different types of relationship attributes to determine the likelihood that the information within the compose message field is intended for the recipient as taught in Zong so that the processor relationship attribute may be specified for a recipient, the specified relationship attribute being used to evaluate information within a compose message field, then evaluate that information within the compose message field by using a natural language processing (NLP) module trained on different types of relationship attributes, to determine the likelihood that the information within the compose message field is intended for the recipient as claimed. Therefore, such a modification of the teachings of Shapira with the teachings of Zong would have yielded nothing more than predictable results to one of ordinary skill in the art.
Shapira and Zong may be interpreted as not expressly teaching wherein the second GUI is further comprising a visual cue graphical rendering control element for enabling a theme for the message thread to be provided as part of the first GUI to reduce a likelihood that the information is sent to an unintended recipient, wherein the theme is selected by the user in response to interacting with the one or more graphical elements, however, as explained above, the combined teachings of Shapira and Zong did teach reducing the likelihood that the information is sent to an unintended recipient.
In an analogous art relating to displaying message threads, Sarafa taught a visual cue graphical rendering control element may be provided in a GUI for enabling a theme for the message thread to be provided as part of another GUI, wherein the theme is selected by the user in response to interacting with one or more graphical elements. (consider column 10, lines 21-37, “Each of the one or more messages may be represented by a particular message bubble, such as message bubble 335. A message bubble may represent an atomic messaging interaction. A message bubble may generally correspond to a defined geometric area in which the contents of a particular messaging exchange (e.g., text, media) are contained within the defined geometric area. A message bubble may have a distinct color or plurality of colors (e.g., one or more gradients) that distinguish it from a background of a message thread interaction display. A message bubble may have a distinctly-colored border, such as a black outline as depicted, or may have a border defined by the interface between differing colors of the message bubble and the background. In some embodiments, the color or colors of either or both the message bubbles and the background may be customized and configured by users of the messaging system 110.”) (consider further column 10, lines 51-61, specifically “To visually differentiate the contributions of different users, different colors may be used for different users in the message thread interaction display 330. Each user may be associated with a particular color…The names of participants in the message thread may each be displayed using the color assigned to each participant.”)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to modify the teachings of Shapira and Zong to include the taught features of Sarafa such that the modification includes every element as claimed. Given Shapira and Zong’s disclosure of a processor outputting a first GUI that allows a user to enter information as an electronic message communication to another user, a second GUI to specify a relationship attribute for the recipient, and then evaluating that information based on the relationship attribute using a NLP module to determine a likelihood that the information is intended to be received by an intended recipient, Sarafa taught that enabling a theme for a message thread so that the theme can be used to distinguish between users in a message thread (again, consider column 10, lines 21-37 and 51-61). Given this specific advantage in Sarafa, one skilled in the art would have been motivated to modify the teachings of Shapira and Zong with the teachings of Sarafa such that the second GUI as taught in Shapira and Zong may be modified to include a visual cue graphical rendering control element may be provided in a GUI for enabling a theme for a message thread to be provided as part of another GUI wherein the theme is selected by the user in response to interacting with the one or more graphical elements as taught in Sarafa so that the second GUI is further comprising a visual cue graphical rendering control element for enabling a theme for the message thread to be provided as part of the first GUI to reduce a likelihood that the information is sent to an unintended recipient wherein the theme is selected by the user in response to interacting with the one or more graphical elements as claimed. Therefore, such a modification of the teachings of Shapira and Zong with the teachings of Sarafa would have yielded nothing more than predictable results to one of ordinary skill in the art.
Shapira, Zong, and Sarafa may be interpreted as not expressly teaching wherein the machine readable instructions further cause the processor to:
select a subset of rules from a set of rules based on the relationship attribute specified for the recipient, each subset of rules of the set of rules specifying a permitted language for a respective relationship attribute of relationship attributes; and evaluate the information within the compose message field to determine whether the information is intended for the recipient further based on the selected subset of rules.
However, in an analogous art relating to providing user assistance in creating electronic messages by analysis of a created electronic message using learned behavior (consider column 3, lines 31-34 regarding the “sender and the recipient and their relationships” and “history of communication”), Mukherjee taught selecting a subset of rules from a set of rules based on a relationship attribute specified for the recipient, each subset of rules of the set of rules specifying a permitted language for a respective relationship attribute of relationship attributes; learn behavior of a user by tracking message in a plurality of message threads between the user and a plurality of message threads between the user and a plurality of recipients and evaluating information within a compose message field to determine whether the information is intended for the recipient based on the selected subset of rules. (consider column 7, lines 9-26, “Classifier modeler 204 constructs a probabilistic topic model using machine learning and based on tokens extracted from prior messages. The tokens can be stored in processed tokens database 210 and annotated with one or more indicators. The indicators can indicate, for example, the time the tokens representing the message were saved (“message time”); an assigned message identifier (“message ID”) used to correlate the token to the particular message from which the token was extracted; a message's author (“sender ID”); a message recipient (“recipient ID”); and/or a messaging group identifier (“group ID”), which as defined herein is an alpha-numeric sequence and/or other symbolic representation corresponding to the message sender and each message recipient. The tokenized representations of prior messages are combined to create a plurality documents corresponding to message text strings, which can be stored in documents database 212 and used for training one or more probabilistic topic models using machine learning.”) (consider further column 7, line 27-50 regarding the “aggregated messages document” for “aggregating multiple prior message exchanged over a specific time period between an identifiable group of users”) (consider column 11, lines 19-30, “More specifically, the appropriateness checker 512 may interpret an appropriateness of a communication prior to a user 532 sending the message 530 to another party 534 according to a plurality of identified contextual factors. In one aspect, by way of example only, the interpreting the appropriateness of the communications based on the plurality of identified contextual factors further includes interpreting a tone, a sentiment, emotion (e.g., levels of stress, frustration, anger, annoyance), keywords, phrases, text font, punctuation, and a sensitively level, relationship status, communication history, or a combination thereof between the user and another party 534.”) (consider column 13, lines 14-37, “The particular relationship between the sender, recipient, the broader audience and the topic of the message, and the nature (e.g., tone, sentiment (e.g., emotional state such as, for example, a stress or agitation level etc.) of the message/comment—negative/positive, factual, insensitive, inappropriate, derogatory, vulgar, inflammatory, judgmental, accusing, defamatory, controversial, unwarranted, (typical sentiment analysis) may be identified. As a next step, the type of media/audience, such as public media, private emails, inside organization forum, and other relationships may be identified. These relationships may be partially or wholly identified from a study of the user's profile. The appropriateness of text content of a communication may then be verified using country-specific, organization-specific rules, general policy rules, social etiquettes rules, which take into consideration the sender/recipient profiles, sender/recipient relationships, historical communications, and other factors previously identified. The appropriateness is then verified using other context specific rules. The sum total of the foregoing analysis is then compared against the aforementioned threshold and the identified, inappropriate content of the communication may be highlighted and an alert notification given where and when warranted.”)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to modify the teachings of Shapira, Zong, and Sarafa to include the taught features of Mukherjee such that the modification includes every element as claimed. Given Shapira, Zong, and Sarafa’s disclosure of a GUI that allows a user to enter information as an electronic message communication to another user and evaluating that information for relationship attributes, Mukherjee specifically taught that using a relationship attribute to evaluate information enables each subset of rules of the set of rules to be part of an analysis in order to determine whether language in the information is appropriate for a particular recipient based on the selected subset of rules (again, consider column 11, lines 19-30 and column 13, lines 14-37). Given this specific advantage in Mukherjee, one skilled in the art would have been motivated to modify the teachings of Shapira, Zong, and Sarafa with the teachings of Mukherjee such that the processor as taught in Shapira, Zong, and Sarafa may be modified to select a subset of rules from a set of rules based on a relationship attribute specified for the recipient, each subset of rules of the set of rules specifying a permitted language for a respective relationship attribute of relationship attributes; and evaluating information within a compose message field to determine whether the information is intended for the recipient based on the selected subset of rules as taught in Mukherjee so that selecting a subset of rules from a set of rules based on a relationship attribute specified for the recipient, each subset of rules of the set of rules specifying a permitted language for a respective relationship attribute of relationship attributes; and the evaluating information within a compose message field to determine whether the information is intended for the recipient may be further based on the selected subset of rules as claimed. Therefore, such a modification of the teachings of Shapira, Zong, and Sarafa with the teachings of Mukherjee would have yielded nothing more than predictable results to one of ordinary skill in the art.
Regarding claim 5, the combined teachings of Shapira, Zong, Sarafa and Mukherjee taught the system of claim 1.
Shapira taught wherein the message thread comprises one or more prior messages between the user and the recipient (consider column 18, lines 4-17 regarding the “user interfaces” on a “user computing device”) (again, consider at least Fig. 4 and column 23, lines 24-54, specifically lines 24-35).
Shapira and Zong may be interpreted as not expressly teaching wherein each recipient message of the one or more prior messages having a unique color to reduce the likelihood that the information within the compose message field is sent to the unintended recipient, however, Sarafa did teach these limitations. (again, consider column 10, lines 21-37, “Each of the one or more messages may be represented by a particular message bubble, such as message bubble 335. A message bubble may represent an atomic messaging interaction. A message bubble may generally correspond to a defined geometric area in which the contents of a particular messaging exchange (e.g., text, media) are contained within the defined geometric area. A message bubble may have a distinct color or plurality of colors (e.g., one or more gradients) that distinguish it from a background of a message thread interaction display. A message bubble may have a distinctly-colored border, such as a black outline as depicted, or may have a border defined by the interface between differing colors of the message bubble and the background. In some embodiments, the color or colors of either or both the message bubbles and the background may be customized and configured by users of the messaging system 110.”) (again, consider further column 10, lines 51-61, specifically “To visually differentiate the contributions of different users, different colors may be used for different users in the message thread interaction display 330. Each user may be associated with a particular color…The names of participants in the message thread may each be displayed using the color assigned to each participant.”)
The motivations regarding the obviousness of claim 1 also apply to claim 5, therefore, claim 5 is rejected under 35 USC § 103 as being unpatentable over the combined teachings of Shapira, Zong, Sarafa and Mukherjee and the same rationale supporting the conclusion of obviousness.
Regarding claim 8, the combined teachings of Shapira, Zong, Sarafa, and Mukherjee taught the system of claim 1.
Shapira, Zong, and Sarafa may be interpreted as not expressly teaching wherein the selected subset of rules identify permitted and/or unpermitted language for the relationship attribute specified for the recipient, however, Shapira did teach that permitted and/or unpermitted language may be identified when evaluating the information (consider column 23, lines 24-35, “FIG. 4 illustrates an example interface 400 usable to present a notification that is selectable to control the display of a potentially offensive message before the potentially offensive message is shared. The message may be a direct message in a messaging or social media application, a text message, an email, an instant message, or any other message type. The example interface 400 includes an editable text input field, in which a first user has input “You suck!!” as a direct message 402 to be shared with a second user (e.g., “anon8”). In some examples, other content types may be included in a direct message, such as images, video, emojis, and so forth as described above in relation to FIG. 1.”) (consider also column 23, line 55-column 24, line 14, “The machine-learned model may take a variety of forms, as discussed in relation to the machine-learned model 114 of FIG. 1. For instance, the machine-learned model may be an image classifier trained to identify potentially offensive images or video in the direct message 402, an audio classifier trained to identify potentially offensive sounds or speech in the direct message 402, and/or a text classifier trained to identify potentially offensive text (including text detected using OCR) in the direct message 402, to name a few examples. In cases where the direct message 402 includes different types of content (e.g., an image, a sound, a text overlay, a text-based message, and so forth), the offensiveness score may be based on combining individual scores associated with an offensiveness of each of the content types included in the direct message 402 (e.g., by summation, weighted average, etc.). In some cases, the potentially offensive content identifiable by the machine-learned model may include different content categories, such as bullying content, hate speech content, nudity content, or others. The machine-learned model may be able to determine that the content included in the direct message 402 is a content category of the known offensive content categories. Based on determining that the direct message 402 is a content category of the known offensive content categories, the social networking system may customize the notification 406 accordingly, such as described in relation to FIGS. 3A-3C.”)
Mukherjee taught wherein the selected subset of rules identify permitted and/or unpermitted language for the relationship attribute specified for the recipient (again, consider column 11, lines 19-30, “More specifically, the appropriateness checker 512 may interpret an appropriateness of a communication prior to a user 532 sending the message 530 to another party 534 according to a plurality of identified contextual factors. In one aspect, by way of example only, the interpreting the appropriateness of the communications based on the plurality of identified contextual factors further includes interpreting a tone, a sentiment, emotion (e.g., levels of stress, frustration, anger, annoyance), keywords, phrases, text font, punctuation, and a sensitively level, relationship status, communication history, or a combination thereof between the user and another party 534.”) (again, consider column 13, lines 14-37, “The particular relationship between the sender, recipient, the broader audience and the topic of the message, and the nature (e.g., tone, sentiment (e.g., emotional state such as, for example, a stress or agitation level etc.) of the message/comment—negative/positive, factual, insensitive, inappropriate, derogatory, vulgar, inflammatory, judgmental, accusing, defamatory, controversial, unwarranted, (typical sentiment analysis) may be identified. As a next step, the type of media/audience, such as public media, private emails, inside organization forum, and other relationships may be identified. These relationships may be partially or wholly identified from a study of the user's profile. The appropriateness of text content of a communication may then be verified using country-specific, organization-specific rules, general policy rules, social etiquettes rules, which take into consideration the sender/recipient profiles, sender/recipient relationships, historical communications, and other factors previously identified. The appropriateness is then verified using other context specific rules. The sum total of the foregoing analysis is then compared against the aforementioned threshold and the identified, inappropriate content of the communication may be highlighted and an alert notification given where and when warranted.”)
The motivations regarding the obviousness of claim 1 also apply to claim 8, therefore, claim 8 is rejected under 35 USC § 103 as being unpatentable over the combined teachings of Shapira, Zong, Sarafa, and Mukherjee and the same rationale supporting the conclusion of obviousness.
Regarding claim 9, the combined teachings of Shapira, Zong, Sarafa, and Mukherjee taught the system of claim 8.
Shapira taught wherein the permitted and/or unpermitted language includes permitted and/or unpermitted words, phrases, and/or sentences. (again, consider column 23, lines 24-35 and column 23, line 55-column 24, line 14)
Mukherjee also taught that the permitted and/or unpermitted language includes permitted and/or unpermitted words, phrases, and/or sentences (again, consider column 11, lines 19-30, “More specifically, the appropriateness checker 512 may interpret an appropriateness of a communication prior to a user 532 sending the message 530 to another party 534 according to a plurality of identified contextual factors. In one aspect, by way of example only, the interpreting the appropriateness of the communications based on the plurality of identified contextual factors further includes interpreting a tone, a sentiment, emotion (e.g., levels of stress, frustration, anger, annoyance), keywords, phrases, text font, punctuation, and a sensitively level, relationship status, communication history, or a combination thereof between the user and another party 534.”)
Regarding claim 10, the combined teachings of Shapira, Zong, Sarafa and Mukherjee taught the system of claim 1.
Shapira may be interpreted as not expressly teaching wherein the machine readable instructions further cause the processor to provide an alert on the display indicating that the information within the compose message field is not intended for the recipient based on the evaluation of the information within the compose message field and the relationship attribute specified for the recipient, however, Shapira did teach wherein the machine readable instructions further cause the processor to provide an alert on the display indicating that the information within the compose message field is not intended for the recipient based on the evaluation of the information within the compose message field. (again, consider column 3, line 62-column 4, line 2, “For instance, a notification may be output on a screen of a process flow when sharing a profile post or a story, but before the content is shared with other users. In the case of a direct message, a notification may be output in real time as the user inputs characters into a message, and/or upon completion of a message (e.g., after the user selects “send” but before the content is shared with the recipient(s) of the message).”) (again, consider column 23, lines 38-54 regarding the “notification associated with the direct message in the user interface”)
Zong taught wherein the machine readable instructions further cause the processor to provide an alert on the display indicating that the information within the compose message field is not intended for the recipient based on the evaluation of the information within the compose message field and the relationship attribute specified for the recipient. (consider column 3, lines 19-48, specifically “In certain embodiments described herein, a machine learning model for determining message unsuitability is trained using example documents created by combining prior messages. The prior messages comprise text strings, which can be collected from text-based messages as well as from speech converted to text by a speech-to-text processor…A system, in accordance with some embodiments, can extract one or more tokens from a message prior to transmitting the message over a communications network. The system can determine from the tokens one or more intended recipients of the message. At least one machine learning model corresponding to the one or more recipients of the message can be selected by the system, the machine learning model(s) trained with tokens extracted from prior messages that were combined to create a plurality training documents. The system can classify the one or more tokens extracted from the message using the machine learning model(s). The system can generate an alert message in response to determining based on the classifying that the chat message is unsuited for sending.”) (consider further column 6, lines 7-27, “Token processor 202 is automatically invoked by system 200 in response to a user creating a message using a device (e.g., smartphone, computer system) that can communicatively couple to a communications network. Prior to delivery of the message to one or more recipients over the communications network, token processor 202 extracts one or more tokens (words) from the message and, based on the extracted token(s), determines one or more intended recipients of the message. System 200 invokes token classifier 206. If token classifier 206, based on the classification model generated by classifier modeler 204, determines from the tokens extracted from the message that the message is unsuitable for sending, then alarm processor 208 is invoked. Alarm processor 208 generates a message warning the user (message sender) that the message is unsuited for sending. As defined herein, “unsuited for sending” refers to a message that is determined based on the machine learning classifier model to include language (identified by tokens representing words) that is likely (statistically) to offend a recipient of the message, create misunderstanding, and/or disclose privileged or confidential information to unauthorized persons.”)
The motivations regarding the obviousness of claim 1 also apply to claim 10, therefore, claim 10 is rejected under 35 USC § 103 as being unpatentable over the combined teachings of Shapira and Zong and the same rationale supporting the conclusion of obviousness.
Regarding claim 11, the combined teachings of Shapira, Zong, Sarafa and Mukherjee taught the system of claim 1.
Shapira, Zong and Sarafa may be interpreted as not expressly teaching wherein the machine readable instructions further cause the processor to provide a notification on the display to confirm that the user intends to send the information within the compose message field to the recipient in response to the evaluation indicating that the information within the compose message field is not likely intended for the recipient based on the relationship attribute specified for the recipient, however, Shapira did teach providing a notification on the display to confirm that the user intends to send the information within the compose message field to the recipient in response to the evaluation of the information with the compose message field. (consider column 24, lines 35-48, “Although not explicitly illustrated in FIG. 4, in some cases, the direct message 402 may be shared with the second user responsive to the duration of time elapsing. By sharing the direct message 402 after a duration of time has elapsed, the techniques described herein give the first user an opportunity to reconsider sharing the potentially offensive content, but may allow the direct message 402 to be shared without the first user having to execute an additional action in the process of sharing the direct message 402 (e.g., by affirmatively selecting an additional control to share the direct message). However, examples are considered in which an affirmative control may be required to be selected by the first user to continue with posting a potentially offensive direct message 402.”)
Mukherjee taught wherein the machine readable instructions further cause the processor to provide a notification on the display to confirm that the user intends to send the information within the compose message field to the recipient in response to the evaluation indicating that the information within the compose message field is not likely intended for the recipient based on the relationship attribute specified for the recipient. (consider column 12, lines 18-28, “In an additional aspect, the appropriateness checker 512 may also detect and determine an intended recipient for receiving the communication. If an alternative email is selected that is not associated with the intended recipient, the message blocker 540 may hold/restrict sending the communication for the selected period of time to ensure the recipient is the correct and intended recipient. In short, the message blocker 540 may intercept and restrict delivery of the communication upon determining from the communication that the communication is being sent or about to be sent to a wrong or unintended user.”) (consider further column 12, lines 29-35, “If the user decides to continue with the communication without accepting the suggested delay, the message may be sent and the learning module 522 will obtain feedback from the user. The user then provides feedback, which may be combined with other user feedback of the communications (or similar communications) from users 534 to form reactions 526.”)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine the teachings of these references such that their combination includes every element as claimed. One skilled in the art could have combined the teachings by known methods such as integration of software routines with no changes to the operation of either reference such that, in combination, each element merely performs the same function as it does separately. Additionally, the Examiner finds that, based on the references’ analogous disclosure regarding GUIs and displaying information to a user, further demonstrates that a combination of their features would have been known and obvious. Therefore, such a combination of the teachings of the references would have yielded nothing more than predictable results to one of ordinary skill in the art.
Regarding claim 12, the combined teachings of Shapira, Zong, Sarafa, and Mukherjee taught the system of claim 11.
Shapira taught wherein the alert is a graphical element alert. (again, consider column 3, line 62-column 4, line 2, “For instance, a notification may be output on a screen of a process flow when sharing a profile post or a story, but before the content is shared with other users. In the case of a direct message, a notification may be output in real time as the user inputs characters into a message, and/or upon completion of a message (e.g., after the user selects “send” but before the content is shared with the recipient(s) of the message).”) (again, consider column 23, lines 38-54 regarding the “notification associated with the direct message in the user interface”)
Zong also taught this limitation. (again, consider column 3, lines 19-48 and column 6, lines 7-27 regarding the “alert”/”alarm”)
Mukherjee also taught this limitation. (consider column 3, lines 9-22, “In one aspect, an appropriateness of a communication may be interpreted prior to a user sending the communication to another party according a plurality of identified contextual factors. The user may be alerted to a possible negative impact of sending the communication to the another party/recipient if the interpreted appropriateness is less than a predetermined threshold. For example, an identified word, phrase, or other part of the communication may be highlighted (e.g., highlight content of an email) to alert the user of the possible negative impact upon a one or more recipients. A delay in sending the communication may be suggested for a selected period of time to implement one or more suggestive corrective actions to the communication.”) (consider further column 14, lines 14-30, “Supervisor “A” may be notified that their communication is potentially inappropriate/offensive given the context described in this example. The notification, for example, may include a message “Alert. It is detected that you have inappropriate content of X, Y, and Z about employee “C.” You have mistakenly listed employee “C” as the recipient of the email as opposed to supervisor “B.” Here is a list of corrective actions you may take to resolve the concern: 1) change the recipient from employee “C” to supervisor “B,” or 2) delete or edit the words X, Y, and Z.” In one aspect, the notification may be one of a variety of notification means such as an alert, a pop-up window over the email, a text message to an approved Internet of Things (“IoT”) device (e.g., a smartwatch that the user may be known to be wearing) and/or any other defined notification according to user preferences.”)
Regarding claim 13, the combined teachings of Shapira, Zong, Sarafa, and Mukherjee taught the system of claim 12.
Shapira, Zong, and Sarafa may be interpreted as not expressly teaching wherein the graphical element alert includes a change recipient option to select a different recipient for sending the information within the compose message field, however, Mukherjee did teach these limitations. (again, consider further column 14, lines 14-30, “Supervisor “A” may be notified that their communication is potentially inappropriate/offensive given the context described in this example. The notification, for example, may include a message “Alert. It is detected that you have inappropriate content of X, Y, and Z about employee “C.” You have mistakenly listed employee “C” as the recipient of the email as opposed to supervisor “B.” Here is a list of corrective actions you may take to resolve the concern: 1) change the recipient from employee “C” to supervisor “B,” or 2) delete or edit the words X, Y, and Z.” In one aspect, the notification may be one of a variety of notification means such as an alert, a pop-up window over the email, a text message to an approved Internet of Things (“IoT”) device (e.g., a smartwatch that the user may be known to be wearing) and/or any other defined notification according to user preferences.”) (consider further column 14, lines 31-46, “The electronic communication coaching service may request a delay in sending the communication and/or automatically delay/restrict sending the communication for a selected period of time. The delay in sending the email may enable supervisor “A” to correct the inappropriate content. In one aspect, the electronic communication coaching service may be enabled, under direction of one or more processor devices, to bypass the request for delay, automatically restrict the delivery of the email, and automatically implement the one or more corrective actions such as, for example, automatically censor/edit the inappropriate contact or change the intended recipient. The email may be sent to the appropriate recipient (e.g., supervisor “B”) upon correcting the inappropriate/offensive content and/or changing the email address for the recipient from employee “C” to supervisor “B.”)
The motivations regarding the obviousness of claim 11 also apply to claim 13, therefore, claim 13 is rejected under 35 USC § 103 as being unpatentable over the combined teachings of Shapira, Zong, Sarafa, and Mukherjee and the same rationale supporting the conclusion of obviousness.
Regarding claim 14, the combined teachings of Shapira, Zong, Sarafa, and Mukherjee taught the system of claim 1.
Shapira may be interpreted as not expressly teaching wherein the specified relationship attribute comprises one of a professional relationship, a romantic relationship, a friendly relationship, and a familiar relationship, however, Zong did teach these limitations. (consider column 10, lines 4-21, “Token classifier 206 determines whether any of the words correspond to topics that make the message unsuitable for sending. A topic can be designated as one whose ass