DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . In communications filed on 07/22/2026. Claims 1, 3, 21, and 25 are amended. Claims 2, and 12-20 are cancelled. Claims 1, 3-11, and 21-29 are pending in this examination.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. This examination is in response to US Patent Application No. 18/810,744.
Examiner Note
Applicant’s amendment to independent claims obviates previously raised claims 1-11, and 21-29, 35 U.S.C .101 Alice decision rejection.
Examiner withdraws the 112(a) , first paragraph rejections for claims 4-5, and 24 per applicant arguments on pages of 14-17 of the argument/remarks filed on 07/22/2026.
Applicants’ arguments filed 07/22/2026 on pages 21-24, and 26-28 regarding claims 5, and 24 rejections have been fully considered and are persuasive. Examiner withdraws claims 5, and 24 rejections.
Response to Arguments
Applicants’ arguments filed 07/22/2026 have been fully considered but they are not persuasive:
Applicant respectfully argues on pages 24-26 of remarks filed on 07/22/2026 that no combination of Bar Noy and Abobaker teaches or suggest “identifying one or more keywords in user agent strings for users in the activity data “and “ applying a machine learning model to the user agent strings”
Examiner respectfully disagrees with applicant argument on pages 24-26 filed with remarks on 07/22/2026. Examiner maintains the rejection.
Bar Noy discloses identifying one or more keywords in user agent strings for users in the activity data [¶64, Potential attack indicators that are to be identified by the detection module 140 may be, for example, any keyword or keyword strings in text or objects in the scanned payloads that are anomalous or suspicious in nature. Examples of potential attack indicators include, but are not limited to, keywords that are indicative of SQL injection attempts and probing attempts via netcat utility to bypass authentication or encryption (i.e., backdoor entry). For example, the keyword string ‘105 or 1=1’ is indicative of an SQL injection (in the HTTP request) to attempt to gain access to restricted information (for example usernames and/or passwords from a database or table that contains usernames and/or passwords). The following is a list of non-limiting examples of sample portions of suspicious HTTP requests and the potential attack indicators (keyword/keyword strings) associated with each sample: see table 1].
Abobaker discloses applying a machine learning model to the user agent strings.
[¶5, Techniques disclosed herein relate generally to text classification. More specifically and without limitation, techniques disclosed herein relate to techniques for fusing word embeddings with word scores for text classification (e.g., in natural language processing). A machine learning approach to text classification may include presenting the text to a machine learning model as one or more low-dimensional continuous feature vectors (e.g., word embeddings)], and
[¶119, A machine learning approach to text classification may include presenting the text to a machine learning model as one or more feature vectors (e.g., word embeddings). The machine learning model may be complex: for example, a long short-term memory (LS™) or a convolutional neural network (CNN). We have found, however, that using a simpler machine learning model may be sufficient. A deep average network or “DAN” (also called a “deep averaging network”), for example, is a simple machine learning model that may be highly performing for the task of text classification].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teaching of Bar Noy by incorporating “a machine learning model”, as taught by Abobaker. One could have been motivated to do so in order to implement a machine learning model to text classification and include techniques for fusing word embeddings with word scores for text classification [ Abobaker, Abstract, ¶¶5, 119].
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-4, 6-9, 11-23, and 25-29 are rejected under 35 U.S.C. 102(a) (1) as being anticipated by BAR NOY ( US2019/0334940).
Regarding claim 1, Bar Noy discloses a method for detecting bots, the method comprising: receiving activity data associated with a website, the activity data including data for a plurality of users of the website [Abstract, Computerized methods and systems receive a request message from a client device that is addressed to a web server hosting at least one web application. the request message is analyzed to identify potential attack indicators that are present in the request message. Each potential attack indicator has a score. A reputation score is assigned to the request message that is associated with behavior of the client device relative other client devices sending request message to the web server] , and [¶80, The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users ( equated to plurality of users)visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users. [¶133, As mentioned above, the bot detection module 154 provides the system 130 with bot detection capabilities. As is generally known in the art, bots are malicious software agents that run on infected client computers( equated to plurality of users) which may collectively form a botnet. Each bot in a botnet is remotely controlled by a malware originator from a command-and-control server, and runs autonomously and automatically on the infected machine, allowing infection of other machines (i.e., computers) via propagation of malicious content by the bots to web applications visited by other machines. The malicious content may be propagated to the web applications via automated HTTP requests sent by the bots. The purpose of the bot detection module 154 is to identify HTTP requests addressed to the web server 120 that are sent by bots], and
identifying a first set of bots in the activity data by identifying self-identifying bots, including identifying one or more keywords associated with self-identifying bots in user agent strings of users of the plurality of users[¶64, Potential attack indicators that are to be identified by the detection module 140 may be, for example, any keyword or keyword strings in text or objects in the scanned payloads that are anomalous or suspicious in nature. Examples of potential attack indicators include, but are not limited to, keywords that are indicative of SQL injection attempts and probing attempts via netcat utility to bypass authentication or encryption (i.e., backdoor entry). For example, the keyword string ‘105 or 1=1’ is indicative of an SQL injection (in the HTTP request) to attempt to gain access to restricted information (for example usernames and/or passwords from a database or table that contains usernames and/or passwords). The following is a list of non-limiting examples of sample portions of suspicious HTTP requests and the potential attack indicators (keyword/keyword strings) associated with each sample: see table 1]; and
identifying a second set of bots in the activity data by applying one or more rules for identifying bots to at least one of an internet protocol (IP) address or a user agent string associated with users of the plurality of users [¶79, The source collection module 142 initially retrieves data from particular fields of the HTTP requests in order to extract source identifiers (that correspond to the particular fields) from the source information associated with each HTTP request. Examples of such source identifiers include, but are not limited to, the IP address of the source, the User-Agent, client computer fingerprint, session ID, and combinations thereof. For each particular source identifier, the source collection module 142 assigns attributes to the source identifier which contain data and information related to the behavior of the source. The attributes include, for example, site coverage, request ratio, vulnerable URL request activity, and response time anomalies (e.g., the elapsed time between receiving an HTTP request and sending an HTTP response message)], and [¶98, In order to better explain the operation of the correlation module 148, the following case is considered as an example. In the example, the suspicious HTITP request has two attributes and the attack database 164 stores four attacks. The attributes of the suspicious HTTP request are: a) the list of potential attack indicators and, b) the IP address of the source that sent the suspicious HTTP request. The four attacks in the attack database 164 are denoted by Attack ID.sub.1, Attack ID.sub.2, Attack ID.sub.3, and Attack ID.sub.4. Attack ID.sub.1 has assigned to it the following attributes: a) one or more potential attack indicators common to the indicators in the list of potential attack indicators in the suspicious HTTP request, and b) an IP address that is different from the IP address of the source that sent the suspicious HTTP request. Attack ID.sub.2 has assigned to it the following attribute: User-Agent that is different from the User-Agent of the suspicious HTTP request. Attack ID.sub.3 has assigned to it the following attribute: User-Agent that is different from the User-Agent of the suspicious HTTP request. Attack ID.sub.4 has assigned to it the following attribute: one or more potential attack indicators common to the indicators in the list of potential attack indicators in the suspicious HTTP request. In such an example case, in response to the query related to the list of potential attack indicators, the attack database 164 will return a first list having the attacks denoted by Attack ID.sub.1 and Attack ID.sub.4. In response to the query related to the IP address of the source that sent the suspicious HTTP request, the attack database 164 will return a second list having the attack denoted by Attack ID.sub.1], and [ ¶¶92, 11, 131, 151]; and
and identifying a third set of bots in the activity data by identifying outlier activity including applying one or more models to generate a plurality of bot confidence scores for the plurality of users in the activity data and, for each user of the plurality of users, comparing a respective bot confidence score of the plurality of bot confidence scores to a threshold value [¶5, The present invention is directed to computerized methods and systems which provide security for web applications. The computerized methods and systems calculate weighted scores based on context-driven inspection of communication protocol request messages (e.g., HTTP requests) and behavioral models of the sources that send the request messages], and
[¶40, Initially, throughout this document, references are made to module output scores that are generated/calculated by modules as a result of analysis of data, by said modules. These scores can generally be characterized as likelihood or confidence scores, which define the probability (and interchangeably the likelihood or confidence) that the data being analyzed falls within a specific category. To determine the category, the output scores are evaluated against a threshold criterion, which in certain embodiments is a threshold value. Data that produces output scores that are above (or equal to) the threshold value are placed in a first category, and data that produces output score which are less than the threshold value are placed in a second category. In such embodiments, the higher the output score the higher the probability that the data is be placed in the first category (and the lower the probability that the data is be placed in the second category). Conversely, the lower the output score the higher the probability that the data is be placed in the second category (and the lower the probability that the data is be placed in the first category)]; and [¶78, In certain embodiments, once the detection module 140 marks (i.e., flags) the HTTP request as suspicious, the behavioral analysis module 146 begins executing functionality. For each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 146 analyzes the HTTP request to make determinations about the HTTP request based on the behavior of the source (i.e., the client device 100) that sent the suspicious HTTP request. The analysis is performed in cooperation with the source collection module 142, which collects information about sources (i.e., client computers) that send HTTP requests. In particular, the source collection module 142 analyzes incoming HTTP requests to extract information pertaining to the sources sending the requests, and subsequently stores the extracted information as entries in a source database 162. In this way, the source collection module 142 compiles web traffic data (i.e., builds up a history) and stores the historical data in the source database 162], and [¶¶85-87] Returning now to the behavioral analysis module 146, for each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 142 queries the source database 162 to retrieve the source identifiers for the particular source that has made the suspicious HTTP request. If the source is not in the source database 162, the behavioral analysis module 146 updates the source database 162 by inserting the source (and the relevant source identifiers and attributes) into the source database 162. Note that initially (i.e., upon initial execution of the functionality of the system 130) the suspicious source will be absent from the source database 162. This is due to the fact that the behavioral analysis module 146 and the source collection module 142 collect information over time as client computers make HTTP requests. Once the behavioral analysis module 146 and the source collection module 142 have collected a sufficient amount of historical data, the source database 162 will return the source identifiers (to the behavioral analysis module 146) for the particular source that has made the suspicious HTTP request. Once the behavioral analysis module 146 retrieves the relevant information from the source database 162, the behavioral analysis module 146 performs attribute comparisons to calculate non-binary attribute scores and non-binary source identifier scores. Specifically, for each source identifier, the behavioral analysis module 146 compares each attribute to like attributes of legitimate sources (from the source database 162) to calculate attribute scores. The behavioral analysis module 146 then combines all of the attribute scores, for the particular source identifier, to calculate a non-binary source identifier score. The behavioral analysis module 146 then combines all of the source identifier score to calculate a non-binary reputation score for the particular HTTP request. The aforementioned source identifier score and reputation score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm], and [¶151, The behavioral analysis module 146 may also generate DoS confidence scores for HTTP requests and source IP addresses associated web applications that are marked (i.e., flagged) as being under DoS attack by the Layer7 DoS protection module 174. The Layer7 DoS protection module 174 may take one or more protective actions against HTTP requests and source IP addresses that correspond to DoS confidence scores indicative of DoS attack. Such protective actions include, but are not limited to, blocking such HTTP requests and/or blacklisting future HTTP requests from that source IP address for a predefined duration (e.g., 5 minutes)], and [¶¶41-42, 60, 66, 87-90, 92, 95-97, 129-130]; and
providing an output to a security system, the output including a classification of whether each user of the plurality of users is a bot [¶88, If the reputation score is above a threshold value, the behavioral analysis module 146 classifies the HTTP request as a legitimate request, and marks (i.e., flags) the HTTP request as a false-positive (referred to hereinafter as “FP”). If the reputation score is below the threshold value, the behavioral analysis module 146 maintains the classification of the HTTP request as a suspicious request, and marks (i.e., flags) the HTTP request as a true-positive (referred to hereinafter as “TP”). The FP/TP marking (i.e., flag), together with the reputation score, are used by the indicator analysis module 144 to update the score entries in the indicator database 160. In other words, the indicator analysis module 144 updates (i.e., adjusts) the scores of the potential attack indicators and the scores for the combinations of potential attack indicators associated with that particular HTTP request, based on the reputation scores and the FP/TP markings (i.e., flags).
[¶113, Upon calculating the composite score, the composite scoring module 152 evaluates the composite score against a threshold criterion, for example a threshold value. If the composite score is greater than or equal to the threshold value, the system 130 deems the HTTP request as a legitimate request, and releases the HTTP request through to the web application 121. If the composite score is less than the threshold value, the system 130 deems the HTTP request as a malicious request, and takes protective action, for example by blocking the HTTP request or redirecting the web client 108 to a homepage]; and
and blocking, by the security system, traffic associated with at least one of an IP address or a user agent string of a bot identified in the output[¶113, Upon calculating the composite score, the composite scoring module 152 evaluates the composite score against a threshold criterion, for example a threshold value. If the composite score is greater than or equal to the threshold value, the system 130 deems the HTTP request as a legitimate request, and releases the HTTP request through to the web application 121. If the composite score is less than the threshold value, the system 130 deems the HTTP request as a malicious request, and takes protective action, for example by blocking the HTTP request or redirecting the web client 108 to a homepage].
Regarding claim 3, Bar Noy discloses wherein identifying the third set of bots in the activity data by identifying outlier activity comprises, for each user of a plurality of users in the activity data[¶¶78, 85-87], and
for each feature of a plurality of features, identifying a feature value for the user for each feature of the plurality of features
[¶78] In certain embodiments, once the detection module 140 marks (i.e., flags) the HTTP request as suspicious, the behavioral analysis module 146 begins executing functionality. For each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 146 analyzes the HTTP request to make determinations about the HTTP request based on the behavior of the source (i.e., the client device 100) that sent the suspicious HTTP request. The analysis is performed in cooperation with the source collection module 142, which collects information about sources (i.e., client computers) that send HTTP requests. In particular, the source collection module 142 analyzes incoming HTTP requests to extract information pertaining to the sources sending the requests, and subsequently stores the extracted information as entries in a source database 162. In this way, the source collection module 142 compiles web traffic data (i.e., builds up a history) and stores the historical data in the source database 162], and
[¶131] The management module 156 is also linked to the display device 170, which enables the system 130 to provide display and user interface capabilities. The management module 156 collects/aggregates data from the modules of the system 130 and processes that data to produce output data for viewing by a system administrator of the system 130 via the display device 170. The output data may include qualitative and quantitative data and information, including, for example, the total number of HTTP requests received by the system 130, the number of times protective actions were taken against HTTP requests, statistical information pertaining to the HTTP requests (e.g., average number of FP, average number of TP, etc.), the scores provided by the various modules, attack IDs (either all of the attack IDs or a snippet of the attack IDs from the attack database 164) and relevant data pertaining to each displayed attack ID (e.g., URL attacked, web server/application attacked, source IP address, etc.), data logs, and the number of bot attacks detected by the bot detection module 154], and
flagging the feature in response to determining that the feature value for the user is more than a distance from a center value ;determining a score based in part on a number of flagged features; and in response to determining that the score is greater than a threshold, identifying the user as a bot.
[¶¶65-66, The potential attack indicators are evaluated against a decision criterion, which may be set by a system administrator that manages the system 130, to determine whether a criterion is satisfied. In certain embodiments, the decision criterion is a threshold value. In such embodiments, the detection module 140 tallies the number of potential attack indicators in the HTTP request and evaluates the number of potential attack indicators against the threshold value. If the potential attack indicators satisfy the decision criterion (e.g., if number of potential attack indicators exceeds the threshold value), the detection module 140 marks (i.e., flags) the HTTP request as suspicious. If the potential attack indicators dissatisfy (i.e., do not satisfy) the decision criterion (e.g., if the number of potential attack indicators is less than or equal to the threshold value), the HTTP request is deemed as a legitimate request and is released through to the web application 121 by the system 130], and
[¶84, The source collection module 142 collects all of the source identifiers and associated attributes and stores the source identifiers and attributes in the source database 162. As a result, the source database 162 stores information about all of the sources that have made HTTP requests, wherein for each source associated with a particular HTTP request, each source identifier of that source is stored in the source database 162 together with all of the attributes associated with each respective source identifier. The source entries in the source database 162 can also be marked (i.e., flagged) with a legitimacy score. For example, sources that have made only legitimate HTTP requests can be marked (i.e., flagged) with a high score in the source database 162, thereby indicating that such sources are legitimate sources], and [¶¶78, 88, 103, 125]
Regarding claim 4, Bar Noy discloses, wherein identifying the third set of bots in the activity data by identifying outlier activity comprises, for each user of a plurality of users in the activity data [¶¶78, 85-87], and
generating a first score for the user by applying a statistical model
[¶131, The management module 156 is also linked to the display device 170, which enables the system 130 to provide display and user interface capabilities. The management module 156 collects/aggregates data from the modules of the system 130 and processes that data to produce output data for viewing by a system administrator of the system 130 via the display device 170. The output data may include qualitative and quantitative data and information, including, for example, the total number of HTTP requests received by the system 130, the number of times protective actions were taken against HTTP requests, statistical information pertaining to the HTTP requests (e.g., average number of FP, average number of TP, etc.), the scores provided by the various modules, attack IDs (either all of the attack IDs or a snippet of the attack IDs from the attack database 164) and relevant data pertaining to each displayed attack ID (e.g., URL attacked, web server/application attacked, source IP address, etc.), data logs, and the number of bot attacks detected by the bot detection module 154], and
generating a second score for the user by applying a clustering model.
[¶151, The behavioral analysis module 146 may also generate DoS confidence scores for HTTP requests and source IP addresses associated web applications that are marked (i.e., flagged) as being under DoS attack by the Layer7 DoS protection module 174. The Layer7 DoS protection module 174 may take one or more protective actions against HTTP requests and source IP addresses that correspond to DoS confidence scores indicative of DoS attack. Such protective actions include, but are not limited to, blocking such HTTP requests and/or blacklisting future HTTP requests from that source IP address for a predefined duration (e.g., 5 minutes)], and
weighing and aggregating at least the first score and second score to generate a bot confidence score; and based on the bot confidence score, determining whether to classify the user as a bot.
[¶40, Initially, throughout this document, references are made to module output scores that are generated/calculated by modules as a result of analysis of data, by said modules. These scores can generally be characterized as likelihood or confidence scores, which define the probability (and interchangeably the likelihood or confidence) that the data being analyzed falls within a specific category. To determine the category, the output scores are evaluated against a threshold criterion, which in certain embodiments is a threshold value…], and
[¶¶75-76, After the list of combinations is generated, for a particular suspicious HTTP request, the indicator analysis module 144 queries the indicator database 160 to check whether each combination in the list exists in the indicator database 160. For each particular combination that is in the indicator database 160, the indicator analysis module 144 retrieves (from the indicator database 160) the non-binary score for the particular combination, as well as the individual scores associated with the potential attack indicators in the particular combination, from the indicator database 160. If a particular combination is absent from the indicator database 160, the database is updated to include the particular combination. The indicator analysis module 144 then retrieves the individual scores associated with the potential attack indicators in the new combination from the indicator database 160. The indicator analysis module 144 then generates a combined non-binary score for the new combination based on the retrieved individual scores and stores the combined score in the indicator database 160. The combined score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm.], and
[¶144, The injected JavaScript code, when executed, also collects user activity related data, and compares the collected data with nominal activity data collected from the non-bot client computers. As a result, the JavaScript code, when executed, calculates a score indicative of the confidence level that the client device 100 is a bot (or a non-bot, i.e., human user). For clarity of illustration, a higher score is indicative of a higher confidence that the client device 100 is a bot, whereas a lower score is indicative of a lower confidence that the client device 100 is a bot. For example, the JavaScript code, when executed, may calculate keystroke speed in response to detected keystroke events, and may further compare the calculated keystroke speed with average human keystroke speed to assign a keystroke score. As a further example, the JavaScript code, when executed, may calculate mouse click speed in response to detected mouse click events, and may further compare the calculated mouse click speed with average human mouse click speed to assign a mouse click score. The various scores may be aggregated/combined to form an overall score, which is provided to the bot detection module 154 via the generated cookie], and [¶¶151].
Regarding claim 6, Bar Noy discloses, wherein identifying the third set of bots in the activity data by identifying outlier activity comprises [¶¶78, 85-87], and
identifying a plurality of features; and for each user of a plurality of users in the activity data, determining whether the user is an outlier based on values of the plurality of features for the user wherein the plurality of features comprises a demand
[¶78, In certain embodiments, once the detection module 140 marks (i.e., flags) the HTTP request as suspicious, the behavioral analysis module 146 begins executing functionality. For each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 146 analyzes the HTTP request to make determinations about the HTTP request based on the behavior of the source (i.e., the client device 100) that sent the suspicious HTTP request. The analysis is performed in cooperation with the source collection module 142, which collects information about sources (i.e., client computers) that send HTTP requests. In particular, the source collection module 142 analyzes incoming HTTP requests to extract information pertaining to the sources sending the requests, and subsequently stores the extracted information as entries in a source database 162. In this way, the source collection module 142 compiles web traffic data (i.e., builds up a history) and stores the historical data in the source database 162], and
[¶¶85-87, Returning now to the behavioral analysis module 146, for each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 142 queries the source database 162 to retrieve the source identifiers for the particular source that has made the suspicious HTTP request. If the source is not in the source database 162, the behavioral analysis module 146 updates the source database 162 by inserting the source (and the relevant source identifiers and attributes) into the source database 162. Note that initially (i.e., upon initial execution of the functionality of the system 130) the suspicious source will be absent from the source database 162. This is due to the fact that the behavioral analysis module 146 and the source collection module 142 collect information over time as client computers make HTTP requests. Once the behavioral analysis module 146 and the source collection module 142 have collected a sufficient amount of historical data, the source database 162 will return the source identifiers (to the behavioral analysis module 146) for the particular source that has made the suspicious HTTP request. Once the behavioral analysis module 146 retrieves the relevant information from the source database 162, the behavioral analysis module 146 performs attribute comparisons to calculate non-binary attribute scores and non-binary source identifier scores. Specifically, for each source identifier, the behavioral analysis module 146 compares each attribute to like attributes of legitimate sources (from the source database 162) to calculate attribute scores. The behavioral analysis module 146 then combines all of the attribute scores, for the particular source identifier, to calculate a non-binary source identifier score. The behavioral analysis module 146 then combines all of the source identifier score to calculate a non-binary reputation score for the particular HTTP request. The aforementioned source identifier score and reputation score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm], and
[¶15, The behavioral analysis module 146 may also generate DoS confidence scores for HTTP requests and source IP addresses associated web applications that are marked (i.e., flagged) as being under DoS attack by the Layer7 DoS protection module 174. The Layer7 DoS protection module 174 may take one or more protective actions against HTTP requests and source IP addresses that correspond to DoS confidence scores indicative of DoS attack. Such protective actions include, but are not limited to, blocking such HTTP requests and/or blacklisting future HTTP requests from that source IP address for a predefined duration (e.g., 5 minutes)], and
and a number of product page views.
[¶80, The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users.
Regarding claim 7, Bar Noy discloses, wherein identifying the third set of bots in the activity data by identifying outlier activity comprises [¶¶78, 85-87], and
providing at least some of the activity data to an outlier detection model; and prior to providing the at least some of the activity data to the outlier detection model, filtering out users that only visited a single page of the website.
[¶80, The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users], and
[¶¶87-88, Once the behavioral analysis module 146 retrieves the relevant information from the source database 162, the behavioral analysis module 146 performs attribute comparisons to calculate non-binary attribute scores and non-binary source identifier scores. Specifically, for each source identifier, the behavioral analysis module 146 compares each attribute to like attributes of legitimate sources (from the source database 162) to calculate attribute scores. The behavioral analysis module 146 then combines all of the attribute scores, for the particular source identifier, to calculate a non-binary source identifier score. The behavioral analysis module 146 then combines all of the source identifier score to calculate a non-binary reputation score for the particular HTTP request. The aforementioned source identifier score and reputation score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm. If the reputation score is above a threshold value, the behavioral analysis module 146 classifies the HTTP request as a legitimate request, and marks (i.e., flags) the HTTP request as a false-positive (referred to hereinafter as “FP”). If the reputation score is below the threshold value, the behavioral analysis module 146 maintains the classification of the HTTP request as a suspicious request, and marks (i.e., flags) the HTTP request as a true-positive (referred to hereinafter as “TP”)].
Regarding claim 8, Bar Noy discloses, wherein the activity data includes a plurality of users that visited the website during a previous day.
[¶74, Initially, the indicator database 160 is populated with scores received from an indicator collection module (not shown), which analyzes suspicious web traffic, for example historical suspicious web traffic, to compile a list of common indicators and associated baseline scores, which may be assigned to the indicators based on a scoring distribution algorithm], and
[¶78, In certain embodiments, once the detection module 140 marks (i.e., flags) the HTTP request as suspicious, the behavioral analysis module 146 begins executing functionality. For each HTTP request marked (i.e., flagged) as suspicious by the detection module 140, the behavioral analysis module 146 analyzes the HTTP request to make determinations about the HTTP request based on the behavior of the source (i.e., the client device 100) that sent the suspicious HTTP request. The analysis is performed in cooperation with the source collection module 142, which collects information about sources (i.e., client computers) that send HTTP requests. In particular, the source collection module 142 analyzes incoming HTTP requests to extract information pertaining to the sources sending the requests, and subsequently stores the extracted information as entries in a source database 162. In this way, the source collection module 142 compiles web traffic data (i.e., builds up a history) and stores the historical data in the source database 162], and
[¶80, The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users], and
[¶111, It is important to note that many of the aforementioned modules influence each other to update respective scores, and that the scores evolve based on the past operation (i.e., history) of the system 130. Furthermore, certain modules rely on data that is built up over time and stored in one or more databases of the system 130 and may not be able to produce valid scores during the initial execution period of the system 130. For example, the behavioral analysis module 146 relies on historical data in the source database 162 in order to produce reputation scores. However, during the initial execution period of the system 130, the source database 162 may be empty or not contain enough data records (i.e., source collection module 142 may not yet have collected source information) to allow the behavioral analysis module 146 to calculate reputation scores. For example, for the first HTTP request received by the system 130, the source database 162 will be empty and therefore the behavioral analysis module 146 will not have any past knowledge], and [¶133].
Regarding claim 9, Bar Noy discloses , further comprising, generating a visualization, the visualization displaying: at least some of the activity data; an indication, for at least some users of a plurality of users in the activity data, whether the user belongs to the first set of bots, the second set of bots, or the third set of bots; and for at least some bots of the third set of bots, a bot confidence score generated by an outlier detection model.
[¶¶130-131, Returning now to FIG. 3, the system 130 further includes (or is linked to) a display device 170 (e.g., display monitor. LCD screen, etc.) and input devices 172 (e.g., keyboard, mouse, etc.). The management module 156 is linked to the input devices 172. The management module 156 may, for example, include a user input interface or a graphical user interface (GUI) that is executed by the CPU 132 and displayed by the display device 170. The input interface enables a user (i.e., a system administrator) to input/adjust parameters and settings of the system 130 via the input devices 172. For example, the user may set or adjust the settings for each of the web applications hosted by the web server 120, including the composite threshold criterion. e.g., threshold score (used in block 474) against which the composite threshold score is evaluated. The management module 156 is also linked to the display device 170, which enables the system 130 to provide display and user interface capabilities. The management module 156 collects/aggregates data from the modules of the system 130 and processes that data to produce output data for viewing by a system administrator of the system 130 via the display device 170. The output data may include qualitative and quantitative data and information, including, for example, the total number of HTTP requests received by the system 130, the number of times protective actions were taken against HTTP requests, statistical information pertaining to the HTTP requests (e.g., average number of FP, average number of TP, etc.), the scores provided by the various modules, attack IDs (either all of the attack IDs or a snippet of the attack IDs from the attack database 164) and relevant data pertaining to each displayed attack ID (e.g., URL attacked, web server/application attacked, source IP address, etc.), data logs, and the number of bot attacks detected by the bot detection module 154].
Regarding claim 11, Bar Noy discloses, wherein applying the one or more rules for identifying bots comprises: identifying a plurality of users associated with an IP address or a user agent string ; determining a demand for the plurality of users; based on the visits to visitor’s ratio and based on the demand, determining that all users associated with the IP address, or the user agent string are bots.
[¶¶79-81, The source collection module 142 initially retrieves data from particular fields of the HTTP requests in order to extract source identifiers (that correspond to the particular fields) from the source information associated with each HTTP request. Examples of such source identifiers include, but are not limited to, the IP address of the source, the User-Agent, client computer fingerprint, session ID, and combinations thereof. For each particular source identifier, the source collection module 142 assigns attributes to the source identifier which contain data and information related to the behavior of the source. The attributes include, for example, site coverage, request ratio, vulnerable URL request activity, and response time anomalies (e.g., the elapsed time between receiving an HTTP request and sending an HTTP response message). The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users. The request ratio attribute contains data and information that includes the ratio between the number of legitimate HTTP requests performed by a specific source and the number of malicious HTTP requests performed by the same source. As should be apparent, the higher the ratio, the greater the confidence that the source is a legitimate source. Note that the inverse ratio may also be stored or calculated. In such configurations, the smaller the ratio, the greater the confidence that the source is a legitimate source], and
[¶98, In order to better explain the operation of the correlation module 148, the following case is considered as an example. In the example, the suspicious HTITP request has two attributes and the attack database 164 stores four attacks. The attributes of the suspicious HTTP request are: a) the list of potential attack indicators and, b) the IP address of the source that sent the suspicious HTTP request. The four attacks in the attack database 164 are denoted by Attack ID.sub.1, Attack ID.sub.2, Attack ID.sub.3, and Attack ID.sub.4. Attack ID.sub.1 has assigned to it the following attributes: a) one or more potential attack indicators common to the indicators in the list of potential attack indicators in the suspicious HTTP request, and b) an IP address that is different from the IP address of the source that sent the suspicious HTTP request. Attack ID.sub.2 has assigned to it the following attribute: User-Agent that is different from the User-Agent of the suspicious HTTP request. Attack ID.sub.3 has assigned to it the following attribute: User-Agent that is different from the User-Agent of the suspicious HTTP request. Attack ID.sub.4 has assigned to it the following attribute: one or more potential attack indicators common to the indicators in the list of potential attack indicators in the suspicious HTTP request. In such an example case, in response to the query related to the list of potential attack indicators, the attack database 164 will return a first list having the attacks denoted by Attack ID.sub.1 and Attack ID.sub.4. In response to the query related to the IP address of the source that sent the suspicious HTTP request, the attack database 164 will return a second list having the attack denoted by Attack ID.sub.1], and [ ¶¶92, 11, 131, 151].
Regarding claim 21, this claim is interpreted and rejected for the same rational set forth in claim 1.
Regarding claim 22, Bar Noy discloses wherein identifying the third set of bots in the activity data by identifying outlier activity comprises: identifying a plurality of features; inputting the activity data into a first model to generate a first score for each user of a plurality of users, wherein the first model determines center values for the plurality of features and generates the first score for each user based on distances of feature values for the user from the center values for the plurality of features: inputting the activity data into a second model to generate a second score for each user of the plurality of users, wherein the second model clusters the plurality of users and generates the second score for each user based on a distance of the user from a center of a cluster to which the user is assigned;
for each user of the plurality of users, aggregating the first score and the second score for the user to determine whether the user is an outlier; and for each user of the plurality of users, in response to determining that the user is an outlier, classifying the user as a bot.
[¶5, The present invention is directed to computerized methods and systems which provide security for web applications. The computerized methods and systems calculate weighted scores based on context-driven inspection of communication protocol request messages (e.g., HTTP requests) and behavioral models of the sources that send the request messages], and
[¶40, Initially, throughout this document, references are made to module output scores that are generated/calculated by modules as a result of analysis of data, by said modules. These scores can generally be characterized as likelihood or confidence scores, which define the probability (and interchangeably the likelihood or confidence) that the data being analyzed falls within a specific category. To determine the category, the output scores are evaluated against a threshold criterion, which in certain embodiments is a threshold value. Data that produces output scores that are above (or equal to) the threshold value are placed in a first category, and data that produces output score which are less than the threshold value are placed in a second category. In such embodiments, the higher the output score the higher the probability that the data is be placed in the first category (and the lower the probability that the data is be placed in the second category). Conversely, the lower the output score the higher the probability that the data is be placed in the second category (and the lower the probability that the data is be placed in the first category)], and [¶¶41-42, 60, 66, 87-90, 92, 95-97, 129-130].
Regarding claim 23, Bar Noy discloses wherein inputting the activity data into the first model to generate the first score for each user of the plurality of users comprises, for each user of the plurality of users: for each feature of the plurality of features, flagging the feature in response to determining that a feature value for the user is greater than a range from a center value for the feature; and generating the first score based at least in part on a number of flagged features.
[¶¶65-66, The potential attack indicators are evaluated against a decision criterion, which may be set by a system administrator that manages the system 130, to determine whether a criterion is satisfied. In certain embodiments, the decision criterion is a threshold value. In such embodiments, the detection module 140 tallies the number of potential attack indicators in the HTTP request and evaluates the number of potential attack indicators against the threshold value. If the potential attack indicators satisfy the decision criterion (e.g., if number of potential attack indicators exceeds the threshold value), the detection module 140 marks (i.e., flags) the HTTP request as suspicious. If the potential attack indicators dissatisfy (i.e., do not satisfy) the decision criterion (e.g., if the number of potential attack indicators is less than or equal to the threshold value), the HTTP request is deemed as a legitimate request and is released through to the web application 121 by the system 130], and
[¶84, The source collection module 142 collects all of the source identifiers and associated attributes and stores the source identifiers and attributes in the source database 162. As a result, the source database 162 stores information about all of the sources that have made HTTP requests, wherein for each source associated with a particular HTTP request, each source identifier of that source is stored in the source database 162 together with all of the attributes associated with each respective source identifier. The source entries in the source database 162 can also be marked (i.e., flagged) with a legitimacy score. For example, sources that have made only legitimate HTTP requests can be marked (i.e., flagged) with a high score in the source database 162, thereby indicating that such sources are legitimate sources], and [¶¶78, 88, 103, 125]
Regarding claim 25, this claim is interpreted and rejected for the same rational set forth in claim 1.
Regarding claim 26, Bar Noy discloses wherein the instructions, when executed by the one or more processors, further cause the system to display a visualization identifying the first set of bots, the second set of bots, and the third set of bots
[¶¶130-131, Returning now to FIG. 3, the system 130 further includes (or is linked to) a display device 170 (e.g., display monitor. LCD screen, etc.) and input devices 172 (e.g., keyboard, mouse, etc.). The management module 156 is linked to the input devices 172. The management module 156 may, for example, include a user input interface or a graphical user interface (GUI) that is executed by the CPU 132 and displayed by the display device 170. The input interface enables a user (i.e., a system administrator) to input/adjust parameters and settings of the system 130 via the input devices 172. For example, the user may set or adjust the settings for each of the web applications hosted by the web server 120, including the composite threshold criterion. e.g., threshold score (used in block 474) against which the composite threshold score is evaluated. The management module 156 is also linked to the display device 170, which enables the system 130 to provide display and user interface capabilities. The management module 156 collects/aggregates data from the modules of the system 130 and processes that data to produce output data for viewing by a system administrator of the system 130 via the display device 170. The output data may include qualitative and quantitative data and information, including, for example, the total number of HTTP requests received by the system 130, the number of times protective actions were taken against HTTP requests, statistical information pertaining to the HTTP requests (e.g., average number of FP, average number of TP, etc.), the scores provided by the various modules, attack IDs (either all of the attack IDs or a snippet of the attack IDs from the attack database 164) and relevant data pertaining to each displayed attack ID (e.g., URL attacked, web server/application attacked, source IP address, etc.), data logs, and the number of bot attacks detected by the bot detection module 154.
Regarding claim 27, Bar Noy discloses wherein the website is a retail website.
[¶80, The site coverage attribute contains data and information that includes the percentage (or fraction or ratio) of common URLs visited by a specific source when browsing the common URLs. Without loss of generality, a “common URL” is defined to be a URL that is typically visited by legitimate users. Examples of common URLs include, http://www.google.com, http://www.facebook.com, http://www.ebay.com, http://www.youtube.com, http://en.wikipedia.org, and http://www.amazon.com. Typically, legitimate users visit various portions/sections of the URL in the web application 121 when browsing the common URL. For example, legitimate users will typically navigate to various sections of the URL by clicking on various hyperlinks throughout the URL page. Legitimate users may also often input user specific information. e.g., user login information, in particular when accessing e-mail, banking, and social media URLs. In contrast, malicious users (i.e., attackers) typically do not navigate URLs with the same prevalence as legitimate users.
Regarding claim 28, Bar Noy wherein the system further comprises a bot detector comprising the one or more processors and the memory.
[ see fig.3 , Bot detection module (154).
Regarding claim 29, Bar Noy discloses wherein identifying the third set of bots in the activity data by identifying outlier activity comprises, for a user associated with the activity data, generating a first score and a second score, generating a weighted aggregation of the first score and the second score, and comparing the weighted aggregation of the first score and the second score to a predetermined threshold.
[¶5, The present invention is directed to computerized methods and systems which provide security for web applications. The computerized methods and systems calculate weighted scores based on context-driven inspection of communication protocol request messages (e.g., HTTP requests) and behavioral models of the sources that send the request messages], and
[¶40, Initially, throughout this document, references are made to module output scores that are generated/calculated by modules as a result of analysis of data, by said modules. These scores can generally be characterized as likelihood or confidence scores, which define the probability (and interchangeably the likelihood or confidence) that the data being analyzed falls within a specific category. To determine the category, the output scores are evaluated against a threshold criterion, which in certain embodiments is a threshold value. Data that produces output scores that are above (or equal to) the threshold value are placed in a first category, and data that produces output score which are less than the threshold value are placed in a second category. In such embodiments, the higher the output score the higher the probability that the data is be placed in the first category (and the lower the probability that the data is be placed in the second category). Conversely, the lower the output score the higher the probability that the data is be placed in the second category (and the lower the probability that the data is be placed in the first category)], and
[¶76, If a particular combination is absent from the indicator database 160, the database is updated to include the particular combination. The indicator analysis module 144 then retrieves the individual scores associated with the potential attack indicators in the new combination from the indicator database 160. The indicator analysis module 144 then generates a combined non-binary score for the new combination based on the retrieved individual scores and stores the combined score in the indicator database 160. The combined score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm.], and
[¶87, Once the behavioral analysis module 146 retrieves the relevant information from the source database 162, the behavioral analysis module 146 performs attribute comparisons to calculate non-binary attribute scores and non-binary source identifier scores. Specifically, for each source identifier, the behavioral analysis module 146 compares each attribute to like attributes of legitimate sources (from the source database 162) to calculate attribute scores. The behavioral analysis module 146 then combines all of the attribute scores, for the particular source identifier, to calculate a non-binary source identifier score. The behavioral analysis module 146 then combines all of the source identifier score to calculate a non-binary reputation score for the particular HTTP request. The aforementioned source identifier score and reputation score may be generated according to a specialized algorithm, for example, a weighted averaging algorithm], and [¶¶41-42, 60, 66, 88-90, 92, 95-97, 129-130].
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, 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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
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 of this title, 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.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over US Patent No. 2019/0334940 issue to Bar Noy , and in view of US Patent Application No. (023/0100508) issued to Abobaker.
Regarding claim 10, Bar Noy discloses wherein identifying the self-identified bots comprises: identifying one or more keywords in user agent strings for users in the activity data [¶64, Potential attack indicators that are to be identified by the detection module 140 may be, for example, any keyword or keyword strings in text or objects in the scanned payloads that are anomalous or suspicious in nature. Examples of potential attack indicators include, but are not limited to, keywords that are indicative of SQL injection attempts and probing attempts via netcat utility to bypass authentication or encryption (i.e., backdoor entry). For example, the keyword string ‘105 or 1=1’ is indicative of an SQL injection (in the HTTP request) to attempt to gain access to restricted information (for example usernames and/or passwords from a database or table that contains usernames and/or passwords). The following is a list of non-limiting examples of sample portions of suspicious HTTP requests and the potential attack indicators (keyword/keyword strings) associated with each sample: see table 1].
Bar Noy does not explicitly disclose , however, Abobaker discloses applying a machine learning model to the user agent strings.
[¶5, Techniques disclosed herein relate generally to text classification. More specifically and without limitation, techniques disclosed herein relate to techniques for fusing word embeddings with word scores for text classification (e.g., in natural language processing). A machine learning approach to text classification may include presenting the text to a machine learning model as one or more low-dimensional continuous feature vectors (e.g., word embeddings)], and
[¶119, A machine learning approach to text classification may include presenting the text to a machine learning model as one or more feature vectors (e.g., word embeddings). The machine learning model may be complex: for example, a long short-term memory (LS™) or a convolutional neural network (CNN). We have found, however, that using a simpler machine learning model may be sufficient. A deep average network or “DAN” (also called a “deep averaging network”), for example, is a simple machine learning model that may be highly performing for the task of text classification].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teaching of Bar Noy by incorporating “a machine learning model”, as taught by Abobaker. One could have been motivated to do so in order for implementing a machine learning model to text classification and include techniques for fusing word embeddings with word scores for text classification [ Abobaker, Abstract, ¶¶5, 119].
Allowable Subject Matter
Claims 5, and 24 are objected to as being dependent upon a rejected base claim but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The reason for allowance will be furnished upon allowance of the application.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
See 892 for more relevant references.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHAHRIAR ZARRINEH whose telephone number is (571)272-1207. The examiner can normally be reached Monday-Friday, 8:30am-5:30pm.
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, Jorge Ortiz-Criado can be reached at 571-272-7624. 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.
/SHAHRIAR ZARRINEH/Primary Examiner, Art Unit 2496