Abstract: Assigning privacy ratings to tracking domains in order to increase user awareness of online privacy violations. A list of known tracking domains and raw data from a set of host websites may be received. A prevalence, frequency, and role for each tracking domain may be determined from the raw data. Each tracking domain may be assigned to a prevalence cluster, a frequency cluster, and a role cluster. A rank may be determined for each prevalence cluster, frequency cluster, and role cluster. A domain privacy rating may be assigned to each tracking domain that is based on the ranks of the clusters to which each tracking domain is assigned. A privacy action may be performed based on the domain privacy rating assigned to each tracking domain.

Abstract: Methods and systems for protecting against harm caused by malicious websites are disclosed. Exemplary embodiments of the present disclosure may protect against harm caused by malicious websites by identifying malicious websites more accurately and reliably. In particular, some embodiments of the present disclosure may receive first resource data from a first web page on a website that is accessed by a first user and second resource data from a second web page on the website that is accessed by a second user. This resource data may be correlated and analyzed. Based on this analysis, a determination may be made that the website is malicious and a security action can be performed.

Abstract: Systems and methods for detecting fraudulent e-commerce websites by identifying fake review systems are disclosed. In particular, some embodiments may identify an e-commerce website and download content contained on one or more product web pages of the e-commerce website. These web pages may be analyzed to identify a product review feature that is within the one or more product web pages. Attributes of the product review feature may then be evaluated to determine that the e-commerce website is fraudulent and a security action may be performed to protect consumers from the e-commerce website.

Abstract: Methods for protecting against malicious websites using repetitive data signatures are disclosed. Some embodiments may identify known malicious websites and known safe websites. A first dataset containing data from one or more artifacts within the known malicious websites and a second dataset containing data from the one or more artifacts within the known safe websites may be created. One or more signatures may be identified from the first dataset. A first frequency of signature repetition within the first dataset and a second frequency of signature repetition within the second dataset may be determined. A level of confidence may be determined based on the frequencies. If a rule establishment threshold for confidence is met or exceeded, a rule may be established that websites containing the one or more signatures are malicious. The rule may be applied to identify a new malicious website. A security action may also be performed.

Abstract: Methods and systems for protecting mobile devices from malicious application encounters are disclosed. In one embodiment, a method may include obtaining application data and feature data from a plurality of mobile devices and identifying one or more malicious applications installed thereon. Based on the application data, a user profile for each mobile device may be determined and each mobile device may be assigned to a cluster based on the user profile identified. A security model based on the featured data may be created for a selected cluster and the security model may be applied to a selected mobile device within the selected cluster. Finally, a security action may be performed on the selected mobile device based on an output of the security model.

Abstract: A computer-implemented method for utilizing graph neural networks for scam website detection may include (i) creating a dataset of target websites including unknown websites and known scam websites with corresponding scam categories, (ii) extracting website constructs utilized for executing scam attacks from the dataset of target websites, (iii) building a graph including nodes for associating each of the target websites with one or more of the corresponding scam categories, (iv) grouping, for each of the nodes, the target websites based on sharing a common construct within the website constructs, and (v) performing a security action that trains a graph neural network for identifying the unknown websites as potential scam websites based on a similarity score determined from the website constructs. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Methods and systems for protecting mobile devices from malicious application encounters are disclosed. In one embodiment, a method may include obtaining application data and feature data from a plurality of mobile devices and identifying one or more malicious applications installed thereon. Based on the application data, a user profile for each mobile device may be determined and each mobile device may be assigned to a cluster based on the user profile identified. A security model based on the featured data may be created for a selected cluster and the security model may be applied to a selected mobile device within the selected cluster. Finally, a security action may be performed on the selected mobile device based on an output of the security model.

Abstract: Methods for protecting against malicious websites using repetitive data signatures are disclosed. Some embodiments may identify known malicious websites and known safe websites. A first dataset containing data from one or more artifacts within the known malicious websites and a second dataset containing data from the one or more artifacts within the known safe websites may be created. One or more signatures may be identified from the first dataset. A first frequency of signature repetition within the first dataset and a second frequency of signature repetition within the second dataset may be determined. A level of confidence may be determined based on the frequencies. If a rule establishment threshold for confidence is met or exceeded, a rule may be established that websites containing the one or more signatures are malicious. The rule may be applied to identify a new malicious website. A security action may also be performed.

Abstract: Methods and systems for protecting against harm caused by malicious websites are disclosed. Exemplary embodiments of the present disclosure may protect against harm caused by malicious websites by identifying malicious websites more accurately and reliably. In particular, some embodiments of the present disclosure may receive first resource data from a first web page on a website that is accessed by a first user and second resource data from a second web page on the website that is accessed by a second user. This resource data may be correlated and analyzed. Based on this analysis, a determination may be made that the website is malicious and a security action can be performed.

Abstract: Systems and methods for detecting fraudulent e-commerce websites by identifying fake review systems are disclosed. In particular, some embodiments may identify an e-commerce website and download content contained on one or more product web pages of the e-commerce website. These web pages may be analyzed to identify a product review feature that is within the one or more product web pages. Attributes of the product review feature may then be evaluated to determine that the e-commerce website is fraudulent and a security action may be performed to protect consumers from the e-commerce website.

Abstract: Distinguishing between functional tracking domains and nonfunctional tracking domains on a host web page. In particular, a list of known tracking domains that load content into host web pages may be received. This list of tracking domains may include tracking domains that are functional and tracking domains that are nonfunctional. The tracking domains that are functional may be determined by evaluating various behaviors and characteristics of the tracking domains. Once functional tracking domains have been determined, these functional tracking domains may be allowed, and other tracking domains may be blocked from loading content onto host web pages thereby preserving the functionality of the web pages.

Abstract: Distinguishing between functional tracking domains and nonfunctional tracking domains on a host web page. In particular, a list of known tracking domains that load content into host web pages may be received. This list of tracking domains may include tracking domains that are functional and tracking domains that are nonfunctional. The tracking domains that are functional may be determined by evaluating various behaviors and characteristics of the tracking domains. Once functional tracking domains have been determined, these functional tracking domains may be allowed, and other tracking domains may be blocked from loading content onto host web pages thereby preserving the functionality of the web pages.

Abstract: Assigning privacy ratings to tracking domains in order to increase user awareness of online privacy violations. A list of known tracking domains and raw data from a set of host websites may be received. A prevalence, frequency, and role for each tracking domain may be determined from the raw data. Each tracking domain may be assigned to a prevalence cluster, a frequency cluster, and a role cluster. A rank may be determined for each prevalence cluster, frequency cluster, and role cluster. A domain privacy rating may be assigned to each tracking domain that is based on the ranks of the clusters to which each tracking domain is assigned. A privacy action may be performed based on the domain privacy rating assigned to each tracking domain.

Abstract: The disclosed computer-implemented method for utilizing user profile data to protect against phishing attacks may include (i) detecting a target user profile associated services accessed by a network-based application, (ii) determining identifiers associated with each of the services, (iii) extracting, for each of the identifiers, feature vectors describing exploitable screen elements in the network-based application associated with phishing attacks, (iv) updating, based on the feature vectors, previously extracted feature vectors in a data repository storing additional profiles for other users associated with the services, (v) predicting, utilizing a machine-learning model, phishing attack threats for target profile user based on a similarity with the additional profiles, and (vi) performing a security action that protects against the phishing attack threats. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Systems and methods for identifying and removing a tracking capability from an external domain that performs a tracking activity on a host web page. Tracking capabilities of an external domain may be removed by altering web requests and/or responses to API calls. Once these tracking capabilities of the external domain have been removed, the altered web requests and/or altered responses to API calls may be transmitted to a web browser and/or entity making the API call thereby protecting user privacy while allowing the external domain to interact with the host web page.

Abstract: Mitigating website privacy issues by automatically identifying cookie sharing risks in a cookie ecosystem. In some embodiments, a method may include generating, by the computing device, a cookie tree representing cookies created on a web browser while browsing a website. The cookie tree may include nodes representing organizations. The cookie tree may further include hierarchical edges between the nodes representing creation chains of the nodes. The method may also include determining, by the computing device, a creator, a sender, and a receiver of each of the cookies. The method may further include mitigating, by the computing device, website privacy issues based on the creator, sender, and receiver of each of the cookies.

Abstract: Identifying and removing a tracking capability from an external domain that performs a tracking activity on a host web page. Tracking capabilities of an external domain may be removed by altering web requests and/or responses to API calls. Once these tracking capabilities of the external domain have been removed, the altered web requests and/or altered responses to API calls may be transmitted to a web browser and/or entity making the API call thereby protecting user privacy while allowing the external domain to interact with the host web page.

Abstract: Analyzing and mitigating website privacy issues by automatically classifying cookies.

Abstract: Analyzing and mitigating privacy issues on a computing device using cookie generation flows. The method includes initiating a headless web browser, monitoring a request made of a website accessed by the headless web browser, monitoring scripts created on the website, instrumenting a function used to create a cookie on the computing device, tracing an initial generation of a call used to create the cookie on the computing device, obtaining a cookie generation flow related to the creation of the cookie, and initiating a security action based on obtaining the cookie generation flow.

Abstract: Building a ground truth dataset for a machine learning-based security application. In one embodiment, a method may include identifying a set of network devices to add to a ground truth dataset. The method may also include, for each network device in the set of network devices, identifying a potentially malicious application stored on the network device, analyzing behavior of the potentially malicious application to determine whether the potentially malicious application has behaved maliciously, and if so, adding the network device to the ground truth dataset as an infected device or, if not, adding the network device to the ground truth dataset as a clean device. The method may further include training a machine learning classifier of a security application using the ground truth dataset, making a security action decision using the machine learning classifier, and performing a security action on a computer system based on the security action decision.