A technique for tracking web browsing activity of a client device that includes storing, in a memory, a client profile having a client identifier associated therewith, providing a client device with a cache file having the client identifier embedded therein, receiving from the client device an identification of a client action and the client identifier, and updating the client profile to include the identification of the client action.

The present disclosure is directed to a technique to reduce latency in feedback-based system performance determination. A system receives, from an application developer device, indications of an in-application event and a first input value for an application content delivery profile. The system receives, via an interface from an application developed by an application developer and executed by a computing device remote from the data processing system and different from the application developer device, a ping indicative of an occurrence of the in-application event on the computing device. The system merges data from the ping with internal data determined by the data processing system to generate merged data. The system determines a predicted performance for the in-application event and provides an indication of the predicted performance. The system configures, responsive to the indication of the predicted performance, the application content delivery profile with a second input value.

Systems and methods for generating, administering and analyzing a user experience study are provided. In particular, intents can be generated from a user experience study by applying one or more screener questions to participants and subjecting the screened participants to one or more tasks. Corresponding clickstreams and success data for each participant engaging in the tasks can be recorded. The success and clickstream data can also be aggregated for all the screened participants as aggregated results. Video data including audio for each of the screened participants can also be recorded.

Techniques and solutions are described for improving user interfaces, such as by analyzing user interactions with a user interface with a machine learning component. The machine learning component can be trained with user interaction data that includes an interaction identifier and a timestamp. The identifiers and timestamps can be used to determine the duration of an interaction with a user interface element, as well as patterns of interactions. Training data can be used to establish baseline or threshold values or ranges for particular user interface elements or types of user interface elements. Test data can be obtained that includes identifiers and timestamps. The time taken to complete an interaction with a user interface element, and optionally an interaction pattern, can be analyzed. If the machine learning component determines that an interaction time or pattern is abnormal, various actions can be taken, such as providing a report or user interface guidance.

A method and a system for arranging a user multi-session from a plurality of user sessions, where the sessions are received from a plurality of computerized client devices communicatively coupled via a communication network to at least one content server. At least some of the client devices may be operated by a same user, and the data content may include at least part of data communicated between any client device and any content server. The method including dividing the received data content into a plurality of sessions, where at least two sessions are associated with the same user, selecting at least two sessions received from at least two respective client devices associated with the same user, and associating the selected at least two sessions to form a multi-session.

The present disclosure is directed to a scalable, extensible, fault-tolerant system for stateful joining of two or more streams that are not fully synchronized, event ordering is not guaranteed, and certain events arrive a bit late. The system can ensure to combine the events or link the data in near real-time with low latency to mitigate impacts on downstream applications, such as ML models for determining suspicious behavior. Apart from combining events, the system can ensure to propagate the needed entities to other product streams or help in entity resolution. If any of the needed data is yet to arrive, a user can configure a few parameters to achieve desired eventual and attribute consistency. The architecture is designed to be agnostic of stream processing framework and can work well with both streaming and batch paths.

A system and method may analyze computer actions on a computer desktop system. Using a data gathering process, a low-level user action information item, describing input by a user (e.g. to the computer desktop system), may be received or gathered. The low-level user action information item may include an input type description and screen window information. Based on a series of low-level user action information items, a process a computer is engaging in with the user may be estimated or determined. The best or most appropriate next low-level user action may be displayed or suggested to the user, e.g. on a computer desktop system to a user.

Systems and methods for conducting a test on a third-party testing platform are provided. A networked system causes presentation of a setup user interface to a third-party user, whereby the setup user interface includes a field for indicating an attribute of a publication to be tested. The networked system receives, via the setup user interface, an indication of the attribute, a subject to be tested, and one or more test parameters. The networked system applies the attribute change to a first version of the publication to generate a second version of the publication. The first version is presented to a first subset of potential users and the second version is presented to a second subset of potential users. Interactions with both the first version and the second version are monitored and analyzed to determine results of the test. The results are then presented to the third-party user.

A computer-implemented method to identify unauthorized use of a device based on a usage pattern. The method includes tracking usage of a device, wherein the usage includes activity by a user interacting with the device. The method includes identifying a usage pattern, wherein the usage pattern is based on usage data. The method further includes generating, based on the usage pattern, a heatmap. The method includes predicting future usage of the device by the user, wherein the predicting includes generating a Markov chain of the predicted future usage. The method also includes determining actual usage is different than the predicted usage. The method further includes calculating, in response to determining the actual usage is different than the predicted future usage, a difference score. The method includes determining the difference score is above a difference threshold, and activating, in response to the difference score being above the difference threshold, an alert.

A graph model of a graphical user interface (GUI) may be generated by processing usage data of the GUI where the usage data comprises sequences of GUI pages and actions between GUI pages. The nodes of the graph model may be determined by obtaining GUI pages from the usage data, identifying dynamic GUI elements in the GUI pages, generating canonical GUI pages by modifying the GUI pages using the dynamic GUI elements, and creating graph nodes using the canonical GUI pages. The edges of the graph may be determined by processing actions from the GUI data that were performed by users to transition from one GUI page to another GUI page. The graph model of the GUI may be used for any appropriate application, such as determining statistics relating to the GUI or statistics relating to individual users of the GUI.