Abstract: This disclosure discloses systems, devices, and methods for parallelized data structure processing in context of machine learning and reverse proxy servers.

Abstract: Systems for assisting users in selecting various promotional offers while shopping online are disclosed. The system is configured to receive promotional information being offered by one or more business merchants to a user, determine a navigation to a webpage on a user device by the user from where an item can be purchased, identify one or more attributes associated to the webpage, determine one or more promotions from the promotional information based on the one or more attributes identified, and display on the user device of the user the one or more promotions associated with the item being purchased by the user on the webpage of the user device.

Abstract: Embodiments of systems and methods for fraud review are disclosed. The systems may comprise multi-tiered computing systems which may receive fraud alerts from multiple sources. A computing system in a tier may receive a fraud alert and use one or more fraud risk metrics to determine whether the fraud alert should be escalated. If the computing system determines that the fraud alert should be escalated, the computing system may transmit an escalation message to a higher tier computing system. If the computing system determines that the fraud alert should not be escalated, the computing system may transmit a message to a fraud prevention computing system. In some embodiments, the computing system may determine that the fraud alert is a false positive and transmit a false positive message to the source of the fraud alert such as a lower tier computing system.

Abstract: Disclosed are method and systems to program a server to identify the value of a fund comprising shares of multiple private entities. The server receives transaction data associated with a fund where the transaction data identifies a proportion of shares within the fund associated with each private entity, price per share of each private entity, and other relevant data. The server then executes multiple artificial intelligence models to identify comparable public entities to each private entity. The server then retrieves stock price data for each public entity and calculates a value for each private entity in real time. The server also displays a value of the fund in real time where identification of each private entity is anonymized.

Abstract: Embodiments disclosed herein describe intelligent e-book readers which provide a significant improvement over the conventional e-books that simply render static content. The intelligent e-book readers may customize a rendered e-book based on, for example, the reading level and preferences of the user, the user's social media profile and activity, and current events. Furthermore, the intelligent e-book reader may provide additional augmented reality (AR)/virtual reality (VR) content associated with one or more portions of the rendered e-book. The intelligent e-book reader may also facilitate virtual, real time communication between multiple users and experts. The intelligent e-book reader may also facilitate one or more users to provide feedback and suggestions to authors and future movie-makers. The intelligent e-book reader may automatically determine difficult portions of an e-book based on the virtual communications and/or real time eye-tracking of a user.

Abstract: Systems and methods for validating outputs from a machine learning model is disclosed. The machine learning model and a statistical model are executed to generate electronic documents in response to customer requests. A random sample of electronic documents generated from the machine learning model and the statistical model are then selected. A comparison is performed between the random sample of electronic documents generated from the machine learning model and the statistical model. The performance of the machine learning model is validated based on results of the comparison.

Abstract: Disclosed herein are embodiments of systems, methods, and products comprises an analytic server, which dynamically predicts future events for web users. The analytic server generates prediction models based on historical click-through analytics data received from the web server. The analytic server captures the current event (e.g., the current operation of the web user) on the web page, and determines the next event by predicting the web user behavior using the prediction models on an event-by-event basis. The analytic server also queries the web user data from a database to better understand the web user's intention, and improve the prediction accuracy. The analytic server modifies the HTML code to display the web page to include a graphical user interface comprising the predicted event. Based on the web users' reactions to the predicted event, the analytic server updates the prediction models.

Abstract: Systems and methods for managing a list of huddle board participants are disclosed. The huddle collaboration system includes a huddle management system having an authentication module, a data processing module, a huddle board management module, and a module manager, among other suitable components. The system runs an automatic process to update a list of huddle boards and huddle board participants, which includes the process of adding or eliminating team members from the list of participants of one or more huddle boards and/or modifying a dotted line member's permissions within one or more huddle boards. The huddle board management module enables the automatic update of permissions assigned to a team member in one or more huddle boards, in a faster and more accurate manner; therefore enhancing the productivity of the huddle and leveraging the human and information technology resource of the company.

Abstract: Methods and systems disclosed herein allow data to be transferred from a data source to a target database with little to no offline period or data corruptions. The methods and systems describe a server that generates a temporary data repository having a similar configuration as the target data repository; transmits the set of new data records from the data source to the temporary data repository; identifies dependency relationship attributes among the data records stored within the target data repository; and when the server identifies that a predetermined number of data records and their respective dependent data records are stored within the temporary data records, the server merges the set of data records and the set of new data records. The server also stores a pre/post merger record of data such that the server can revert to a previous version of data or roll forward to another version.

Abstract: Various systems and methods use a value in a data file for a data process, as the data process is scaled up in terms of dataset dimensionality, data volume, data types, data content, data source quantity, and data source speed, while remaining compliant with ACID principles. As such, these technologies provide for sourcing of data from various data sources, where the data includes the data file storing the value. The data is cleansed and fused, which enables a report to be generated. In response to the value in the data file being modified, the data, inclusive of the data file storing the value, is again cleansed and fused based on the value being modified. This processing in-turn enables the report to modified based on the value being modified.

Abstract: Methods and systems described herein are directed at least to a system to generate a database structure with a low-latency key architecture. The system can identify one or more natural keys of at least one dimensional object, the dimensional object corresponding to at least a portion of a database structure, identify at least one dependency of at least one fact object on the dimensional object, the fact object corresponding to the database structure, generate, based on one or more of the natural keys, at least one dimensional identifier associated with the dimensional object, assign, to the dimensional object, the dimensional identifier to the dimensional object, and link, concurrently with the assigning, the fact object to the dimensional object by associating the dimensional identifier with the fact object, to generate the database structure.

Abstract: A method comprising receiving a request from a first user to generate an employee dataset for a second user associated with one or more attributes; generating an instruction configured to be transmitted to a plurality of network nodes associated with a blockchain to receive a latest valid blockchain associated with one or more users; upon receiving the latest valid blockchain, generating a score corresponding to each user based on the data within each user's latest valid blockchain and based on a scoring algorithm received from the computing device associated with the first user; generating a graphical user interface comprising a plurality of fields configured to display data associated the one or more users comprising each user's score; upon displaying, the graphical user interface on the first computing device, receiving a selection of users; and generating, a notification to the selected users comprising identification information regarding the first user.

Abstract: The present systems and methods for customizing sub-applications and dashboards in a digital huddle environment are disclosed. The disclosed system and method may be employed for customizing sub-applications according to predefined options, including text, images and numbers; as well as customizing dashboards according to user requirements for showing goals progress, next meetings notifications, schedule, meetings alerts, and the like, according to the requirements of a particular team working in a digital huddle.

Abstract: Methods and systems for determining relationships between two or more nominally unrelated data sources utilizing a combination of probabilistic modeling and graphical clustering are described. The systems and methods for utilizing probabilistic model functions as a way of determining and judging the likelihood that two records from different systems are related to the same entity.

Abstract: A DI recovery management system generates a plurality of ranked claimant records and recovery scores. A predictive machine learning model inputs disability income claim data and disability income claimant data into an event history model utilizing discrete-time survival analysis in conjunction with a gradient boosting machine learning model. The claim termination event is one of a plurality of preselected recovery events that indicate that a claimant has achieved return to work capacity. Claimant data used in modeling includes diagnosis data representative of workplace disability duration guidelines. The predictive machine learning model is continually trained using updated disability income claims data. The training procedure transforms claimant records extracted from a DI claims database into a longitudinal format that includes multiple person-year records corresponding to each claimant record.

Abstract: Systems and methods for remote huddle collaboration are disclosed. The huddle collaboration system may include a huddle management system which may include an authentication module, a data processing module, and a module manager, among other components. Huddle collaboration system may allow huddle members of a company to fully engage in remote huddle sessions through different kind of client computing devices, in a consistent way regardless of their location. The system may act as an alternative to traditional whiteboards, displaying, collecting, and storing information during an active huddle session, where this information may be available to the users in the form of one or more standard and/or custom sub-applications, according to the user's and/or project's requirements.

Abstract: A system and method for automated body mass index is disclosed. The disclosed method operates within a system architecture including one or more computing devices, one or more servers, and one or more databases. A processor operating within the one or more servers executes one or more algorithms for detecting relevant features associated with a potential client's multimedia information. The method may include calculating feature values, such as abdomen circumference, face width, face height, cheekbone width, jaw width, and neck width, and the like as well as calculating the body mass index of the potential client using one or more regression algorithms. A baseline and updated BMI may be determined, and used for determining a baseline and updated value.

Abstract: This disclosure discloses systems, devices, and methods for parallelized data structure processing in context of machine learning and reverse proxy servers.

Abstract: System and methods for an agent tool that identifies high value targets within a particular area, based on an individual's social media activity are disclosed. A computing device displays a user interface generated by an analysis module, and receives an agent's search criteria. Then, the analysis module scans a plurality of social media sites and selects the profile of individuals according to the search criteria. Next, the analysis module identifies the profile of members who are associated to the profile of each previously selected individual, and compares them with the profile of pleased customers. If both profiles match, then the analysis module has identified a high value target. The system also provides the full profile of each high value target found, when needed. The agent tool identifies high value targets in a simpler, less time-consuming and inexpensive way, while increasing the company's revenue.