Abstract: Data stream based event sequence anomaly detection for mobility customer fraud analysis is presented herein. A system obtains a sequence of events comprising respective modalities of communication that correspond to a subscriber identity associated with a communication service—the sequence of events having occurred within a defined period. Based on defined classifiers representing respective fraudulent sequences of events, the system determines, via a group of machine learning models corresponding to respective machine learning processes, whether the sequence of events satisfies a defined condition with respect to likelihood of representing a fraudulent sequence of events of the respective fraudulent sequences of events. In response to the sequence of events being determined to satisfy the defined condition, the system sends, via a user interface of the system, a notification indicating that the sequence of events has been determined to represent the fraudulent sequence of events.

Abstract: A processing system may determine a plurality of input features of a first machine learning model that is deployed in a telecommunication network for a prediction task associated with an operation of the telecommunication network and apply a time series forecast model to a historical data set of a first data source associated with at least one of the plurality of input features to generate a forecast upper bound of a first characteristic of the first data source for a first time period and a forecast lower bound of the first characteristic of the first data source for the first time period. The processing system may then detect that the first characteristic exceeds one of the forecast upper bound or the forecast lower bound during the first time period and generate an alert that an output of the first machine learning model may be faulty, in response to the detecting.

Abstract: An example method includes receiving data to be provided to an application using a scoring model for calculating a score, determining that the data is incompatible with a current feature set of the scoring model applied by the application, receiving a next best model of features in response to the determining that the data is incompatible with the current feature set, executing the application to calculate the score with the data and the features of the next best model, and generating an output in accordance with the score.

Abstract: The described technology is generally directed towards a machine learning feature recommender, for use in connection with a feature store. By collecting data and recommending machine learning features to users based on collected data, embodiments can facilitate data scientists' discovery of features that have been used by their colleagues and that are likely to make their machine learning models more performant. The disclosed machine learning feature recommender can reduce the effort involved in developing machine learning models.

Abstract: Aspects of the subject disclosure may include, for example, receiving input data via a transformation UI, generating transformation configuration data, causing the transformation UI to present transformation object data per the transformation configuration data, where the transformation object data identifies data objects each including an input and output field name and a data type, detecting, from the transformation UI, an instruction defining a mapping for the input data, including a modification to the output field name of a data object such that the input field name of the data object is mapped to the modified output field name, based on the detecting the instruction, modifying the first transformation configuration data per the mapping to derive second transformation configuration data, performing a transformation of the input data based on the second transformation configuration data, and causing the transformation UI to present a transformation output. Other embodiments are disclosed.

Abstract: In another example, a device includes a processor and a computer-readable medium storing instructions which, when executed by the processor, cause the processor to perform operations. The operations include acquiring a plurality of data items from a plurality of data sources, wherein the at least two data sources data sources of the plurality of data sources are maintained by different entities, normalizing attributes of the plurality of data items, using a first machine learning technique, matching at least two data items of the plurality of data items to form a grouping, wherein the matching is based on similarities observed in the attributes of the at least two data items subsequent to the normalizing, and creating a single profile for an individual associated with the at least two data items, based on the grouping, wherein the single profile consolidates the attributes of the at least two data items.

Abstract: Disclosed are systems, methods and computer-readable media for identifying failed points in a network in real time. The system and method employ a topology database against which parsed and enhanced fault notifications are compared to identify the location of the fault notifications. The fault notifications are associated into a single event. A root cause analysis module having machine learning capabilities is used to match the single event with a predicted root cause by accessing a root cause database established with existing historic data and heuristically derived failure scenarios.

Abstract: Techniques are provided for assisting users to share specific locations within videos. Specifically, controls are provided to enable a viewer of a video to drop a “video pin” on a location within a video. In response to the user dropping a video pin at a particular location in the video, a “video pin record” that indicates the selected location is automatically generated by a video pin application. The video pins may be used to identify specific time points in the video and/or specific time segments of the video. Video pin records may be shared with other users to allow the other users to immediately jump to the locations, within the video, at which the corresponding video pins were dropped.

Abstract: One or more simulations are generated for in-home monitoring. The simulations model sensory detection of a user's physical activities using a different number and/or a different combination of sensors. Each different simulation may thus be associated with an accuracy and a cost, depending on the number and/or combination of sensors. The simulations thus present a range of sensory configurations that balance accuracy and affordability, from which an optimum sensory solution may be determined for the in-home monitoring.

Abstract: Techniques are provided for assisting users to share specific locations within videos. Specifically, controls are provided to enable a viewer of a video to drop a “video pin” on a location within a video. In response to the user dropping a video pin at a particular location in the video, a “video pin record” that indicates the selected location is automatically generated by a video pin application. The video pins may be used to identify specific time points in the video and/or specific time segments of the video. Video pin records may be shared with other users to allow the other users to immediately jump to the locations, within the video, at which the corresponding video pins were dropped.