Abstract: Concepts and technologies disclosed herein are directed to optimizing and reducing redundant dispatch tickets via a network knowledge graph. According to one aspect disclosed herein, a network knowledge graph generation system (“NKGGS”) can construct a machine learning model to determine a probability of an installation job needing a helper job to fulfill a service order. The NKGGS can execute the machine learning model to determine the probability. The machine learning model can determine the probability of the installation job needing the helper job to fulfill the service order based upon a network knowledge graph and a dependency score. The NKGGS can cluster the installation job with a plurality of installation jobs.

Abstract: Historical time-series data can be analyzed using a probabilistic model to determine one or more distributions, including at least a normal distribution and an anomaly distribution. These distributions can be analyzed to obtain values for distribution parameters, such as mean, standard deviation, and density, as well as other statistical parameters, for use in building a forecasting model. This model can analyze the time-series data to predict or forecast actionable anomalies at one or more future points or periods in time, such as may exceed a determined anomaly threshold with at least a minimum amount of confidence. A determination can be made as to one or more actions to take in anticipation of the anomalous event, or volume of events, such as to attempt to prevent the occurrence or to be better positioned to handle the occurrence. Such forecasting or prediction can utilize both modeling and feature engineering.

Abstract: Aspects of the subject disclosure may include, for example, receiving a plurality of proposed machine learning solutions to a machine learning problem including receiving, for each respective proposed machine learning solution of the plurality of proposed machine learning solutions, one or more of a machine learning model, a dataset and a data pipeline output; automatically determining hybrid solutions to the machine learning problem, including combining, by the processing system, at least one of a first component from a first proposed machine learning solution with at least one of a second component from a second proposed machine learning solution; and ranking the hybrid solutions including determining a log loss score for each hybrid solution and sorting the hybrid solutions according to the log loss score for each hybrid solution. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, receiving a plurality of proposed machine learning solutions to a machine learning problem including receiving, for each respective proposed machine learning solution of the plurality of proposed machine learning solutions, one or more of a machine learning model, a dataset and a data pipeline output; automatically determining hybrid solutions to the machine learning problem, including combining, by the processing system, at least one of a first component from a first proposed machine learning solution with at least one of a second component from a second proposed machine learning solution; and ranking the hybrid solutions including determining a log loss score for each hybrid solution and sorting the hybrid solutions according to the log loss score for each hybrid solution. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, receiving a plurality of proposed machine learning solutions to a machine learning problem including receiving, for each respective proposed machine learning solution of the plurality of proposed machine learning solutions, one or more of a machine learning model, a dataset and a data pipeline output; automatically determining hybrid solutions to the machine learning problem, including combining, by the processing system, at least one of a first component from a first proposed machine learning solution with at least one of a second component from a second proposed machine learning solution; and ranking the hybrid solutions including determining a log loss score for each hybrid solution and sorting the hybrid solutions according to the log loss score for each hybrid solution. Other embodiments are disclosed.

Abstract: A processing system including at least one processor may collect a first set of time series features relating to requests for a content item at a content distribution node in a communication network, generate a first prediction model based upon the first set of time series features to predict levels of demand for the content item at the content distribution node at future time periods, identify, via the first prediction model, a first time period of the future time periods when a predicted level of demand for the content item exceeds a threshold level of demand, identify a second time period of the future time periods when a predicted level of utilization of the communication network is below a threshold level of utilization, the second time period being prior to the first time period, and transfer the content item to the content distribution node in the second time period.

Abstract: Techniques for processing of electronic documents comprising tables to desirably extract and/or recreate tables, including information in the tables, are presented. A document processing management component (DPMC) can perform a multi-stage process to extract a table from a document and recreate the table, including the table structure and information, in an editable form. During first stage, DPMC can identify candidate cells of the table based on analysis of the document, including identifying border lines that can represent cell borders, identifying any free floating candidate cells, and identifying characters of the candidate cells. During second stage, DPMC can determine structural relationships between respective candidate cells and respective neighbor candidate cells in all directions, based on applicable rules, and record the respective associations between those candidate cells.

Abstract: A processing system including at least one processor may collect a first set of time series features relating to requests for a content item at a content distribution node in a communication network, generate a first prediction model based upon the first set of time series features to predict levels of demand for the content item at the content distribution node at future time periods, identify, via the first prediction model, a first time period of the future time periods when a predicted level of demand for the content item exceeds a threshold level of demand, identify a second time period of the future time periods when a predicted level of utilization of the communication network is below a threshold level of utilization, the second time period being prior to the first time period, and transfer the content item to the content distribution node in the second time period.

Abstract: Aspects of the subject disclosure may include, for example, training a machine learning model on training data, generating, by the machine learning model, a plurality of prediction data records which each has an associated probability, and promoting prediction data records of the plurality of prediction data records having an associated probability exceeding a threshold. The subject disclosure may further include combining the promoted prediction data records with the training data to form new training data, retraining the machine learning model on the new training data and generating, by the machine learning model, new prediction data records. The subject disclosure may further include identifying a real-time condition based on the new prediction data records, the real-time condition being one that requires prompt attention, and resolving the real-time condition. Other embodiments are disclosed.

Abstract: A processing system including at least one processor may collect a first set of time series features relating to requests for a content item at a content distribution node in a communication network, generate a first prediction model based upon the first set of time series features to predict levels of demand for the content item at the content distribution node at future time periods, identify, via the first prediction model, a first time period of the future time periods when a predicted level of demand for the content item exceeds a threshold level of demand, identify a second time period of the future time periods when a predicted level of utilization of the communication network is below a threshold level of utilization, the second time period being prior to the first time period, and transfer the content item to the content distribution node in the second time period.

Abstract: Concepts and technologies disclosed herein are directed to optimizing and reducing redundant dispatch tickets via a network knowledge graph. According to one aspect disclosed herein, a network knowledge graph generation system (“NKGGS”) can construct a machine learning model to determine a probability of an installation job needing a helper job to fulfill a service order. The NKGGS can execute the machine learning model to determine the probability. The machine learning model can determine the probability of the installation job needing the helper job to fulfill the service order based upon a network knowledge graph and a dependency score. The NKGGS can cluster the installation job with a plurality of installation jobs.

Abstract: A processing system may obtain historical job feature, network plant feature, and calendar feature data associated with customer premises installation jobs of a telecommunication network which include a dispatch of a customer premises technician. A plurality of the customer premises installation jobs may further include a dispatch of a network-based technician.

Abstract: A processing system including at least one processor may collect a first set of time series features relating to requests for a content item at a content distribution node in a communication network, generate a first prediction model based upon the first set of time series features to predict levels of demand for the content item at the content distribution node at future time periods, identify, via the first prediction model, a first time period of the future time periods when a predicted level of demand for the content item exceeds a threshold level of demand, identify a second time period of the future time periods when a predicted level of utilization of the communication network is below a threshold level of utilization, the second time period being prior to the first time period, and transfer the content item to the content distribution node in the second time period.

Abstract: Aspects of the subject disclosure may include, for example, receiving a plurality of proposed machine learning solutions to a machine learning problem including receiving, for each respective proposed machine learning solution of the plurality of proposed machine learning solutions, one or more of a machine learning model, a dataset and a data pipeline output; automatically determining hybrid solutions to the machine learning problem, including combining, by the processing system, at least one of a first component from a first proposed machine learning solution with at least one of a second component from a second proposed machine learning solution; and ranking the hybrid solutions including determining a log loss score for each hybrid solution and sorting the hybrid solutions according to the log loss score for each hybrid solution. Other embodiments are disclosed.

Abstract: A processing system including at least one processor may collect a first set of time series features relating to requests for a content item at a content distribution node in a communication network, generate a first prediction model based upon the first set of time series features to predict levels of demand for the content item at the content distribution node at future time periods, identify, via the first prediction model, a first time period of the future time periods when a predicted level of demand for the content item exceeds a threshold level of demand, identify a second time period of the future time periods when a predicted level of utilization of the communication network is below a threshold level of utilization, the second time period being prior to the first time period, and transfer the content item to the content distribution node in the second time period.

Abstract: A processing system collects data points, wherein each data point indicates a location of user equipment in a telecommunication service provider network at a point in time, generates, for a first data point, a set of features over a plurality of time windows, wherein the set includes time series features, estimates an importance of each feature, wherein the importance indicates an accuracy with which the feature allows for estimation of an unseen location of the user equipment, wherein the unseen location is a location that is not identified in the plurality of data points, selects a threshold number of features from the set, wherein the threshold number of features have a greatest importance relative to all features in the set, generates a plurality of predictions of the unseen location, using the threshold number of features, aggregates the plurality of predictions, and estimates the unseen location based on the aggregating.

Abstract: A processing system may obtain historical job feature, network plant feature, and calendar feature data associated with customer premises installation jobs of a telecommunication network which include a dispatch of a customer premises technician. A plurality of the customer premises installation jobs may further include a dispatch of a network-based technician.

Abstract: A processing system collects data points, wherein each data point indicates a location of user equipment in a telecommunication service provider network at a point in time, generates, for a first data point, a set of features over a plurality of time windows, wherein the set includes time series features, estimates an importance of each feature, wherein the importance indicates an accuracy with which the feature allows for estimation of an unseen location of the user equipment, wherein the unseen location is a location that is not identified in the plurality of data points, selects a threshold number of features from the set, wherein the threshold number of features have a greatest importance relative to all features in the set, generates a plurality of predictions of the unseen location, using the threshold number of features, aggregates the plurality of predictions, and estimates the unseen location based on the aggregating.