Abstract: An anomaly detection and analysis system generates analysis or summary of the anomalies detected from key performance indicators (KPIs). The system receives anomaly data reporting anomalies detected in key performance indicator (KPI) data. The system classifies the reported anomalies into a plurality of anomaly items, wherein anomalies from KPI data that share a set of features are assigned to one anomaly item. The system computes a ranking score for each anomaly item by assigning predefined weights for different anomaly types that are present in the anomaly item. The system sorts a list of anomaly items from the plurality of anomaly items into a sorted list of anomaly items according to the ranking scores computed for the plurality of anomaly items. The system sends the sorted list of anomaly items to a user device for presentation.

Abstract: The disclosed system identifies and resolves telecommunication network problems and/or problems with devices utilizing the network. The system segments the user devices into groups having similar characteristics, by grouping together devices more likely to leave the network, and devices less likely to leave the network. Each group of devices has a trained machine learning model, to predict the NEX score for each device. The system focuses on devices with a low NEX score. The system determines whether the problem is with the network and/or the problem is with the device. The system gathers information from network monitoring software and determine which tower caused, for example, the call to drop. The system determines technological capability of the device from the device TAC number. The system suggests solutions to the network problems by suggesting adding a sector, tuning an antenna, etc., or if the device has problems, adjusting phone settings.

Abstract: A network performance degradation detection system is provided. The system receives Key Performance Indicator (KPI) values for devices of different device types running different software versions. The system determines baseline values of a first device type by averaging the KPI values of the different software versions running on devices of the first device type. The system compares KPI values for a first software version running on devices of the first device type with the determined baseline values to produce a set of comparison results. The system applies the set of comparison results to a classification model to determine whether the first software version running on devices of the first device type causes network performance degradation.

Abstract: The disclosed system identifies international calling performance issues of a wireless telecommunication network. The system receives network traffic data for international calls including information about call attempts to a country. The system categorizes the country into a major category and a minor category based on the call attempts information. For a subset of countries, and for each key performance indicator in a subset of selected key performance indicators, the system monitors performance using an anomaly detection model to identify an anomaly in network performance, determines an actual value of the key performance indicator for the detected anomaly, and computes a variation value of the determined actual value based on a predicted range of values. The system ranks countries using the computed variation values, to indicate problematic parts of the wireless telecommunication network.

Abstract: The disclosed system identifies international calling performance issues of a wireless telecommunication network. The system receives network traffic data for international calls including information about call attempts to a country. The system categorizes the country into a major category and a minor category based on the call attempts information. For a subset of countries, and for each key performance indicator in a subset of selected key performance indicators, the system monitors performance using an anomaly detection model to identify an anomaly in network performance, determines an actual value of the key performance indicator for the detected anomaly, and computes a variation value of the determined actual value based on a predicted range of values. The system ranks countries using the computed variation values, to indicate problematic parts of the wireless telecommunication network.

Abstract: Techniques for identifying certain types of network activity are disclosed, including parsing network traffic to automatically recognize anonymous identifiers. Such techniques may be used to identify and eliminate malicious and/or undesirable network traffic, and to identify topics relevant to a user of a particular network device so that communications to such a user are more likely to relate to a topic of interest to the user.

Abstract: An anomaly detection and analysis system detects anomalies in time series data from key performance indicators (KPIs). The system decomposes samples of the time series data received during a first time interval into a trend component, a seasonality component, and a randomness component. The system identifies an upper bound and a lower bound based on the trend component, the seasonality component, and a variance of the randomness component. The system reports a sample received after the first time interval as an anomaly when the sample exceeds the upper bound or the lower bound. The system recalculates the trend component, the seasonality component, and the randomness component when more than a threshold percentage of the samples of the time series data received during a second time interval are reported as being anomalous.

Abstract: Techniques for identifying certain types of network activity are disclosed, including parsing of a Uniform Resource Locator (URL) to identify a plurality of key-value pairs in a query string of the URL. The plurality of key-value pairs may include one or more potential anonymous identifiers. In an example embodiment, a machine learning algorithm is trained on the URL to determine whether the one or more potential anonymous identifiers are actual anonymous identifiers (i.e., advertising identifiers) that provide advertisers a method to identify a user device without using, for example, a permanent device identifier. In this embodiment, a ranking threshold is used to verify the URL. A verified URL associate the one or more potential anonymous identifiers with the user device as actual anonymous identifiers. Such techniques may be used to identify and eliminate malicious and/or undesirable network traffic.

Abstract: The disclosed system identifies and resolves telecommunication network problems and/or problems with devices utilizing the network. The system segments the user devices into groups having similar characteristics, by grouping together devices more likely to leave the network, and devices less likely to leave the network. Each group of devices has a trained machine learning model, to predict the NEX score for each device. The system focuses on devices with a low NEX score. The system determines whether the problem is with the network and/or the problem is with the device. The system gathers information from network monitoring software and determine which tower caused, for example, the call to drop. The system determines technological capability of the device from the device TAC number. The system suggests solutions to the network problems by suggesting adding a sector, tuning an antenna, etc., or if the device has problems, adjusting phone settings.

Abstract: The disclosed system identifies international calling performance issues of a wireless telecommunication network. The system receives network traffic data for international calls including information about call attempts to a country. The system categorizes the country into a major category and a minor category based on the call attempts information. For a subset of countries, and for each key performance indicator in a subset of selected key performance indicators, the system monitors performance using an anomaly detection model to identify an anomaly in network performance, determines an actual value of the key performance indicator for the detected anomaly, and computes a variation value of the determined actual value based on a predicted range of values. The system ranks countries using the computed variation values, to indicate problematic parts of the wireless telecommunication network.

Abstract: The disclosed system identifies and resolves telecommunication network problems and/or problems with devices utilizing the network. The system segments the user devices into groups having similar characteristics, by grouping together devices more likely to leave the network, and devices less likely to leave the network. Each group of devices has a trained machine learning model, to predict the NEX score for each device. The system focuses on devices with a low NEX score. The system determines whether the problem is with the network and/or the problem is with the device. The system gathers information from network monitoring software and determine which tower caused, for example, the call to drop. The system determines technological capability of the device from the device TAC number. The system suggests solutions to the network problems by suggesting adding a sector, tuning an antenna, etc., or if the device has problems, adjusting phone settings.

Abstract: The disclosed system identifies international calling performance issues of a wireless telecommunication network. The system receives network traffic data for international calls including information about call attempts to a country. The system categorizes the country into a major category and a minor category based on the call attempts information. For a subset of countries, and for each key performance indicator in a subset of selected key performance indicators, the system monitors performance using an anomaly detection model to identify an anomaly in network performance, determines an actual value of the key performance indicator for the detected anomaly, and computes a variation value of the determined actual value based on a predicted range of values. The system ranks countries using the computed variation values, to indicate problematic parts of the wireless telecommunication network.

Abstract: A computing system may automatically infer one or more events that occur during an application session involving activity on a network, such as the Internet. Such an application session may be interactions with, for example, social networking websites, banking websites, news websites, and so on. Events are any of a number of activities or transactions that may occur during the application session. The computing system may automatically infer an event by gathering network transaction data for network transactions performed by one or more client devices of a wireless communication network. The computing system may generate a network activity signature based, at least in part, on the network transaction data and apply pattern recognition and/or machine learning to the network activity signature to infer events associated with the network activity signature.

Abstract: A network performance degradation detection system is provided. The system receives Key Performance Indicator (KPI) values for devices of different device types running different software versions. The system determines baseline values of a first device type by averaging the KPI values of the different software versions running on devices of the first device type. The system compares KPI values for a first software version running on devices of the first device type with the determined baseline values to produce a set of comparison results. The system applies the set of comparison results to a classification model to determine whether the first software version running on devices of the first device type causes network performance degradation.

Abstract: An anomaly detection and analysis system generates analysis or summary of the anomalies detected from key performance indicators (KPIs). The system receives anomaly data reporting anomalies detected in key performance indicator (KPI) data. The system classifies the reported anomalies into a plurality of anomaly items, wherein anomalies from KPI data that share a set of features are assigned to one anomaly item. The system computes a ranking score for each anomaly item by assigning predefined weights for different anomaly types that are present in the anomaly item. The system sorts a list of anomaly items from the plurality of anomaly items into a sorted list of anomaly items according to the ranking scores computed for the plurality of anomaly items. The system sends the sorted list of anomaly items to a user device for presentation.

Abstract: An anomaly detection and analysis system generates analysis or summary of the anomalies detected from key performance indicators (KPIs). The system receives anomaly data reporting anomalies detected in key performance indicator (KPI) data. The system classifies the reported anomalies into a plurality of anomaly items, wherein anomalies from KPI data that share a set of features are assigned to one anomaly item. The system computes a ranking score for each anomaly item by assigning predefined weights for different anomaly types that are present in the anomaly item. The system sorts a list of anomaly items from the plurality of anomaly items into a sorted list of anomaly items according to the ranking scores computed for the plurality of anomaly items. The system sends the sorted list of anomaly items to a user device for presentation.

Abstract: The disclosed system identifies and resolves telecommunication network problems and/or problems with devices utilizing the network. The system segments the user devices into groups having similar characteristics, by grouping together devices more likely to leave the network, and devices less likely to leave the network. Each group of devices has a trained machine learning model, to predict the NEX score for each device. The system focuses on devices with a low NEX score. The system determines whether the problem is with the network and/or the problem is with the device. The system gathers information from network monitoring software and determine which tower caused, for example, the call to drop. The system determines technological capability of the device from the device TAC number. The system suggests solutions to the network problems by suggesting adding a sector, tuning an antenna, etc., or if the device has problems, adjusting phone settings.

Abstract: An analyzer configured to monitor a radio access network (RAN) of a cellular network is provided. The RAN includes multiple clusters that each includes multiple sites and multiple cells. The analyzer receives a multiple key performance indicator (KPI) measurements from the multiple clusters. Each KPI measurements generated for one of several KPI types. The analyzer receives information identifying anomalous KPI measurements in the received KPI measurements. For a cluster of the RAN, the analyzer identifies one or more common anomalous KPI types that satisfy a ubiquity criterion. The analyzer ranks the identified common anomalous KPI types for the cluster based on an anomaly metric that is derived from the anomalous KPI measurements generated for each identified common anomalous KPI type. The analyzer outputs a list of common anomalous KPI types for the cluster based on the ranking.

Abstract: An analyzer configured to monitor a radio access network (RAN) of a cellular network is provided. The RAN includes multiple clusters that each includes multiple sites and multiple cells. The analyzer receives a multiple key performance indicator (KPI) measurements from the multiple clusters. Each KPI measurements generated for one of several KPI types. The analyzer receives information identifying anomalous KPI measurements in the received KPI measurements. For a cluster of the RAN, the analyzer identifies one or more common anomalous KPI types that satisfy a ubiquity criterion. The analyzer ranks the identified common anomalous KPI types for the cluster based on an anomaly metric that is derived from the anomalous KPI measurements generated for each identified common anomalous KPI type. The analyzer outputs a list of common anomalous KPI types for the cluster based on the ranking.

Abstract: An anomaly detection and analysis system detects anomalies in time series data from key performance indicators (KPIs). The system decomposes samples of the time series data received during a first time interval into a trend component, a seasonality component, and a randomness component. The system identifies an upper bound and a lower bound based on the trend component, the seasonality component, and a variance of the randomness component. The system reports a sample received after the first time interval as an anomaly when the sample exceeds the upper bound or the lower bound. The system recalculates the trend component, the seasonality component, and the randomness component when more than a threshold percentage of the samples of the time series data received during a second time interval are reported as being anomalous.