Abstract: Systems and methods are provided that, in some embodiments that extract user data from at least one data warehouse. The user data is sorted within each dimension, and partitions each dimension into bins. Clusters are defined as each bin that includes user data for a number of users that exceeds a threshold. Clusters are determined for every combination of dimensions. Each combination of clusters that exceed the threshold is defined as clusters that are formed from multiple dimensions. All clusters and other clusters are stored into a cluster definition table. The clusters are used to analyze the profile of specific users.

Abstract: An ordered combination of machine-learning models may be used to identify users who are likely to abandon use of an application, predict the reasons why those users are likely to abandon, and identify intervening actions that the application can perform to reduce the probability that the users will abandon the application. For example, a first machine-learning model determines a retention-prediction value indicating a probability that the user will complete a target action in the application before a session terminates. If the retention-prediction value satisfies a threshold condition, a second machine-learning model determines a reason why the session is likely to terminate before the user completes the target action. A third machine-learning model determines an intervention action for the application to perform to increase the probability that the user will complete the target action before the session terminates.

Abstract: Systems described herein apply an ordered combination machine-learning models to identify users who are likely to abandon use of an application, predict the reasons why those users are likely to abandon, and identify intervening actions that the application can perform to reduce the probability that the users will abandon the application. A first machine-learning model determines a retention-prediction value indicating a probability that the user will complete a target action in the application before a session terminates. If the retention-prediction value satisfies a threshold condition, a second machine-learning model determines a reason why the session is likely to terminate before the user completes the target action. A third machine-learning model determines an intervention action for the application to perform to increase the probability that the user will complete the target action before the session terminates.

Abstract: A method for rule-based composition of user interfaces. A machine-learned rule repository is established based on previously observed combinations of UI states, UI features, and user features. A classifier classifies users into segments. Each segment includes users for which a combination of user features and UI states are defined. A first machine learning model (MLM) estimates a user segment-content preference including preferred UI content. A second MLM estimates a seen content-seen content similarity UI content preferences estimated according to prior UI content a user has seen. Based on the UI state and based on the user ID, rule-based recipes are obtained. Each rule-based recipe specifies a corresponding UI content suitable for an interaction between the user and the interface. A selected rule-based recipe is selected from the rule-based recipes. Specific UI content specified by the selected rule-based recipe is obtained, and the interface is updated with the specific UI content.

Abstract: A method for rule-based composition of user interfaces involves obtaining a user identity (ID) of a user accessing an application using a user interface and obtaining a user interface (UI) state of the user interface. Based on the UI state and based on the user ID, a plurality of rule-based recipes are obtained. Each rule-based recipe specifies a UI content suitable for an interaction between the user and the user interface. The method further includes ranking each of the rule-based recipes of the plurality of rule-based recipes based on a likeliness that the rule-based recipe is suitable, given the UI state and the user ID, identifying, from the ranked plurality of rule-based recipes, a highest-ranked rule-based recipe, obtaining the UI content specified by the highest-ranked rule-based recipe, and updating the user interface with the UI content.

Abstract: Method and apparatus for providing personalized self-help experience in online application. A predictive model is trained to learn a relationship between one or more user features and one or more tags using historical user feature data. High-dimensional vectors representing each of a plurality of questions are generated and stored in the lookup table. The trained predictive model outputs tags probabilities from the incoming user data, using the learned relationship. A user high-dimensional vector is formed based on the tags probabilities. Similarity metrics are calculated between the high-dimensional vector for the respective question and the user high dimensional vector. One or more of the most relevant question titles are returned to a client device for presentation to a user.

Abstract: Method and apparatus for providing a personalized self-support service to a user of an online application coupled with an online community forum. Embodiments include obtaining a plurality of questions from the online community forum and obtaining historical user data. Embodiments further include identifying one or more part-of-speech words in the plurality of questions and generating a high-dimensional vector for each question of the plurality of questions based on a frequency of the one or more part-of-speech words. Embodiments further include identifying one or more user features of the plurality of users based on the historical user data and establishing, based on the historical user data, one or more statistical correlations between user features and part-of-speech words. Embodiments further include training a predictive model based on the one or more statistical correlations. Embodiments further include using the predictive model to predict to provide one or more relevant questions to the user.

Abstract: Certain aspects of the present disclosure provide techniques for providing a routing system to a user of a product. An example technique includes receiving from a user of a product a query and a personal ID. Based on the personal ID of the user, the user's profile is retrieved which comprises user attribute data, a clickstream history of the user, and a product SKU of the product. Based on the query and the user profile, processed user data is generated. Additionally, agent profile data for each available agent is retrieved, and based on the user attribute data, the processed user data, and the agent profile data of each agent, a predicted quality score is generated for each agent. The agent with the highest predicted quality score is determined, and the user is routed to the agent with the highest predicted quality score.

Abstract: Systems described herein apply an ordered combination machine-learning models to identify users who are likely to abandon use of an application, predict the reasons why those users are likely to abandon, and identify intervening actions that the application can perform to reduce the probability that the users will abandon the application. A first machine-learning model determines a retention-prediction value indicating a probability that the user will complete a target action in the application before a session terminates. If the retention-prediction value satisfies a threshold condition, a second machine-learning model determines a reason why the session is likely to terminate before the user completes the target action. A third machine-learning model determines an intervention action for the application to perform to increase the probability that the user will complete the target action before the session terminates.

Abstract: An ordered combination of machine-learning models may be used to identify users who are likely to abandon use of an application, predict the reasons why those users are likely to abandon, and identify intervening actions that the application can perform to reduce the probability that the users will abandon the application. For example, a first machine-learning model determines a retention-prediction value indicating a probability that the user will complete a target action in the application before a session terminates. If the retention-prediction value satisfies a threshold condition, a second machine-learning model determines a reason why the session is likely to terminate before the user completes the target action. A third machine-learning model determines an intervention action for the application to perform to increase the probability that the user will complete the target action before the session terminates.

Abstract: This disclosure relates to customizing deployment of an application to a user interface of a client device. An exemplary method includes training a model based on historical context information of a plurality of users by identifying correlations between the historical context information and a plurality of user interface components. The method further includes receiving context information from the client device. The method further includes determining a user intent based on the context information using the model. The method further includes customizing one or more widgets by selecting one or more user interface components to include in the one or more widgets based on the user intent. The method further includes generating a custom user interface definition comprising the one or more widgets. The method further includes transmitting, to the user interface of the client device, the custom user interface definition.

Abstract: Systems and methods are provided that, in some embodiments that extract user data from at least one data warehouse. The user data is sorted within each dimension, and partitions each dimension into bins. Clusters are defined as each bin that includes user data for a number of users that exceeds a threshold. Clusters are determined for every combination of dimensions. Each combination of clusters that exceed the threshold is defined as clusters that are formed from multiple dimensions. All clusters and other clusters are stored into a cluster definition table. The clusters are used to analyze the profile of specific users.

Abstract: Systems and methods are provided that, in some embodiments that extract user data from at least one data warehouse. The user data is sorted within each dimension, and partitions each dimension into bins. Clusters are defined as each bin that includes user data for a number of users that exceeds a threshold. Clusters are determined for every combination of dimensions. Each combination of clusters that exceed the threshold is defined as clusters that are formed from multiple dimensions. All clusters and other clusters are stored into a cluster definition table. The clusters are used to analyze the profile of specific users.

Abstract: Method and apparatus for providing a personalized self-support service to a user of an online application coupled with an online community forum. Embodiments include obtaining a plurality of questions from the online community forum and obtaining historical user data. Embodiments further include identifying one or more part-of-speech words in the plurality of questions and generating a high-dimensional vector for each question of the plurality of questions based on a frequency of the one or more part-of-speech words. Embodiments further include identifying one or more user features of the plurality of users based on the historical user data and establishing, based on the historical user data, one or more statistical correlations between user features and part-of-speech words. Embodiments further include training a predictive model based on the one or more statistical correlations. Embodiments further include using the predictive model to predict to provide one or more relevant questions to the user.

Abstract: Method and apparatus for providing a personalized self-support service to a user of an online application coupled with an online community forum. Embodiments include obtaining a plurality of questions from the online community forum and obtaining historical user data. Embodiments further include identifying one or more part-of-speech words in the plurality of questions and generating a high-dimensional vector for each question of the plurality of questions based on a frequency of the one or more part-of-speech words. Embodiments further include identifying one or more user features of the plurality of users based on the historical user data and establishing, based on the historical user data, one or more statistical correlations between user features and part-of-speech words. Embodiments further include training a predictive model based on the one or more statistical correlations. Embodiments further include using the predictive model to predict to provide one or more relevant questions to the user.

Abstract: The present disclosure relates to dynamically generating user interfaces based on a user's emotional state. An example method generally includes a computer system receiving emotional response data from a client device. The computer system identifies a version of a user experience to present on the client device based on the received emotional response data and generates code for rendering a user interface associated with the identified version of the user experience. The generated code is transmitted to the client device for rendering and presentation on the client device.

Abstract: Method and apparatus for providing a personalized self-support service to a user of an online application coupled with an online community forum. Embodiments include obtaining a plurality of questions from the online community forum and obtaining historical user data. Embodiments further include identifying one or more part-of-speech words in the plurality of questions and generating a high-dimensional vector for each question of the plurality of questions based on a frequency of the one or more part-of-speech words. Embodiments further include identifying one or more user features of the plurality of users based on the historical user data and establishing, based on the historical user data, one or more statistical correlations between user features and part-of-speech words. Embodiments further include training a predictive model based on the one or more statistical correlations. Embodiments further include using the predictive model to predict to provide one or more relevant questions to the user.

Abstract: Method and apparatus for providing personalized self-help experience in online application. A predictive model is trained to learn a relationship between one or more user features and one or more tags using historical user feature data. High-dimensional vectors representing each of a plurality of questions are generated and stored in the lookup table. The trained predictive model outputs tags probabilities from the incoming user data, using the learned relationship. A user high-dimensional vector is formed based on the tags probabilities. Similarity metrics are calculated between the high-dimensional vector for the respective question and the user high dimensional vector. One or more of the most relevant question titles are returned to a client device for presentation to a user.

Abstract: Systems and methods are provided that, in some embodiments that extract user data from at least one data warehouse. The user data is sorted within each dimension, and partitions each dimension into bins. Clusters are defined as each bin that includes user data for a number of users that exceeds a threshold. Clusters are determined for every combination of dimensions. Each combination of clusters that exceed the threshold is defined as clusters that are formed from multiple dimensions. All clusters and other clusters are stored into a cluster definition table. The clusters are used to analyze the profile of specific users.

Abstract: Devices, systems and methods are disclosed for estimating characteristics of noise included in one-dimensional data. For example, a number of data points associated with noise below each of a plurality of thresholds may be determined to calculate a cumulative distribution function. A probability density function may be derived from the cumulative distribution function. A variance may be calculated from the cumulative distribution function and/or the probability density function. The noise may be modeled using the variance and other characteristics determined from the cumulative distribution function and/or the probability density function.