Abstract: Systems and methods for generating augmented search results are disclosed. An example method may be performed by one or more processors of a ranking system and include receiving a search query, generating, for each respective data asset of a plurality of data assets, a query relevancy score indicating a likelihood that the respective data asset is relevant to the search query, retrieving, for each of a number of potentially relevant data assets, a utility score indicating a likelihood that the potentially relevant data asset is useful, generating, for each of the number of potentially relevant data assets, an augmented query score indicating a likelihood that the potentially relevant data asset is relevant to and useful for the search query based on the query relevancy scores and the utility scores, and generating augmented search results for the search query based on the augmented query scores.

Abstract: A method including transcribing, into digital tokens, utterances from a conversation between an agent and a person. The method also includes embedding the digital tokens into an utterances tensor including sequences of the digital tokens. The method also includes obtaining a metadata tensor by encoding metadata related to the utterances into the metadata tensor. The method also includes executing a machine learning model which takes, as input, the utterances tensor and the metadata tensor, and which outputs a predicted source article predicted to be related to the utterances. The method also includes generating an interactive link to the predicted source article.

Abstract: A testing system and method for testing application code against various failure scenarios. The testing system and method generate a test workflow including test source code implementing a series of actions that affect an application component and or an infrastructure component included in application code. The testing system and method execute the test workflow to determine the performance of the application code during one or more failure scenarios caused by the series of actions included in the test workflow. Performance data generated by the test code is analyzed by a performance analysis service or method to identify limitations of the application code and build resiliency patterns that address the limitations and improve the performance of the application code.

Abstract: A processor may receive user interaction data of a user for a plurality of electronically-presented offers. The processor may generate a plurality of labels, the generating comprising generating a label for each respective offer according to a comparison of the quality of the user interactions of the respective offer to the frequency of the user interactions of the respective offer. Each label may be a positive label or a negative label. The processor may determine whether the generating produced both positive and negative labels. The processor may select one of a plurality of available ML models, wherein a two-class ML model is chosen in response to determining that the generating produced both positive and negative labels and a one-class ML model is chosen in response to determining that the generating did not produce both positive and negative labels. The selected ML model may be trained and/or may be used to process user profile data and provide recommendations.

Abstract: A computing system identifies an incoming voice call from a user device to an agent device associated with the computing system. The computing system generates a transcription of the incoming voice call using one or more natural language processing techniques. The computing system extracts a problem description from the transcription. The problem description indicates a topic for the incoming voice call. A first machine learning model estimates a situation vector from the problem description. A second machine learning model identifies a pre-existing situation vector that closely matches the estimated situation vector. The computing system retrieves a situation description that corresponds to the identified pre-existing situation vector.

Abstract: Systems and methods for throttling requests to web services are disclosed. A system is configured to receive, at a host, one or more requests during a first time period. Each request is for a web service hosted on a backend. The host is one of a plurality of hosts of an application programming interface (API) gateway to receive a plurality of requests for the web services. The system is further configured to: process at least a portion of the one or more requests for the one or more web services; count, by a local counter in a local cache of the host, the one or more requests received at the host during the first time period; compare a local count of the local counter to a local bucket size associated with the host; and provide an instruction to update a remote count of a remote counter based on the comparison.

Abstract: Predictive modeling within a special purpose hardware platform to determine scenarios that are most likely to increase conversion potential for each trial user and retention potential for each active subscriber of a service, collectively referred to as a propensity score. The predictive models are integrated with a contextual marketing system that uses a loss risk assessment to learn user behavior and optimize content messaging designed to improve actual conversion or retention behavior for the user.

Abstract: Embodiments disclosed herein may extract trending topics from phone call transcripts or any type of text data. The phone call transcripts may be collected for a time period and the time period may be divided into time spans. For each time span having more than a threshold number of phone call transcripts, n-grams from the phone call transcripts may be extracted. The extracted n-grams may be contextually clustered by converting the n-grams into their embedding vectors, reducing the dimensionality of the embedding vectors, and clustering similar reduced dimensionality embedding vectors. Normalized occurrences of one or more clusters may be generated. The recent mean of the number of occurrences of the normalized clusters may be compared with the historical mean and offset by historical standard deviation to generate a modified Z-score. N-grams corresponding to the clusters with high Z-scores may be identified as trending topics.

Abstract: A computer-implemented system and method for predicting and flagging an anomaly entered in a digital form. A server computing device classifies a plurality of data fields of the digital form to identify a set of non-zero value data fields; and obtains an anomaly detection model comprising a statistical tree structure associated with the data field of the digital form. The server computing device receives datasets including a target value of a data field and values of a set of cohorting data features; traverses a statistical tree structure of the anomaly detection model with the target dataset to form a set of target cohorts to determine a target statistic value for the data field; flags the data field value of the target dataset as an anomaly item; and generates one or more confidence scores for a runtime prediction based on one or more variance changes for the data field.

Abstract: Certain aspects of the present disclosure provide techniques for processing north-south queries in an application programming interface (API) generated for a plurality of associated services. The method generally includes receiving a request to invoke a function exposed by a first service in the API, wherein the function commits data to one or more data repositories. A data record is generated by the first service based on the received request based on a data object definition and a service definition for the first service. A message is generated including the generated data record and an indication that the message is to be consumed by a data lake processing system to commit data to a data lake. The message is published on a message bus for consumption by the data lake processing system.

Abstract: A method including receiving, at a mobile device, a request to provision a first geofence. The first geofence has a first predicted duration that expires before a second predicted geofence is provisioned by the mobile device. Provisioning the first geofence and the second predicted geofence together consume a first amount of power. The method also includes determining a predictive geofence radius based on a predicted speed of the mobile device. The method also includes provisioning a first predictive geofence having the predictive geofence radius and further having a second predicted duration that is different than the first predicted duration and that expires before a second predictive geofence is provisioned by the mobile device. Provisioning the first predictive geofence and the second predictive geofence together consume a second amount of power, the second amount of power less than the first amount of power.

Abstract: Systems and methods for verifying an identity of a party to a video call are disclosed. An example method may be performed by one or more processors of a verification system and include storing first facial encodings, each of the first facial encodings associated with one or more images of a corresponding registered user of a plurality of registered users associated with the verification system, receiving, during a video call associated with a specified registered user of the plurality of registered users, a second facial encoding associated with one or more images of a person participating in the video call, the second facial encoding received from a device with which the person uses to participate in the video call, identifying, among the first facial encodings, a first facial encoding corresponding to the specified registered user, and determining whether the person participating in the video call is the specified registered user.

Abstract: Certain aspects of the present disclosure provide techniques for generalized metric for machine learning model evaluation for unsupervised classification including: for each unsupervised machine learning model of one or more unsupervised machine learning models: generating a first set of synthetic inputs for the model of the one or more unsupervised machine learning models; providing the first set of synthetic inputs to the model trained to output a prediction for each input of the first set of synthetic inputs, wherein the prediction indicates whether the input is of a first class; identifying, based on an output of the model, a second set of synthetic inputs predicted to be of the first class; determining, based on a set of expected normal inputs for the model and the second set of synthetic inputs, an accuracy score for the unsupervised machine learning model; and providing the accuracy score for display.

Abstract: A location system of a mobile device may detect a first location and a second location of the mobile device. A processor of the mobile device may determine that the first location and second location are within a geofence surrounding the location of interest. The processor may determine a speed at which the mobile device traveled from the first location to the second location. The processor may determine that the mobile device is loitering at the location of interest in response to the speed being below a threshold value. In response to determining that the mobile device is loitering at the location of interest, the processor may perform check in processing to check a user of the mobile device as being on site at the location of interest.

Abstract: A document extraction system executed by a processor, may process documents using manual and automated systems. The document extraction system may efficiently route tasks to the manual and automated systems based on a predicted probability that the results generated by the automated system meet some baseline level of accuracy. To increase document processing speed, documents having a high likelihood of accurate automated processing may be routed to an automated system. To ensure a baseline level of accuracy, documents having a smaller likelihood of accurate automated processing may be routed to a manual system.

Abstract: Systems and methods are used to generate contact type predictions that route user customer service requests within a support platform. The contact type predictions are generated using a hybrid model that includes a deep learning component and a business logic component. The deep learning component may generate a multi-channel output based on text features and context features. The multi-channel output is modified based on one or more business rules to generate the contact type predictions.

Abstract: Systems and methods may be used to generate and use a merchant community graph generated based on merchant financial transaction data. Connections between merchants and other data within the merchant community graph can be used to detect fraud, target product offerings and or other advertisements, detect similar communities, generate dynamic attributes that may be used to develop machine learning models, and develop new user interfaces (UIs) and other features of an information service.

Abstract: A method optimizes questions to retain engagement. The method includes generating, using a machine learning model, a churn risk from user interaction data. The method includes selecting, when the churn risk satisfies a threshold, a field, from multiple fields, using multiple prediction confidences corresponding to multiple prediction values generated for the multiple fields. The method includes obtaining a prediction value for the field and obtaining a question, corresponding to the field, using the prediction value. The method includes presenting the question and receiving a user input in response to the question.

Abstract: Processing compliance documents based on an artificial intelligence (AI) model is described herein. A system is configured to obtain a compliance document and obtain seed data associated with the compliance document. The seed data includes a plurality of sample text inputs and a plurality of sample computer readable operations associated with the plurality of sample text inputs. The system is also configured to parse text in the compliance document into one or more text segments, provide the one or more text segments and the seed data to the AI model, and obtain, from the AI model, one or more computer readable operations associated with the one or more text segments. The one or more computer readable operations are generated by the AI model based on few-shot learning using the seed data. The system is also configured to store the one or more computer readable operations for completing the compliance document.

Abstract: A method including transforming metrics, related to a computer network environment, into a digital image including pixels that represent the metrics. The computer network environment initially is load balanced by a first load balancing scheme selected from among load balancing schemes. The method also includes generating a classification of the digital image. The method also includes selecting, based on the classification of the digital image, a selected load balancing scheme from among the load balancing schemes. The method also includes changing the first load balancing scheme to the selected load balancing scheme such that the selected load balancing scheme is applied to the computer network environment.