Abstract: A machine learning system executed by a processor may generate predictions for a variety of natural language processing (NLP) tasks. The machine learning system may include a single deployment implementing a parameter efficient transfer learning architecture. The machine learning system may use adapter layers to dynamically modify a base model to generate a plurality of fine-tuned models. Each fine-tuned model may generate predictions for a specific NLP task. By transferring knowledge from the base model to each fine-tuned model, the ML system achieves a significant reduction in the number of tunable parameters required to generate a fine-tuned NLP model and decreases the fine-tuned model artifact size. Additionally, the ML system reduces training times for fine-tuned NLP models, promotes transfer learning across NLP tasks with lower labeled data volumes, and enables easier and more computationally efficient deployments for multi-task NLP.

Abstract: Systems and methods for k-anonymizing a corpus of documents using linguistic similarities and embeddings distances between words. For instance, a word pair is selected based on linguistic similarity (e.g., belonging to the same part of speech) and small embeddings distance. For the selected word pair, a plurality of words is retrieved, also based on linguistic similarity to, and embeddings distances from, the selected word pair. Out of the plurality of words, a third word is identified that has a closer linguistic similarity to the word pair and also has smaller embeddings distances from the word pair. Each word in the word pair is then replaced by the third word. The process is repeated until k-anonymity is achieved.

Abstract: A method implements a streaming machine learning platform. The method includes processing event data with a preprocessing application to generate model input data and cardinality data in response to receiving the event data from a streaming application. The method further includes forwarding, using the cardinality data, the model input data to a machine learning model. The method further includes processing the model input data with the machine learning model to generate model output data stored to an output record. The method further includes processing the model output data from the output record with a post processing application to generate post data stored to a post record. The method further includes generating notification data using the post data of the post record and transmitting a notification of the notification data.

Abstract: Certain aspects of the present disclosure provide techniques for processing transactions in a computing system. An example method generally includes receiving a request to perform an operation with respect to an object included in the request. A system identifies an archetype defining properties of the object included in the request. Based on the identified archetype, the system identifies data repositories to commit data to in order to perform the requested operation and rules for performing the operation with respect to the object. One or more actions are executed against the identified data repositories according to the identified rules.

Abstract: Described herein are example implementations for handling of phishing attempts. A system receives a request to perform an electronic transaction, with the request including information regarding a user account. The system generates one or more probabilities of the request being valid based on the request and processing of a plurality of electronic transactions associated with one or more user accounts, identifies whether the request is valid based on the one or more probabilities, and in response to identifying that the request is not valid, provides an indication that the request is not valid.

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: A method involves receiving a first command. The first command includes a data extraction expression applied to fields of a dataset of a data source. The first command also includes a first report configuration expression applied to first dimensions of a first report. The method also involves generating, by executing the data extraction expression on the dataset, records of the dataset. The method also involves generating, by executing the first report configuration expression on the records, a first tree of subsets of the records. The method also involves populating, using the first report configuration expression and the first tree of subsets, cells of the first dimensions to obtain first populated dimensions. The method also involves generating, in response to receiving the first command and by traversing the first tree of subsets, the first report including the first populated dimensions.

Abstract: Described herein are example implementations for the automatic detection and handling of personal identifiable information (PII) in electronic records. In some aspects, a system receives one or more computer readable logs of information for one or more computer services, with each log including a string of characters. The system performs one or more string search algorithm based operations on the entirety of the one or more strings of the one or more computer readable logs to identify a range of the one or more strings to be searched for PII that is less than the entirety of the one or more strings. The system also performs one or more regular expression algorithm based operations on the range of the one or more strings to identify one or more instances of PII. The system generates and outputs an indication of the one or more instances of the PII that are identified.

Abstract: Certain aspects of the present disclosure provide techniques for intelligent grouping of travel data for review through a user interface. In one embodiment, a method for providing grouped travel data to a user interface of an application, comprises: receiving a plurality of trip records from an application running on a remote device; providing a first subset of the plurality of trip records to a prediction model; providing a second subset of the plurality of trip records to a model training module; receiving labels for each trip record of the first subset of the plurality of trip records from the prediction model; grouping the first subset of the plurality of trip records based on the received labels; and transmitting the grouped first subset of the plurality of trip records to the application running on the remote device.

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: 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: This disclosure relates to systems and methods for processing electronic payments for customer purchases. In some implementations, a mobile computing device receives a payment request identifying a sales transaction between a merchant and a customer. The payment request indicates a purchase amount owed by the customer to the merchant, and carries a set of instructions. The mobile computing device activates a card reader based at least in part on the set of instructions, receives customer authorization for electronic payment of the purchase amount, instructs the card reader to process a credit card for the purchase amount based on receiving the customer authorization, and receives confirmation of payment of the purchase amount from the card reader.

Abstract: A method for mitigating cold starts in recommendations includes receiving a request that identifies a requested page and identifying a content vector of the requested page. The content vector is generated based on providing text of the requested page to a neural network text encoder. The method further includes selecting, based on the content vector, a link to a cold start page that does not satisfy a threshold level of interaction data. The selected link is ranked above a second link to a warm page that does satisfy the threshold level of the interaction data. The method further includes presenting the requested page with the selected link.

Abstract: Systems and methods for recognizing domain specific named entities are disclosed. An example method may be performed by one or more processors of a text incorporation system and include extracting a number of terms from a text under consideration, identifying, among the number of terms, a set of unmatched terms that do not match any of a plurality of known terms, passing each respective unmatched term to a vectorization module, embedding a vectorized version of each respective unmatched term in a vector space, comparing each vectorized version to known term vectors, passing, to a machine learning model, candidate terms corresponding to known term vectors closest to the vectorized versions, identifying, using the machine learning model, a best candidate term for each respective unmatched term, mapping the best candidate terms to unmatched terms in the text under consideration, and incorporating the text under consideration into the system based on the mappings.

Abstract: Aspects of the present disclosure provide techniques for machine learning and rules integration. Embodiments include receiving input values corresponding to a subset of a set of input variables associated with an automated determination. Embodiments include generating a directed acyclic graph (DAG) representing a set of constraints corresponding to the set of input variables. The set of constraints relate to one or more machine learning models and one or more rules. Embodiments include receiving one or more outputs from the one or more machine learning models based on one or more of the input values. Embodiments include determining outcomes for the one or more rules based on at least one of the input values. Embodiments include populating the DAG based on the input values, the one or more outputs, and the outcomes. Embodiments include making the automated determination based on logic represented by the DAG.

Abstract: A method including embedding, by a trained issue MLM (machine learning model), a new natural language issue statement into an issue vector. An inner product of the issue vector with an actions matrix is calculated. The actions matrix includes centroid-vectors calculated using a clustering method from a second output of a trained action MLM which embedded prior actions expressed in natural language action statements taken as a result of prior natural issue statements. Calculating the inner product results in probabilities associated with the prior actions. Each of the probabilities represents a corresponding estimate that a corresponding prior action is relevant to the issue vector. A list of proposed actions relevant to the issue vector is generated by comparing the probabilities to a threshold value and selecting a subset of the prior actions with corresponding probabilities above the threshold. The list of proposed actions is transmitted to a user device.

Abstract: A method extracts explainable corpora embeddings. The method includes constructing a graph with nodes representing terms from a text sequence and edges that include pointwise values generated between pairs of terms. The method further includes generating a rank vector from the graph. Elements of the rank vector correspond to the edges of the graph. A rank value, of the rank values, corresponds to a term from the text sequence. The method further includes selecting the term by comparing the rank vector to a previous rank vector generated for a previous text sequence.