Abstract: Systems and methods for generating synthetic data are disclosed. For example, a system may include one or more memory units storing instructions and one or more processors configured to execute the instructions to perform operations. The operations may include receiving a dataset including time-series data. The operations may include generating a plurality of data segments based on the dataset, determining respective segment parameters of the data segments, and determining respective distribution measures of the data segments. The operations may include training a parameter model to generate synthetic segment parameters. Training the parameter model may be based on the segment parameters. The operations may include training a distribution model to generate synthetic data segments. Training the distribution model may be based on the distribution measures and the segment parameters.

Abstract: Various embodiments are directed to a password security warning system. An artificial neural network or other types of models may be used to determine whether a password that is created, input, or proposed by a user via an interface includes one or more predictable or typical transformations or combinations of characters derived from user-specific information. Based on the determination, a warning may be provided to the user.

Abstract: Systems and methods for selectively obscuring data in a digital statement are disclosed. The systems and methods can relate to multiple user accounts that access a shared account. The systems can include means to determine which account completed a transaction. Device location tracking and hardware characteristics of the devices can be used to determine which account completed the transaction. Information related to transactions that fall outside of expected spend categories can be selectively obscured on the digital statement of the non-transacting account. The information related to the transaction can be selectively viewable by non-transacting account.

Abstract: Disclosed embodiments may include a method that includes setting an influence level for each index that a neural network can accept in one sample to a same level for a neural network, receiving a training corpus including training input samples and a corresponding correct training prediction samples, generating, using the neural network, prediction samples, identifying an accuracy for each index by comparing the prediction samples with the corresponding correct training prediction samples, adjusting the influence level for each index based on the accuracy for each index, identifying one or more poorly accurate indexes for the neural network, receiving a first input sample including one or more characters, generating one or more normalized first input samples by applying one or more buffers to the one or more poorly accurate indexes, and generating, using the neural network, a categorization of each character in the one or more normalized first input samples.

Abstract: A device detects a communication involving a user associated with an account and a service representative, and sends, to a user device associated with the account, an authentication notification that causes the user device to display an authentication field for the user. The device sends, to a service representative device associated with the service representative, a message that indicates that the service representative is to request, via the communication, the user to enter personal information associated with the user into the authentication field, where the user device is configured to generate a first authentication code based on a user input received from the user device in the authentication field. The device generates a second authentication code based on personal information associated with the account from a data structure, receives the first authentication code, and performs an action based on the first authentication code and the second authentication code.

Abstract: In some embodiments, gaze-tracking-based image downscaling for multi-party video communication may be provided. In some embodiments, a set of gaze locations may be received from a set of receiving devices during concurrent video communication sessions between the receiving devices and a sending device. Different collections of positions may be determined based on the gaze locations, and, for each such collection, a spatial indicator (e.g., a bounded region) may be determined. A first downscaled encoding of the source image (from the sending device) may be generated based on the first spatial indicator, and a second downscaled encoding of the source image may be generated based on the second spatial indicator, etc. The downscaled encodings may then be sent to the respective receiving devices during the concurrent video communication sessions.

Abstract: The present disclosure relates to systems and methods for parsing unstructured data with neural networks. In one implementation, a system for parsing unstructured data may include at least one processor and at least one non-transitory memory storing instructions that, when executed by the at least one processor, cause the system to: receive unstructured data; apply a classifier to the unstructured data to identify a type of the unstructured data; based on the identification, select a corresponding neural network; apply the selected neural network to the unstructured data to obtain structured data; and output the structured data.

Abstract: A user device may obtain sensor data related to a physical environment. The user device may process the sensor data to determine that a potential threat exists in the physical environment and a threat score associated with the potential threat. The user device may determine that the threat score satisfies a threat score threshold. The user device may obtain, based on determining that the threat score satisfies the threat score threshold, additional information related to the physical environment. The user device may generate, based on the additional information related to the physical environment and at least one of the sensor data, information indicating the potential threat in the physical environment, or the threat score associated with the potential threat, threat determination information associated with the physical environment. The user device may perform, based on the threat determination information associated with the physical environment, one or more actions.

Abstract: In certain embodiments, document-based resolution of user inquiries may be facilitated. A predicted intent of the user is determined based on chat activity information associated with the user. A response to a user inquiry may be provided to the user via a chat interface based on the predicted intent. User response may be obtained for the response, which may indicate a failure in providing a resolution to the user (e.g., regarding a user inquiry). Upon detecting a failure in providing the resolution, a document associated with the user and matching the predicted intent may be obtained and presented to via the chat interface. The document may have content related to the predicted intent. The document may be presented as a response to the user response to seek a confirmation from the user regarding its relevance to the user inquiry.

Abstract: Methods, systems, and computing devices for visualizing natural language processing algorithm processes are described herein. A plurality of categories may be determined. Each color of a plurality of colors may correspond to the categories. Text content may be processed using a natural language processing algorithm. Confidence values indicating, for each of a plurality of portions of the text content, a degree of confidence corresponding to one or more of the plurality of categories may be determined. Display colors may be determined based on the confidence values. A user interface comprising a visualization of the text content may be displayed, and the user interface may be configured to show each portion of the text content using a display color such that the user interface indicates changes in confidence across the plurality of characters.

Abstract: Disclosed herein are system, method, and computer program product embodiments for a text-to-speech system. An embodiment operates by identifying a portion of content of the document, that is to be replaced with a summary associated with the content responsive to a request for an audible version of the document. A first request for the audible version of the document is received. A first summary of the content at a first level of detail is generated. The first summary is audibly output. A second request for additional information is received. A second summary of the content at different level of detail is generated. The second summary of the content at the second level of detail is audibly output.

Abstract: Methods and systems are described herein for improvements to authenticate users, particularly authenticating a user based on data known to the user. For example, methods and systems allow for users to be securely authenticated based on data known to the users over remote communication networks without storing the data known to the users. Specifically, methods and systems authenticate users by requiring users to select images that are known to the users. For example, the methods and systems may generate synthetic images based on the user's own images and require the user to select the synthetic image, from a set of a set of images, that is known to the user to authenticate the user. Moreover, the methods and systems alleviate storage and privacy concerns by not storing the data known to the users.

Abstract: A secure method for resetting the password for an account is disclosed. During the setup of the account, the user can provide the service provider with a media file, and when the user asks the service provider to reset the password for the account, the user will be prompted with several media files. The user can be asked to identify the media file that the user provided to the service provider at the time of the setup of the account. If the user properly identifies the media file, the password will be reset.

Abstract: Systems, methods, articles of manufacture, and computer program products to: train a prediction model using a machine learning process, the prediction model configured to estimate whether further application of a hyperparameter tuning technique will cause an improvement in at least one of the hyperparameters; select the hyperparameters using the tuning technique; apply the prediction model to determine if further adjustment of the hyperparameters is likely to improve the success metric; and terminate the tuning technique when: accuracy of the prediction model in predicting improvement in a hyperparameter is above a predetermined accuracy threshold, and the prediction model predicts that further application of the tuning technique will not result in an improvement to the hyperparameter; or the accuracy of the prediction model in predicting improvement in the parameter is below the predetermined accuracy threshold, and an accuracy of hyperparameter adjustment is determined to be below a predetermined adjustment

Abstract: Systems and methods for classifying data are disclosed. For example, a system may include at least one memory storing instructions and at least one processor configured to execute the instructions to perform operations. The operations may include receiving training data comprising a class. The operations may include training a data classification model using the training data to generate a trained data classification model. The operations may include receiving additional data comprising labeled samples of an additional class not contained in the training data. The operations may include creating a synthetic data generator. The operations may include training the synthetic data generator to generate synthetic data corresponding to the additional class. The operations may include generating a synthetic classified dataset comprising the additional class. The operations may include retraining the trained data classification model using the synthetic classified dataset.

Abstract: Systems and methods for architecture embeddings for efficient dynamic synthetic data generation are disclosed. The disclosed systems and methods may include a system for generating synthetic data configured to perform operations. The operations may include retrieving a set of rules associated with a first data profile and generating, by executing a hyperparameter search, a plurality of hyperparameter sets for generative adversarial networks (GANs) that satisfy the set of rules. The operations may include generating mappings between the hyperparameter sets and the first data profile and storing the mappings in a hyperparameter library. The operations may include receiving a request for synthetic data, the request indicating a second data profile and selecting, from the mappings in the hyperparameter library, a hyperparameter set mapped to the second data profile. The operations may include building a GAN using the selected hyperparameter set and generating, using the GAN, a synthetic data set.

Abstract: Systems and methods for ATM fraud detection are presented. ATM fraud detection may be provided in the context of including a currency sensor in the ATM to count currency notes. ATM fraud detection may be obtained by using a currency sensor, such as a scanner, to image and count any currency notes that remain after being dispensed to a user, and determining if the count is correct for the currency notes dispensed. A count discrepancy resulting from removal of some but not all dispensed currency notes would indicate attempted theft or fraud.

Abstract: A method includes receiving, by a processor, bias data categories. A data input from a user for classification in data categories is received. A classification machine learning model is utilized to classify the data input in at least one data category and determine a first confidence probability in a classification outcome. A bias filter machine learning model is utilized to determine a second confidence probability that the classification outcome of classifying the data input into the at least one data category is based on at least one bias characteristic associated with at least one bias data category. A gate machine learning model is utilized to determine when to output the classification outcome of classifying the data input into the at least one data category to a computing device of a user based at least in part on the first confidence probability, the second confidence probability, and a predefined bias threshold.

Abstract: Systems and methods for formatting data are disclosed. For example, a system may include at least one memory storing instructions and one or more processors configured to execute the instructions to perform operations. The operations may include receiving data comprising a plurality of sequences of data values and training a recurrent neural network model to output conditional probabilities of subsequent data values based on preceding data values in the data value sequences. The operations may include generating conditional probabilities using the trained recurrent neural network model and the received data. The operations may include determining a data format of a subset of the data value sequences, based on the generated conditional probabilities, and reformatting at least one of the data value sequences according to the determined data format.

Abstract: The present disclosure provides systems and methods for synthetic data generation. A recurrent neural network can be trained for synthetic data generation by obtaining a sequence of elements and determining, using a classifier, that the sequence corresponds to a token. In response to the determination, a recurrent neural network configured to use a first vocabulary including the elements can be modified to use a second vocabulary, the second vocabulary including the token and the first vocabulary. The modified recurrent neural network can be trained using the token and the sequence of elements. The trained recurrent neural network can be used to generate synthetic data. A classifier can detect sequences of elements in the synthetic data corresponding to tokens. The tokens can replace the sequences of elements in the generated synthetic data and can be provided to the trained recurrent neural network to continue synthetic data generation.