Abstract: A method and system adaptively improves potential customer conversion rates, revenue metrics, and/or other target metrics by providing effective user experience options to some users while concurrently testing user responses to other user experience options, according to one embodiment. The method and system selects the user experience options by applying user characteristics data to an analytics model, according to one embodiment. The method and system analyzes user responses to the user experience options to update the analytics model, and to dynamically adapt the personalization of the user experience options, according to one embodiment. The method and system filters out user experience options from delivery to users, if the user experience options are non-compliant with one or more business rules, to maintain business relations for the service provider and to maintain user confidence in the services provided by the service provider (e.g., a tax return preparation system), according to one embodiment.

Abstract: A method that predicts business income from user transaction data. A multinomial classifier is trained, using a vector of features from data related to a historical transaction and a label associated with the historical transaction, to generate a probability that the historical transaction belongs to a specific classification with respect to income. Data related to a new transaction is split into a set of unigrams. A new vector of features is generated from the data related to the new transaction. The new vector includes a set of values that correspond and are assigned to the set of unigrams. A classification with respect to income is determined for the new transaction by applying the multinomial classifier to the new vector. The new transaction is labeled with the classification. One or more fields of a form that is maintained by an online service is populated using the classification.

Abstract: A method for presence based mobile payment. The method includes registering, by a mobile payment service, a user based on a request for registration with a mobile payment service, wherein registering includes storing information items identifying the user. The method also includes receiving, from a point of sale system (PoS), an information item from the information items, and a payment authorization for a transaction. The information item is broadcast through an advertising channel by a mobile device to the PoS in accordance with a communication standard without pairing the PoS to the mobile device. The advertising channel is a channel of the communication standard and is used for identification of available nearby devices. The method also includes facilitating, based on the information item and by the mobile payment service, payment for the transaction from an electronic financial account of the user.

Abstract: The invention relates to a method for characterizing time series. The method includes obtaining a time series that includes data points for consecutive time intervals, decomposing the time series into a trend component, a seasonal component, and a residual component, and clustering the seasonal component into a number of clusters. Each of the consecutive time intervals is assigned to one of the clusters based on a level of the data points of the seasonality component in the time interval. The method further includes labeling one of the clusters as baseline with the baseline cluster containing more time intervals than any one of the other clusters, determining cluster labels for the other clusters based on the level of the data points of the time intervals assigned to the other clusters; and for each time interval, storing the corresponding cluster label in a vector.

Abstract: A method for generating a software container includes receiving a software application and a containerization file. The method also includes building an image file using the containerization file, the image file containing the software application. The method also includes recording, in the image file, an image lineage. The method also includes performing a security scan of the image file to obtain a result, the security scan comprising checking the image file for inadequacies. The method also includes assigning, to the image file, a security level selected from among a plurality of different predetermined security levels. Assigning is based on a combination of the image lineage and the result of the security scan. The method also includes signing the image file with the security level to create a signed image file. The method also includes storing the signed image file as the software container.

Abstract: Systems and methods that implement a paired-consistency-based process for evaluating and or regulating fairness in machine learning models.

Abstract: Systems and methods that may implement an anomaly detection process for time series data. The systems and methods may implement a model ensemble process comprising at least one machine learning model in a supervised class and at least one machine learning model in an unsupervised class.

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 that may be used to automatically generate inventory templates for use with an accounting platform. The automatically generated templates may be for a first user within a particular industry and may be based on established inventory trees of other system users within the same industry that have similar demographics of the first user.

Abstract: Modular computerized tax return preparation systems that allow tax data determinations to be performed when the user indicates that tax data will not be entered, which may be for various reasons including the user declining to provide personal information (such as a social security number or other personal identification information), not knowing the requested data, or not having the requested data readily available. A user interface controller presents an interview screen requesting data tax return data and the user responds that the requested data applies to the user but is not being entered or will be entered at a later time. A placeholder is generated or selected for interim use until the requested data is provided by the user, and when provided, tax data determinations and calculations can be repeated as necessary if the provided data differs from the placeholder.

Abstract: A computer-implemented method for estimating a likelihood of an individual qualifying for a benefit program includes obtaining a profile corresponding to the individual. The method also includes determining respective likelihoods that the profile would satisfy each completeness graph in a set of completeness graphs by running the profile against each completeness graph using an evaluation algorithm. Each completeness graph corresponds to a respective benefit program.

Abstract: A system and method for information extraction character level features. The system and method may be used for data extraction for various types of content including a receipt or a tax form.

Abstract: The method and system involves instant loan decisions by generating a risk profile of a small business (SMB). The risk profile is generated based on accounting data and other third party business management application (BMA) data of the SMB. In particular, the accounting data and other third party BMA data are retrieved from a BMA (e.g., accounting application, payroll application, tax preparation application, personnel application, etc.) as a software-as-an-service (SaaS) used by the SMB. Specifically, the risk profile represents the likelihood of the SMB to be delinquent and/or to default on a loan. The risk profile is then provided to a lender for making an expedient lending decision with respect to the SMB. In addition, statistics of lenders' lending decisions based on provided risk profiles are analyzed to generate a correlation. Accordingly, the algorithm(s) used to generate the risk profile from the accounting data and other third party BMA data are adjusted to maximize the correlation.

Abstract: A method and computer program product for identifying errors in a computerized preparation of an electronic financial document to be submitted to an agency. A financial document preparation program is executed, a first connection is initiated with a computer of the agency, the schema requirements are received from the agency via the first connection, data entry fields are generated, data entered into the data entry fields is received, error rules are generated from the schema requirements, each error rule associated with a respective one of the data entry fields associated with the electronic financial document, the data entered into the data entry fields is checked against the respective error rules to identify schema errors in the data entered into the data entry fields, the electronic financial document is generated based on the data entered into the data entry fields, and the electronic financial document is electronically transmitted to the agency.

Abstract: A method may include obtaining a first dependency of a first calling service on a called service and a second dependency of a second calling service on the called service. The method may further include generating a first routing configuration for the first calling service that maps a name of the called service to an ingress point of a first cluster, generating a second routing configuration for the second calling service that maps the name of the called service to the ingress point of the first cluster, detecting a relocation of the called service to a second cluster, modifying the first routing configuration to map the name of the called service to an ingress point of the second cluster, and modifying the second routing configuration to map the name of the called service to a local name of the called service within a namespace of the second cluster.

Abstract: A method and a system for fingerprinting signatures and enhanced signature capturing for charge card transactions on mobile communication devices identify and transform a signature into a normalized representation such as a bytestream including coordinate data and combine the normalized representation with some transaction information into combined data. The combined data is further transformed into fixed length data stored for further analyses or uses. Upon the receipt of a dispute, the stored representation(s) of the signature and some of the stored transaction information of the transaction in dispute may be transformed into a first fixed length data. The first fixed length data may be further compared with the stored fixed length data obtained at the time of the transaction to resolve the dispute. The representations of signatures of multiple transactions may be used to determine whether these multiple transactions belong to the same category or to different categories.

Abstract: A method may include obtaining a first dependency of a first calling service on a called service and a second dependency of a second calling service on the called service. The method may further include generating a first routing configuration for the first calling service that maps a name of the called service to an ingress point of a first cluster, generating a second routing configuration for the second calling service that maps the name of the called service to the ingress point of the first cluster, detecting a relocation of the called service to a second cluster, modifying the first routing configuration to map the name of the called service to an ingress point of the second cluster, and modifying the second routing configuration to map the name of the called service to a local name of the called service within a namespace of the second cluster.

Abstract: This disclosure provides systems, methods and apparatus for manipulating virtual objects in a virtual reality (VR) environment. In some implementations, an augmented reality (AR) system determines an orientation of a physical object in the real-world space based at least in part on images or video of the physical object captured by an image capture device, and generates a virtual object representative of the physical object based at least in part on the orientation and the at least one detected feature. The AR system detects movement of the physical object in the real-world space based at least in part on the captured images or video, and manipulates the virtual object based at least in part on the detected movements of the physical object. In some aspects, the AR system can determine the orientation and detect movement of the physical object without receiving control signals or communications from the physical object.

Abstract: A processor of a remote crypto cluster (RCC) may receive a public key from a client device through at least one network. The processor of the RCC may obtain an encrypted specific key and a blinded project key from at least one data source through the at least one network. The processor of the RCC may derive a derived key in blind based on the encrypted specific key and the blinded project key.

Abstract: Techniques are disclosed for automatically adding accessibility properties to an application. One embodiment includes a method for adding accessibility. The method includes, for at least a first interface control provided by a first software application, the first interface control including a set of one or more accessibility properties: determining a control type of a plurality of control types based on one or more attributes of the first interface control, and associating the control type with the set of accessibility properties of the first interface control. The method further includes matching at least a second interface control provided by a second software application to one of the plurality of control types based on one or more attributes of the second interface control, the second interface control not including accessibility properties. The method further includes adding accessibility properties to the second interface control based on accessibility properties of the matched control type.