Abstract: An application programming interface (API) hub architecture establishes relationships among the provisioning systems, the IAM target systems, and an intermediate API hub system. A provisioning system can send the API hub system a message in a format associated with the IAM system. The API hub determines and identifies which of one or more available IAM systems should receive the message. The API hub system translates the message into a correct format for each of the determined target IAM systems and sends the translated messages to the appropriate target systems. Similarly, the API hub system can receive a message from an IAM system, identify, based on the message, which of one or more provisioning systems should receive the message, translate the message into an appropriate format for each of the determined one or more provisioning systems, and send the formatted message to the determined provisioning systems.

Abstract: The systems and methods provide a machine learning model that can exploit long time dependency for time-series sequences, perform end-to-end learning of dimension reduction and clustering, or train on long time-series sequences with low computation complexity. For example, the methods and systems use a novel, unsupervised temporal representation learning model. The model may generate cluster-specific temporal representations for long-history time series sequences and may integrate temporal reconstruction and a clustering objective into a joint end-to-end model.

Abstract: In certain embodiments, multi-modal-based generation of settlement instructions may be facilitated. In some embodiments, a portfolio of a live environment may be emulated in a projected environment. A target portfolio may be generated in the projected environment based on the emulated portfolio. Partial synchronization between the target portfolio of the projected environment and the portfolio of the live environment may be performed such that a first subset of changes to the portfolio of the live environment are reflected in the target portfolio of the projected environment. Subsequent to the partial synchronization, the target portfolio of the projected environment may be updated such that the update of the target portfolio accounts for the first subset of changes. Subsequent to the update of the target portfolio, settlement instructions may be generated based on differences between the target portfolio of the projected environment and the portfolio of the live environment.

Abstract: A system stores transaction data in a ring chain architecture. A ring chain comprises blocks of data stored as a length-limited block chain in a ring buffer configuration. A block of transactions is stored on a ring chain until enough new blocks are added to overwrite the ring buffer with new blocks. The system stores multiple ring chains that update at varying frequencies. A new block on a lower frequency ring chain stores an aggregation of data from the blocks that were added to a higher frequency ring chain in the time since the previous addition of a block to the lower frequency ring chain. Thus, a system of ring chains stores progressively summarized state transition data over progressively longer time intervals while maintaining immutability of the record and reducing storage requirements.

Abstract: The system uses the non-repudiatory persistence of blockchain technology to store all task statuses and results across the distributed computer network in an immutable blockchain database. Coupled with the resiliency of the stored data, the system may determine a sequence of processing tasks for a given processing request and use the sequence to detect and/or predict failures. Accordingly, in the event of a detected system failure, the system may recover the results prior to the failure, minimizing disruptions to processing the request and improving hardware resiliency.

Abstract: Methods and systems are described for balancing loads in distributed computer networks for computer processing requests with variable rule sets and dynamic processing loads. The methods and systems may include determining an initial allocation of the plurality of processing requests to the plurality of available domains that has a lowest initial sum excess processing load. The methods and systems may then retrieve an updated estimated processing load for at least one of the plurality of processing requests and determine a secondary allocation of the plurality of processing requests to the plurality of available domains.

Abstract: Methods and systems for predicting successful data transmission during mass communications across computer networks featuring disparate entities and imbalanced data sets using machine learning models. For example, the methods and systems provide a prediction as to whether or not a communication will be successful prior to the transmission being sent. Moreover, in some embodiments, the methods and systems described herein provide probability of a successful transmission as a function of time. For example, the methods and system provide a probability of how likely a communication will succeed (or fail) if it is sent at various times. Additionally, in some embodiments, the methods and systems may alert a sender prior to the transmission of a communication that the transmission is likely to succeed or fail.

Abstract: A dataset management system organizes datasets using a data relationship graph that serves as a representation of datasets that are related to one another. For example, the data relationship graph includes nodes that each represent a dataset as well as edges that each connect two nodes. Each edge represents a similarity in characteristics of the similar datasets such as a similarity in the datasets' origin, organizational schema, lineage, or data values. When a user is to be provided recommended datasets, the dataset management system identifies candidate datasets by traversing the nodes and edges of the data relationship graph. Amongst these candidate datasets, the dataset management system evaluates users that have accessed the candidate datasets as well as the context in which the candidate datasets were accessed to identify recommended datasets for presentation to the user.

Abstract: The system uses the non-repudiatory persistence of blockchain technology to store all task statuses and results across the distributed computer network in an immutable blockchain database. Coupled with the resiliency of the stored data, the system may determine a sequence of processing tasks for a given processing request and use the sequence to detect and/or predict failures. Accordingly, in the event of a detected system failure, the system may recover the results prior to the failure, minimizing disruptions to processing the request and improving hardware resiliency.

Abstract: Methods and systems are described for implementing automated controls assessment through an application programming interface (“API”) driven software development kit. For example, the system may receive a response from an API-based agent to an automated controls assessment audit. The system may process the response, using a library of reusable features for controls assessment audits for a plurality of computer domains, to generate a result of the automated controls assessment audit. The system may then generate an outcome of the first automated controls assessment audit.

Abstract: A central service provider manages verifiable and cryptographically strong transactions in a block chain network. For each transaction, the central service provider maintains and updates one or more directed graphs that describe the traceable positions of an asset held by parties of the block chain network. Therefore, transactions can be reconciled by tracing along the directed graphs. The central service provider also leverages crypto-hashing to enforce the immutability of each executed transaction. For example, each transaction is cryptographically hashed and written into a block in the block chain. Each transaction refers to a cryptographic hash of a prior transaction and each block refers to a cryptographic hash of a prior block. Therefore, the executed transactions are cryptographically strong in that if a single transaction is altered, then the entire block of transactions including the transaction is invalidated.

Abstract: In some embodiments, a request may be received from a customer via an interface, wherein the request indicates (i) a social media stream associated with the customer to be monitored for instances of investment triggering content (ITC). A customer account of the customer may be configured based on the request. The social media stream may be monitored for instances of ITC based on the customer account configuration. Responsive to the monitoring detecting an instance of ITC, a first amount may be withdrawn from the customer account, and a purchase of a quantity of shares of an investment fund may be executed on behalf of the customer using the withdrawn first amount.

Abstract: Methods and system are described for optimizing data processing in cloud-based, machine learning environments. For example, through the use of a machine learning model utilizing a self organizing map and/or the use of specific processing nodes in a computer system to perform specific tasks the methods and system may more efficiently distribute tasks through a cloud computing environment and increase overall processing speeds despite increasing amounts of data. The methods and system described herein are particularly related to collateral allocation computer systems that automate the management of numerous collateral assets. For example, as the amount of collateral assets and the complexity of given transactions grow, typical allocation systems face frequent processing delays related to collateral allocations (e.g., allocations of collateral associated with Tri-Party Repos).

Abstract: Methods and systems for predicting successful data transmission during mass communications across computer networks featuring disparate entities and imbalanced data sets using machine learning models. For example, the methods and systems provide a prediction as to whether or not a communication will be successful prior to the transmission being sent. Moreover, in some embodiments, the methods and systems described herein provide probability of a successful transmission as a function of time. For example, the methods and system provide a probability of how likely a communication will succeed (or fail) if it is sent at various times. Additionally, in some embodiments, the methods and systems may alert a sender prior to the transmission of a communication that the transmission is likely to succeed or fail.

Abstract: Methods and systems are described for improving hardware resiliency during serial processing tasks in distributed computer networks. In particular, the system uses may use the non-repudiatory persistence of blockchain technology to store all task statuses and results across the distributed computer network in an immutable blockchain database. Coupled with the resiliency of the stored data, the system may determine a sequence of processing tasks for a given processing request and use the sequence to detect and/or predict failures. Accordingly, in the event of a detected system failure, the system may recover the results prior to the failure, minimizing disruptions to processing the request and improving hardware resiliency.

Abstract: In certain embodiments, transaction-subset-assignment of computer processing nodes may be facilitated to perform collateral allocation. In some embodiments, a query may be performed to obtain a set of transactions. Computer processing nodes may be selected from a set of available nodes for performing collateral allocations. Collateral allocation rules associated with a lender may be obtained, and each of the computer processing nodes may be caused to perform collateral allocations for one subset of the transaction set in accordance with the collateral allocation rules by assigning transactions of the transaction set respectively to the computer processing nodes such that the computer processing nodes collectively perform collateral allocations for the transaction set.

Abstract: The present disclosure relates to a system and method for matching a signature extracted from an electronic document (e.g., extracted as .jpg, .gif .png, etc.) with one or more other signatures. The system determines various signature attributes including a number of black pixels in the extracted image, the slopes of four different lines that bound the signature in the extracted image, and a normalized width to height ratio. Two (or more) signatures are considered a match if the attributes match to within specified tolerances. Individual attributes have their own tolerances. These tolerances are adjustable independent of the other tolerances for other attributes.

Abstract: A system for analyzing liquidity data for a client of a financial services provider includes processor(s), at the financial services provider, configured to execute computer program module(s). The modules are configured to receive a request from the client to generate a liquidity analysis based on client account(s), and, upon receiving the request, execute on the processors of the system, module(s) configured to aggregate liquidity data from a plurality of financial subsystems of the financial services provider as aggregated liquidity data. The aggregated data includes cash flows in transactions associated with the client accounts in one or more of the financial subsystems. The modules are further configured to execute on the processors, module(s) configured to generate a graphical depiction of at least a portion of the aggregated liquidity data over a time interval, based on the request, and transmit, for displaying on an electronic display at the client, the graphical depiction.

Abstract: An online database system receives a tenant dataset from a tenant and automatically generates a mapping that can be used to convert the tenant dataset from a tenant schema to a database schema used by the online database system. To generate the mapping, the online database system analyzes individual entries of the tenant dataset received from the tenant and determines a likelihood of usage. The mapping maintains entries in the tenant dataset that are likely to be used as individual entries in the converted dataset and compresses entries that are less likely to be used into fewer entries to satisfy the database schema. The online database system converts the tenant dataset from the tenant schema to the database schema using the mapping.

Abstract: An API mediation system can allow an original service designed to interact with client systems over a first API to interact with other client systems over a second API without significant modifications to the original service. The API mediation system can automatically generate a mediation layer designed to translate communications to and from the original service between the first and second APIs. Similarly, the API mediation system can implement a mapping between functions of the original service through the first API and equivalent functions over the second API, including mapping security operations or other security methods between the first and second APIs. To generate the mapping layer, the API mediation system can generate a model of the original service outlining the underlying functionality of the original service.