Abstract: A system for generating and inferencing using enterprise knowledge graphs is provided. The system receives first input data related to one or more entities from one or more data sources. The system extracts a first set of data components from the first input data and determines, based upon the extracted data components, a second set of data components. The system identifies one or more relationships between the first set of data components and the second set of data components and generate a knowledge graph comprising a plurality of nodes. A first node of the knowledge graph can represent a first respective data component of the first set of data components and a second node of the knowledge graph can represent a second respective data component of the second set of data components. The first node can be associated with the second node based on an identified relationship between the nodes.

Abstract: Provided herein are methods and systems for identifying a related data values in a plurality of datasets. The method can include extracting data from the plurality of datasets, identifying potential linking categories, and determining a validity of the potential linking categories. One or more linking categories can be selected from the potential linking categories based on the validity, and related data values can be identified between the plurality of datasets based on the selected linking categories. Also provided herein are methods and systems for classifying data values as reconciled or non-reconciled. The method can include identifying a plurality of related data values in a plurality of datasets. For each of the plurality of related data values, a similarity score and confidence score can be determined, and the plurality of related data values can be classified as reconciled or non-reconciled based on the similarity score and/or the confidence score.

Abstract: Provided herein are methods and systems for determining a roll forward amount. The method can include identifying a starting balance, extracting reconciled data values from a database, each reconciled data value having a confidence score that indicates a confidence level that a reconciled data value of a first dataset is the same as a corresponding reconciled data value of a second dataset. The method can include identifying a first subset of the reconciled data values that are a first type and a second subset of the reconciled data values that are a second type, and determining a roll forward amount based on the starting balance, the first subset, and the second subset.

Abstract: Provided herein are systems and methods for decentralized privacy preserving audits comprising receiving by a first local system, from an orchestrator system, instructions to perform a local audit; executing, by the first local system, in response to receiving the instructions to perform the local audit, a first local audit model epoch, wherein executing the first local audit model epoch comprises: analyzing information from one or more data sources from a local data set; receiving first input information from a second local system through a privacy preserving communication framework in response to requesting the first input information; determining whether information from the one or more data sources from the local data set is reconciled based on the first input information; and in accordance with a determination that the information from the one or more data sources from the local data set is not reconciled, executing a second local audit model epoch.

Abstract: An exemplary system for constructing data structures that can perform inferential reasoning to answer input queries may receive input data, extract and cluster entities in the input data into topic clusters, and for a first topic cluster construct a data structure comprising a plurality of nodes, wherein nodes of the data structure respectively represent a topic entity extracted from the input data and grouped into the first topic cluster, and wherein a first node of the data structure is associated with a second node of the data structure based on the first node and the second node respectively representing a first topic entity and a second topic entity associated in the input data with a common one of the one or more identified linguistic modalities. An exemplary system comprising the data structure may receive an input query and generate a response to the input query using the data structure.

Abstract: Disclosed herein are methods and systems for generating a merged dataset, comprising: accessing data comprising a core dataset and an additional dataset; identifying a plurality of common attributes between the core dataset and the additional dataset; determining a plurality of similarity scores between an inquiring entity in the core dataset and a plurality of candidate entities in the additional dataset, including, for each candidate entity of the plurality of candidate entities: calculating a similarity score for the candidate entity based at least in part on a distance-based score and a weight influence score; selecting one or more matches for the inquiring entity in the core dataset from the plurality of candidate entities in the additional dataset based at least in part on the plurality of similarity scores; and generating the merged dataset by adding the one or more selected matches for the inquiring entity to the core dataset.

Abstract: A first step in training a deep learning model may include generating data representing a plurality of historical episodes. Each historical episode may be divided into a sequence of time units, and historical information may be associated with each time unit. Next, for each historical episode of the plurality of episodes, a respective training action sequence may be generated using an evolutionary algorithm. A training data set comprising a plurality of training data points may then be generated. Each of the plurality of training data points may comprise an action extracted from a training action sequence generated by the evolutionary algorithm. The deep learning model may be trained using training data set to generate future actions to be executed at current or future time units.

Abstract: Provided herein are methods and systems for generating a chatbot. In response to a user selection of a content repository, the method can include generating a hierarchy map based on the selected content repository, translating the hierarchy map into a series of chatbot prompts, and generating a chatbot based on the series of chatbot prompts. The chatbot can be configured to guide the user to selected content within the content repository.

Abstract: The present disclosure relates generally to storing computer models, and more specifically to a platform for achieving replicability of a computer model (e.g., a trained machine-learning algorithm) by storing and providing access to data associated with the computer model using an immutable and decentralized ledger system (e.g., a blockchain ledger) and a distributed database. An exemplary computer-enabled method for storing a computer model, the method comprises: receiving data associated with the computer model; generating one or more asset files based on the data associated with the computer model; generating one or more hash values corresponding to the one or more asset files; generating one or more of location trackers corresponding to the one or more asset files; generating a ledger entry comprising the one or more hash values and the one or more location trackers; and adding the ledger entry to a blockchain ledger.

Abstract: A system for generating digital flowcharts is provided. The system receives sketch image data comprising a plurality of shapes and text, and processes the sketch image data to generate flowchart data by applying a first model configured to generate shape data, applying a second model configured to generate text data, and generating linking data that associates shape data and text data. The system may generate and display a visualization of the flowchart data. The system may map the flowchart data to a region of a presentation slide and display a visualization of the flowchart data on the presentation slide.

Abstract: Systems and methods for generating and utilizing an interactive causal loop diagram using a causal loop designer are provided. In one or more examples, a computer-implemented method for creating a causal loop diagram comprising visually emphasized elements can comprise displaying a first element and a second element, wherein the first element and the second element comprise visually emphasizable elements. In response to receiving a user command to connect the first element to the second element, the method can comprise displaying a connection between the first element and the second element, wherein the connection comprises a visually emphasizable element. In one or more examples, the method comprises displaying a causal loop diagram comprising the first element, the second element, and the connection. In response to a user activating a visual emphasis tool, the method can comprise visually emphasizing one or more of the visually emphasizable elements.

Abstract: Provided are asset tracking systems for determining a location of an asset, comprising a mobile device comprising one or more detection antennas and a transmitter, and a device associated with an asset and emitting Bluetooth low energy signals, wherein the system is configured to: detect a plurality of electromagnetic signals of the environment and one or more Bluetooth low energy signals from the one or more Bluetooth low energy devices; generate a signal profile based on the plurality of electromagnetic signals; determine, based on a comparison of the signal profile to data of signal profiles at a plurality of locations in the environment, a mobile-device location in the environment; and determine, based on the mobile-device location and based on the one or more Bluetooth low energy signals, one or more asset locations in the environment.

Abstract: Systems and methods for secure and accountable execution of computer scripts are disclosed. A system for validating an execution of a set of computer instructions may be configured to receive a first cryptographic hash, the first cryptographic hash corresponding to the set of computer instructions, to receive a second cryptographic hash, the second cryptographic hash corresponding to a runtime utility, wherein the runtime utility is configured to execute the set of computer instructions, to generate a ledger entry comprising the first cryptographic hash, the second cryptographic hash, and an indicator of success, and to add the ledger entry to a blockchain ledger, wherein the blockchain ledger is configured to receive the ledger entry from an authenticated node.

Abstract: Described are system, method, and computer-program product embodiments for developing an object detection model. The object detection model may detect a physical object in an image of a real world environment. A system can automatically generate a plurality of synthetic images. The synthetic images can be generated by randomly selecting parameters of the environmental features, camera intrinsics, and a target object. The system may automatically annotate the synthetic images to identify the target object. In some embodiments, the annotations can include information about the target object determined at the time the synthetic images are generated. The object detection model can be trained to detect the physical object using the annotated synthetic images. The trained object detection model can be validated and tested using at least one image of a real world environment. The image(s) of the real world environment may or may not include the physical object.

Abstract: The present disclosure relates generally to storing computer models, and more specifically to a platform for achieving replicability of a computer model (e.g., a trained machine-learning algorithm) by storing and providing access to data associated with the computer model using an immutable and decentralized ledger system (e.g., a blockchain ledger) and a distributed database. An exemplary computer-enabled method for storing a computer model, the method comprises: receiving data associated with the computer model; generating one or more asset files based on the data associated with the computer model; generating one or more hash values corresponding to the one or more asset files; generating one or more of location trackers corresponding to the one or more asset files; generating a ledger entry comprising the one or more hash values and the one or more location trackers; and adding the ledger entry to a blockchain ledger.

Abstract: Methods and systems for artificial intelligence (AI)-assisted document annotation and training of machine learning-based models for document data extraction are described. The methods and systems described herein take advantage of a continuous machine learning approach to create document processing pipelines that provide accurate and efficient data extraction from documents that include structured text, semi-structured text, unstructured text, or any combination thereof.

Abstract: An indoor geolocation system for determining a location in three-dimensional space includes a plurality of base stations and a mobile device movable about an indoor environment in three dimensions. The mobile device detects electromagnetic signals in the indoor environment emitted by devices other than the base stations, and generates a signal profile based on the signals. The mobile device transmits the signal profile to one or more of the base stations, which forward the signal profile to a remote server. The system determines a location of the in three-dimensional space of the mobile device by comparing the signal profile to data regarding signal profiles at a plurality of locations in the indoor environment. The data regarding signal profiles in the indoor environment may have been captured by a detection device other than the mobile device at a time prior to the detection of the electromagnetic signals by the mobile device.

Abstract: Methods and systems for artificial intelligence (Al)-assisted document annotation and training of machine learning-based models for document data extraction are described. The methods and systems described herein take advantage of a continuous machine learning approach to create document processing pipelines that provide accurate and efficient data extraction from documents that include structured text, semi-structured text, unstructured text, or any combination thereof.

Abstract: Methods and systems for artificial intelligence (AI)-assisted document annotation and training of machine learning-based models for document data extraction are described. The methods and systems described herein take advantage of a continuous machine learning approach to create document processing pipelines that provide accurate and efficient data extraction from documents that include structured text, semi-structured text, unstructured text, or any combination thereof.

Abstract: Described are methods and systems for calibrating simulation models to generate digital twins for physical entities. In some embodiments, a method includes receiving a plurality of datasets for a plurality of corresponding physical entities. A calibration request is enqueued to a calibration requests queue for each received dataset and includes information indicating a dataset and a corresponding physical entity. A plurality of calibration engines and a plurality of corresponding simulation clusters for generating a plurality of calibration results for a plurality of calibration requests dequeued from the calibration requests queue can be deployed.