Resumen de: US2023091076A1

Hybrid machine-learning systems and methods can be used to perform automatic keyphrase extraction from input text, such as historical records. For example, a computer-implemented method for extracting keyphrases from input text can include receiving input text having a plurality of words and identifying a set of candidate phrases from the plurality of words and a score for each of the candidate phrases using one or more unsupervised machine-learning models. The method can also include identifying named entities from the set of candidate phrases using one or more supervised machine-learning models and determining an updated set of scores for at least some of the candidate phrases within the set based on the named entities identified using the supervised machine-learning model. The method can also include identifying a keyphrase from the set of candidate phrases based on the updated set of scores.

Resumen de: AU2023201031A1

SYSTEMS AND METHODS FOR IMPROVED CORE SAMPLE ANALYSIS Provided herein are methods and systems for improved core sample analysis. At least one image of a core sample may be analyzed to determine structural data associated with the core sample (e.g., attributes of the core sample). A machine learning model may analyze the at least one image and determine one or more attributes associated with the core sample. The machine learning model may generate a segmentation mask. An output image may be generated. A user may interact with the output image and provide one or more user edits. The one or more user edits may be provided to the machine learning model for optimization thereof.

Resumen de: US2023085684A1

A method of recommending data, a device, and a medium, which relate to a field of an artificial intelligence technology, in particular to fields of deep learning, natural language processing and intelligent recommendation technologies. The method of recommending the data includes: acquiring operation data of an operation object, and the operation data is associated with first content data and first target object data; determining an operation object feature, a content feature and a target object feature based on the operation data; determining a fusion feature based on the operation object feature and the content feature; and recommending second content data and second target object data in an associated manner based on the fusion feature and the target object feature.

Resumen de: US2023088561A1

A method of generating a synthetic training dataset for training a machine learning model using an original training dataset including a plurality of features includes selecting a feature ci of the original training dataset as a target vector yi, selecting remaining features of the original training dataset as a set of training input vectors X\i, where X\i includes all features of the training dataset other than a feature corresponding to the selected feature ci, and training a prediction model f(yi|X\i). The method generates an estimate y′i of the target vector yi by applying the prediction model to the set of training vectors X\i, and inserts a synthetic feature c′i corresponding to the estimate y′i of the target vector yi into a synthetic training dataset.

Resumen de: US2023088325A1

A computer system for assessing sound to analyze a user's sleep includes a processor configured to perform operations including: (i) storing sample sound data associated with a plurality of sample sleep events, the sample sound data including a plurality of sample characteristics each respectively associated with at least one sample sleep event of the plurality of sample sleep events; (ii) receiving, from a client device, subject sound data collected during a sleep interval; (iii) analyzing, using a machine learning algorithm, the subject sound data collected during the sleep interval; (iv) identifying, based upon the analyzing, a subject characteristic associated with the subject sound data; (v) comparing the subject characteristic with the plurality of sample characteristics; and (vi) determining, based upon the comparing, whether the subject characteristic substantially matches at least one sample characteristic to identify one or more subject sleep events occurring during the sleep interval.

Resumen de: US2023087026A1

A device may obtain user information associated with a user and first account information associated with the user. The device may determine, based on the user information, user employment information and may determine, based on the first account information, user compensation information. The device may process, using a first machine learning model, the user employment information and the user compensation information to determine predicted future user compensation information. The device may obtain second account information associated with the user and may determine, based on the second account information, new user compensation information. The device may determine whether the new user compensation information is consistent with the predicted future user compensation information. The device may determine a predicted reason for the new user compensation information not being consistent with the predicted future user compensation information. The device may cause, based on the predicted reason, at least one action to be performed.

Resumen de: US2023086724A1

Techniques for mining training data for use in training a dependency model are disclosed herein. In some embodiments, a computer-implemented method comprises: obtaining training data comprising a plurality of reference skill pairs, each reference skill pair comprising a corresponding first reference skill and a corresponding second reference skill, the plurality of reference skill pairs being included in the training data based on a co-occurrence of the corresponding first and second reference skills for each reference skill pair in the plurality of reference skill pairs, the co-occurrence comprising the corresponding first and second reference skills co-occurring for a same entity; and training a dependency model with a machine learning algorithm using the training data, the dependency model comprising a logistic regression model or a data gradient boosted decision tree (GBDT) model. The dependency model may then be used to identify corresponding dependency relations for a plurality of target skill pairs.

Resumen de: US2023086465A1

A method for rule-based composition of user interfaces. A machine-learned rule repository is established based on previously observed combinations of UI states, UI features, and user features. A classifier classifies users into segments. Each segment includes users for which a combination of user features and UI states are defined. A first machine learning model (MLM) estimates a user segment-content preference including preferred UI content. A second MLM estimates a seen content-seen content similarity UI content preferences estimated according to prior UI content a user has seen. Based on the UI state and based on the user ID, rule-based recipes are obtained. Each rule-based recipe specifies a corresponding UI content suitable for an interaction between the user and the interface. A selected rule-based recipe is selected from the rule-based recipes. Specific UI content specified by the selected rule-based recipe is obtained, and the interface is updated with the specific UI content.

Resumen de: US2023085786A1

Techniques for generating recommendations using machine learning with respect to semantic concepts defined in a knowledge graph. A hair profile is determined for a user based on inputs related to the user. Determining the hair profile includes extracting attributes of the user from the inputs using natural language processing, computer vision, or both, and identifying respective nodes for the extracted attributes in the knowledge graph. The knowledge graph is created via machine learning using population data including hair-related data in order to identify relationships between semantic concepts represented by nodes of the knowledge graph. The nodes include discrete properties such as individual hair attributes, ingredients of products, or otherwise discrete characteristics of factors that may affect a user's hair or related health conditions. A generalized recommendation is generated based on the hair profile. A personalized recommendation may be generated based on the generalized recommendation and progress logged by the user.

Resumen de: US2023085704A1

Systems and methods for dynamic demand sensing in a supply chain in which constantly-updated data is used to select a machine learning model or retrain a pre-selected machine learning model, for forecasting sales of a product at a specific location. The updated data includes product information and geographic information.

Resumen de: US2023085687A1

Various disclosed embodiments can resolve output inaccuracies produced by many machine learning models. Embodiments use content order as input to machine learning model systems so that they can process documents according to the position or rank of instances in a document or image. In this way, the model is less likely to misclassify or incorrectly detect instances or the ordering between predicted instances. The content order in various embodiments can be used as an additional signal to classify or make predictions.

Resumen de: US2023093444A1

A data processing method performed by an information processing device operating a machine learning algorithm. The machine learning algorithm jointly processes operational data originating from a real-life context and human mental brain activity data relating to implicit human participation with this context, provided by a passive brain-computer interface. The operational data and human mental brain activity data to be jointly processed are identified by sensing human mental engagement with an aspect of the context. Based on the processing analysis and inferencing by the machine learning algorithm and aspects of the context may be controlled and adapted by the information processing device. Included are a program product, a trained machine-learning model, and a data processing system.

Resumen de: US2023088431A1

In one embodiment, a method includes executing server operations for processing a data item, wherein the server operations are based on feature values associated with the data item, executing a heuristic to determine a heuristic result value based on the feature values, determining a prediction result by a machine-learning model based on the feature values and the heuristic result value, wherein the prediction result is associated with an assessment score indicating an effectiveness of the machine-learning model, invoking a feedback function configured to update the machine-learning model based on the prediction result and the assessment score, and updating the machine-learning model when resources on the server are available.

Resumen de: US2023088182A1

Provided are systems and methods directed to identifying relationships between colloquial place names in a relational database. In some embodiments, a method of identifying relationships between colloquial place names in a relational database comprises receiving geographic location information; generating a vector corresponding to the geographic location; comparing the geographic location information vector to a plurality of colloquial place name vectors in a relational database that maps a plurality of colloquial place names to a plurality of corresponding colloquial place name vectors in a vector space, to generate a plurality of similarity scores that is calculated based on the geographic location information vector and each colloquial place name vector of the plurality of colloquial place name vectors; and identifying that one or more colloquial place names in the relational database are related to the geographic location information based on the plurality of similarity scores.

Resumen de: US2023087969A1

A method for providing treatment recommendations for a patient to a physician is disclosed. The method includes receiving health information associated with the patient, determining a first risk score for the patient based on the health information using a trained predictor model, determining a second risk score for the patient based on the health information and at least one artificially closed care gap included in the health information using the predictor model, determining a predicted risk reduction score based on the first risk score and the second risk score, determining a patient classification based on the predicted risk reduction score, and outputting a report based on at least one of the first risk score, the second risk score, or the predicted risk reduction score.

Resumen de: US2023088436A1

A method of reducing a future amount of electronic fraud alerts includes receiving data detailing a financial transaction, inputting the data into a rules-based engine that generates an electronic fraud alert, transmitting the alert to a mobile device of a customer, and receiving from the mobile device customer feedback indicating that the alert was a false positive or otherwise erroneous. The method also includes inputting the data detailing the financial transaction into a machine learning program trained to (i) determine a reason why the false positive was generated, and (ii) then modify the rules-based engine to account for the reason why the false positive was generated, and to no longer generate electronic fraud alerts based upon (a) fact patterns similar to fact patterns of the financial transaction, or (b) data similar to the data detailing the financial transaction, to facilitate reducing an amount of future false positive fraud alerts.

Resumen de: US2023087672A1

The AI-Based Real-Time Prediction Engine Apparatuses, Methods and Systems (“AIRTPE”) transforms machine learning training input, order placement input inputs via AIRTPE components into machine learning training output, order placement output, information leakage alert outputs. An order placement datastructure associated with a security identifier is obtained. An order placement allocation for the security identifier is determined. An order placement request datastructure for a first order is sent to a server associated with a first venue. A set of trade tick data messages associated with the first venue is obtained. A set of inferred labels is determined for each obtained trade tick data message using a real-time prediction logic generated using a machine learning technique. The inferred labels of a selected inferred label type are grouped into buckets. When it is determined that the grouped inferred labels correspond to execution data generated by the first order, an information leakage alert is generated.

Resumen de: US2023087583A1

A system, platform, program product, and/or method for generating new composite insight templates that includes: running a machine learning model on a data set to obtain for each of a plurality of entities a risk score and feature-based insights; generating a list of top “n” features input to the machine learning model that contributes to the risk score for each entity; grouping entities based upon similar features input to the machine learning model that contributes to the risk score for each entity; generating a decision tree for at least one of the group of entities; extracting, from the decision tree generated for the at least one of the group of entities, one or more feature-based insights; generating, by applying subject matter input, a new composite insight based upon the one or more feature-based insights; and adding the new composite insight to insight templates.

Resumen de: US2023093371A1

A computer-implemented method may include: receiving financial information regarding a user; categorizing transaction information of the user based on the financial information; displaying the categorized transaction information of the user; receiving information regarding at least one financial preference and at least one transaction preference of the user; training a machine learning engine based on the at least one financial preference and at least one transaction preference of the user to determine one or more activities available to the user; calculating, for each of the one or more activities available to the user, an estimated influence on the at least one financial preference; displaying the estimated influence on the at least one financial preference based on a user selected one of the one or more activities available to the user; filtering the one or more activities available to the user with a positive estimated influence to the at least one financial preference; and presenting a recommendation of action relating to the one of the one or more activities available to the user, wherein the recommendation of action relating to the one of the one or more activities available to the user is presented by at least one of voice notification, application notification, tactile notification, or graphic notification.

Resumen de: US2023092866A1

Disclosed herein is a machine learning platform and system for data analysis including for purposes of providing digital evaluations and therapeutics.

Resumen de: US2023092716A1

A system and method for executing a record within an immutable sequential data structure, the system including a computing device, the computing device configured to transmit a communication to a remote device, receive a remark from the remote device, retrieve an input related to a user, wherein the input is stored as an encrypted proof-linked assertion on at least an immutable sequential data structure for authorized party access, generate a record as a function of the input, transmit the record to the remote device, and store an executed record within the at least an immutable sequential data structure.

Resumen de: WO2023041150A1

Embodiments herein disclose a method for facilitating a privacy-aware representation in a system. The method comprises determining using a content analyzer a layout of a scene. Thereafter, one or more objects in the scene is identified using the content analyzer to define a relationship between the objects. A privacy status to be tagged to the one or more identified objects is inferred by using a machine learning model. At least one object is processed based on the privacy status inferred. A privacy-aware representation of the scene is rendered, wherein the privacy-aware representation displays the scene with at least one processed object.

Resumen de: GB2610984A

An artificial intelligence (AI) platform to support a continuous integration and deployment (CI/CD) pipeline for software development and operations (DevOps). One or more dependency graphs are generated based on application artifacts. A machine learning (ML) model is leveraged to capture a relationship between components in the dependency graph (s) and one or more pipeline artifacts. Responsive a change of an application artifact, the captured relationship is leveraged to identify an impact of the detected change on the pipeline artifact (s). The CI/CD pipeline is selectively optimized and executed based on the identified impact to improve the efficiency of the pipeline and the deployment time.

Resumen de: WO2021231317A1

A method comprises: receiving data corresponding to a plurality of data categories of a patient; selecting, from a plurality of trained machine learning models and based on the plurality of data categories, a first machine learning model and a second machine learning model, the first machine learning model being trained using first data of a first subset of the plurality of data categories and having a first weight indicative of a first performance metric value, the second machine learning model being trained using second data of a second subset of the plurality of data categories and having a second weight indicative of a second performance metric value; generating a first prediction result and a second prediction result using, respectively, the first model and the second model; and generating a combined prediction result based on the first prediction result, the second prediction result, the first weight and the second weight.

Resumen de: WO2023036714A1

A computer-implemented method comprising: obtaining activity data indicative of an anomalous activity within a computer system; processing the activity data to generate confidence data representative of a set of confidence values, each confidence value representative of a confidence that the anomalous activity comprises a respective type of activity; and determining, based on at least the confidence data, mitigating action to take to mitigate the anomalous activity. Further examples relate to a computer system configured to implement an intrusion detection system and an intrusion response system, and to a computer-implemented method of calibrating a system comprising a machine learning model trained to generate output uncalibrated confidence data representative of a set of output uncalibrated confidence values.