Alerta

Improved machine learning systems

Absstract of: AU2025202625A1

Systems and methods for training a machine learning (ML) model to predict outcomes is disclosed. The ML model includes a plurality of ML sub-models and an ensemble model. The method includes: receiving a plurality of data records; creating the ML sub-models based on the data records; assigning at least a subset of the data records to each of the ML sub-models; training each ML sub-model using the assigned subset of data records, each sub-model trained to determine a predicted outcome based on a given user data record; providing the predicted outcomes generated by each of the sub-models to the ensemble model, the ensemble model trained to combine the predicted outcomes from the sub-models to obtain a combined predicted outcome; using the trained ML model to determine a predicted outcome for an individual data record; and reusing the determined predicted outcome for the individual data record to retrain the ML model. Systems and methods for training a machine learning (ML) model to predict outcomes is disclosed. The ML model includes a plurality of ML sub-models and an ensemble model. The method includes: receiving a plurality of data records; creating the ML sub-models based on the data records; assigning at least a subset of the data records to each of the ML sub-models; training each ML sub-model using the assigned subset of data records, each sub-model trained to determine a predicted outcome based on a given user data record; providing the predicted outcomes generated by e

ARTIFICIAL INTELLIGENCE (AI) ASSISTED DIGITAL DOCUMENTATION FOR DIGITAL ENGINEERING

Absstract of: AU2024214090A1

A digital documentation system for preparation of engineering documents utilizing one or more artificial intelligence (Al) algorithms is provided. The system includes a user interface for selecting and populating templates with data, and one or more Al algorithms for creating and recommending templates, and preparing documents based on the recommended templates. The system uses natural language processing and semantic analysis algorithms to understand the content of the templates, documents, and associated engineering data, and to generate and recommend relevant templates to the user based on user prompts. The system also uses machine learning and predictive modeling and decision-tree algorithms to assist with the preparation of documents, by generating suggestions for data fields and values based on the user's previous inputs and the overall context of the document and available engineering data, including model data and metadata from digital models accessed in a zero-trust framework.

Detecting Compatibility Mismatch by Generative Artificial Intelligence

Absstract of: US2025292309A1

Detecting compatibility mismatch by generative artificial intelligence is described. Compatibility data is obtained (e.g., by accessing a database). The compatibility data is associated with a compatibility between items (e.g., items and categories of vehicles or an item and another item) and includes a list of recommended compatibilities between the items and a user reported compatibility for at least one item. A machine learning model is generated for detecting a compatibility mismatch between a first item and a second item and/or between an item and a category of vehicle. At least a portion of the compatibility data is provided as input to generative artificial intelligence to generate the machine learning model. An update to the list of recommended compatibilities is determined based on the detected compatibility mismatch.

CYBERSECURITY EVENT DETECTION, ANALYSIS, AND INTEGRATION FROM MULTIPLE SOURCES

Absstract of: WO2025193502A1

The present disclosure presents methods and systems for determining cybersecurity risk exposure for entities. In one aspect, a method is provided that includes providing first text data to a trained LLM to identify data associated with a first candidate cybersecurity event for an entity, comparing the entity's identifier to domain information to verify the entity's identifier, determining if the first candidate cybersecurity event represents a new cybersecurity event based on com with previous data, and updating a cybersecurity risk score for the entity based on this determination. Further enhancements include training the LLM with cybersecurity event data, outputting documentation of the event source, and various methods for evaluating the novelty and severity of the cybersecurity event, including similarity measures and manual review triggers. The techniques leverage LLMs, machine learning models, and automated actions to provide a comprehensive approach to cybersecurity risk assessment and response. Other aspects are also provided.

AUTOMATED SENSING AND CONTROL SYSTEM WITH DATA ANALYTICS AND ARTIFICIAL INTELLIGENCE

Absstract of: WO2025193810A1

The present disclosure relates to training a machine learning model based on a dataset comprising sales volume, product distribution logistics records, and product manufacturing data. The present disclosure further relate to extracting from the model a prediction of at least one item selected from a group consisting of future sales volume of an existing product, consumer interest for new products, and failure rates for product distribution equipment.

FEED ENRICHMENT USING SELF-SUPERVISED MULTIMODAL LARGE LANGUAGE MODELS

Absstract of: WO2025193673A1

A system and method for enhancing data feed using a machine learning (ML) model are disclosed. In some embodiments, the method includes receiving multimodal data associated with a plurality of data items and providing, from the received multimodal data, a set of multimodal data samples to the ML model, each multimodal data sample associated with two or more modalities. The method also includes training the ML model using the set of multimodal data samples by optimizing a similarity value computed for each multimodal data sample based on whether the multimodal data sample is associated with a same data item or from different data items. The method further includes receiving new data associated with a new data item, the new data including one or more data components to be enriched, and automatically populating the one or more data components using the trained ML model.

METHOD AND SYSTEM FOR WRITE-PROTECTING DATA IN MIXED-MEDIA DATABASES

Absstract of: WO2025193266A1

Write protection can be provided in mixed-media datasets. Contextual details may be extracted from a set of media to form a mixed-media dataset. The mixed-media dataset may be used to train a machine-learning model. A request to modify the mixed-media dataset may be received causing the machine-learning model to determine if implementing the request to modify the mixed-media dataset will introduce conflict or a deviation from the current mixed-media dataset. Upon confirming that implementing the request will not introduce a conflict or deviate from the from the current mixed-media, the mixed-media dataset may be modified according to the request and the machine-learning model may be retrained using the modified mixed-media dataset.

SYSTEMS AND METHODS FOR AUTOMATED QUALIFICATION ANALYSIS

Absstract of: WO2025193249A1

Qualification decisioning systems and techniques are described. For instance, a system receives user information that is indicative of a user's income stream and/or asset(s). The system receives product qualification criteria data corresponding to products, with different products corresponding to different product-specific qualification criteria. The product qualification criteria data can change over time. The system dynamically analyzes the user information and the product qualification criteria data using a trained machine learning (ML) model in real-time as the user information and the product qualification criteria data continue to be received. The trained ML model identifies a subset of the plurality of products that the user qualifies for at a specific time. The system outputs recommendations for the subset of the plurality of products. The system dynamically trains the trained ML model further, using the recommendations and the user information as training data, to update the trained ML model for future qualification decisions.

AI-POWERED PLATFORM FOR GENERATION OF MATERIALS AND PREDICTION OF DESIRED PARAMETERS, CHARACTERISTICS, QUALITIES OR PROPERTIES THEREOF

Absstract of: WO2025189301A1

The invention encompasses innovative AI-powered platforms that predict various parameters (e.g. compostability, sustainability, material mechanical properties, formulate solutions properties, economic factors, etc.) and generates formulations for sustainable materials, while accounting for the various physical and non-physical properties as wells the applications they serve. The invention encompasses systematic search algorithms and/or machine learning techniques for identification of biomaterials that are predicted to have applicable properties. This invention encompasses a powerful method and tool to streamline the design and development of materials that are sustainable. In various embodiments, by employing advanced data processing, polymer chemistry knowledge and predictive modeling, the embodiments of the present invention provides precise and efficient recommendations, enhancing the research and development of sustainable biomaterials. Leveraging evolutionary and machine learning methodologies, the embodiments of the present invention identify polymer materials that adhere to real-world renewable constraints and other parameters, emphasizing a multitude of pertinent physical and non-physical properties.

METHODS FOR ESTIMATING AND SERVING WAKE WINDOW PREDICTIONS BASED ON SLEEP DATA

Absstract of: US2025292903A1

According to certain aspects of the present disclosure, systems and methods are disclosed for tracking and predicting the optimal wake windows of individual infants at scale and thus, the optimal sleep times of a user on a screen customized to each child prioritizing the use of highly personalized values based on machine learning to augment existing expert opinions of wake windows by age that will be automatically adaptive to the changes in child sleep and to generate predictions personalized for each child.

HUMAN INPAINTING UTILIZING A SEGMENTATION BRANCH FOR GENERATING AN INFILL SEGMENTATION MAP

Absstract of: US2025292377A1

The present disclosure relates to systems, methods, and non-transitory computer-readable media that modify digital images via scene-based editing using image understanding facilitated by artificial intelligence. For example, in one or more embodiments the disclosed systems utilize generative machine learning models to create modified digital images portraying human subjects. In particular, the disclosed systems generate modified digital images by performing infill modifications to complete a digital image or human inpainting for portions of a digital image that portrays a human. Moreover, in some embodiments, the disclosed systems perform reposing of subjects portrayed within a digital image to generate modified digital images. In addition, the disclosed systems in some embodiments perform facial expression transfer and facial expression animations to generate modified digital images or animations.

PREDICTION DEVICE, PREDICTION METHOD, AND PREDICTION PROGRAM

Absstract of: US2025292915A1

A prediction device includes: an acquisition unit that acquires chemical substance information of a drug and pharmacological information of the drug; an estimation unit that estimates estimated information of the drug by performing machine learning using the chemical substance information and the pharmacological information; and an output unit that predicts and outputs both efficacy and side effects of the drug on an organism by retraining a model of the machine learning on the basis of the estimated information.

Asset-Exchange Feedback In An Asset-Exchange Platform

Absstract of: US2025292296A1

An asset-exchange feedback system is implemented for performing asset-exchange feedback operations. The asset-exchange feedback system collects historical asset-listing data from an asset-exchange platform. The historical asset-listing data comprises, for each asset listing of a plurality of previous asset listings, a plurality of asset-listing attributes and a result of the asset listing. The asset-exchange feedback system uses a first machine learning model to determine, based on the historical asset-listing data, a first set of attribute-importance scores. Each attribute-importance score in the first set of attribute-importance scores corresponds to a respective asset-listing attribute in the plurality of asset-listing attributes and indicates an importance of the respective asset-listing attribute to one or more offerees participating in the asset-exchange platform. The asset-exchange feedback system performs an asset-exchange feedback operation based on the first set of attribute-importance scores.

MACHINE LEARNING TECHNIQUES TO CONSTRUCT AND APPLY HOME VALUATION MODELS THAT TAKE INTO ACCOUNT INFORMATION DERIVED FROM PHOTOGRAPHS OF HOMES

Absstract of: US2025292289A1

A home valuation facility is described. The facility accesses information about each of a plurality of homes sold in a geographic area during a distinguished period of time. The accessed information includes, for each home, a selling price for the home and one or more photos depicting the home. The facility uses the accessed information to train a statistical model for predicting the value of a home in the geographic area based on information about the home, including one or more photos depicting the home. The facility receives information about a distinguished home, including one or more photos depicting the distinguished home. The facility subjects the received information about the distinguished home to the trained statistical model to obtain a prediction of the distinguished home's value. The facility causes the obtained prediction of the distinguished home's value to be displayed together with information identifying the distinguished home.

IDENTIFYING AMBIGUOUS PATTERNS AS MALWARE USING GENERATIVE MACHINE LEARNING

Absstract of: US2025291921A1

A request is received to scan a package integration for a malicious dependency, the package integration to be integrated into an application. Using a known package cache, a subset dependencies of the package integration that have not been previously scanned is determined. Content of each file of the subset is input into a malware detection model, and an identification of an ambiguous pattern is received from the malware detection model. Responsive to receiving the identification of the ambiguous pattern, the ambiguous pattern is input into a severity model, and a level of severity that the ambiguous pattern would impose on an assumption that malware is present is received. Where the level of severity is above a threshold minimum level of severity, a query is transmitted to a generative machine learning model to determine whether malware is present.

DETECTING COMPATIBILITY MISMATCH BY GENERATIVE ARTIFICIAL INTELLIGENCE

Absstract of: EP4617989A1

Detecting compatibility mismatch by generative artificial intelligence is described. Compatibility data is obtained (e.g., by accessing a database). The compatibility data is associated with a compatibility between items (e.g., items and categories of vehicles or an item and another item) and includes a list of recommended compatibilities between the items and a user reported compatibility for at least one item. A machine learning model is generated for detecting a compatibility mismatch between a first item and a second item and/or between an item and a category of vehicle. At least a portion of the compatibility data is provided as input to generative artificial intelligence to generate the machine learning model. An update to the list of recommended compatibilities is determined based on the detected compatibility mismatch.

DIGITAL LIFESTYLE INTERVENTION SYSTEM USING MACHINE LEARNING AND REMOTE MONITORING DEVICES

Absstract of: WO2024102668A1

Systems and methods for a digital lifestyle intervention system using machine learning and remote monitoring devices is described herein. The disclosed systems can include a processor configured to train, based on historical user data, a personal machine learning model to generate an output indicative of a blood pressure prediction and lifestyle feature impact on blood pressure prediction. The trained personal machine learning model can be further configured to receive user data including blood pressure data for the user and generate output including lifestyle feature impact and a blood pressure prediction by applying the trained personal machine learning model to the received user data. At least one lifestyle recommendation can be generated based on the output of the trained personal machine learning model and/or output from a population model applied to the received user data. The at least one lifestyle recommendation can be provided to a user via a user interface.

MACHINE LEARNING TRAINING DURATION CONTROL

Absstract of: US2024152798A1

Some embodiments select a machine learning model training duration based at least in part on a fractal dimension calculated for a training data dataset. Model training durations are based on one or more characteristics of the data, such as a fractal dimension, a data distribution, or a spike count. Default long training durations are sometimes replaced by shorter durations without any loss of model accuracy. For instance, the time-to-detect for a model-based intrusion detection system is shortened by days in some circumstances. Model training is performed per a profile which specifies particular resources or particular entities, or both. Realistic test data is generated on demand. Test data generation allows the trained model to be exercised for demonstrations, or for scheduled confirmations of effective monitoring by a model-based security tool, without thereby altering the model's training.

HYBRID MACHINE LEARNING METHODS OF TRAINING AND USING MODELS TO PREDICT FORMULATION PROPERTIES

Absstract of: AU2023407504A1

Methods include training a machine learning module to predict one or more target product properties for a prospective chemical formulation, including (a) constructing or updating a training data set from one or more variable parameters; (b) performing feature selection on the training data set; (c) building one or more machine learning models using one or more model architectures; (d) validating the one or more machine learning models; (e) selecting at least one of the one or more machine learning models and generating prediction intervals; (g) interpreting the one or more machine learning models; and (h) determining if the one or more target product properties calculated are acceptable and deploying one or more trained machine learning models, or optimizing the one or more machine learning models by repeating steps (b) to (g). Methods also include application of trained machine learning modules to predict formulation properties from prospective data.

SOFTWARE ASSESSMENT TOOL FOR MIGRATING COMPUTING APPLICATIONS USING MACHINE LEARNING

Absstract of: US2024152869A1

A computing system includes a processor; and a memory having stored thereon instructions that, when executed by the one or more processors, cause the system to: receive content migration project parameters, resource migration project parameters and one or more services parameters of a user; scan a tenant computing environment; process the parameters by applying a multiplier display the costs, profits and pricing information. A method includes receiving content migration project parameters, resource migration projecting parameters and one or more services parameters of a user; scanning a tenant computing environment; processing the parameters by applying a multiplier displaying the costs, profits and pricing information. A non-transitory computer readable medium includes program instructions that when executed, cause a computer to: receive content migration project parameters, resource migration project parameters and one or more services parameters of a user; scan a tenant computing environment; process the parameters by applying a multiplier display the costs, profits and pricing information.

SEARCH REQUEST PROCESSING

Nº publicación: EP4617903A1 17/09/2025

Applicant:

AMADEUS SAS [FR]
Amadeus S.A.S

Absstract of: EP4617903A1

Method, systems and computer programs for handling search requests at a search platform are provided. The search platform determines, using a cache with a number of incomplete search results, one or more of the incomplete search results with first data fields that correspond to the least one search parameter. For each determined incomplete search result, the search platform generates at least one second data field using a machine learning model. The at least one second data field corresponds to at least one search parameter and the at least one first data field of each determined incomplete search result. The search platform assembles a number of completed search results on the basis of the determined incomplete search results and the generated at least one second data field and returns at least one of the completed search results.