Alerta

SYSTEM AND METHOD FOR SENTIMENT ANALYSIS BASED INTERACTIVE ENGAGEMENT USING MACHINE LEARNING

Absstract of: WO2026154498A1

An integrated system (106) for sentiment analysis using machine learning. The system (106) receives data associated with one or more users, where the data comprises one or more parameters associated with interaction among the one or 5 more users during a predefined schedule. The system (106) analyzes, the one or more parameters to determine one or more sentiments associated with the one or more users. The system (106) dynamically configures, via a machine learning engine (214), the predefined schedule based on the determined one or more sentiments. The system (106) enables, interaction among the one or more users 10 using the dynamically configured predefined schedule.

PROCESS ANALYSIS AND AUTOMATION THROUGH MACHINE LEARNING PROCESSING OF NETWORK DATA

Absstract of: US20260211637A1

0000 Automated generation of scripts includes receiving by a script generation engine, a user input prompt for a task and packages associated with the task, via a user interface. A machine learning engine parses the prompt into words representing a process associated with the task. The machine learning engine determines context from the words related to the task and associate the context with the packages. The machine learning engine identifies one or more sections of packages with context that are dynamic values within the process as potential variables. The script generation engine queries a retrieval of stored packages associated with the task. The machine learning engine replaces the dynamic values with a variable. The script generation engine generates a script configured to perform the task, wherein the script includes the variable as a placeholder for a user entry of the dynamic values.

SYSTEM AND METHOD FOR GENERATING FORMULATIONS USING MACHINE LEARNING ARCHITECTURES

Absstract of: US20260212164A1

0000 There is provided systems and methods for generating formulations with improved performance and lower resource consumption. The population of formulations may be created by a Differential Evolution (DE) process. In each successive generation, lower performing formulations may be replaced with new formulations predicted to have better performance by a modeling pipeline. The modeling pipeline may perform a search for ML architectures and hyperparameter optimization for a suitable ML architecture, and then train one or more machine learning models to minimize error (otherwise maximizing the resulting score for a formulation). The machine learning models may be an ensemble of Random Forest models. These new formulations may form part of the next generation in the DE process. This modified version of the Differential Evolution process using machine learning techniques may result in populations of formulations which are superior in performance, and/or require fewer iterations of generations to achieve formulations which meet performance objectives.

DATA-DRIVEN SIMULATION AND ANALYSIS FRAMEWORK FOR MULTI-MODAL URBAN MOBILITY SYSTEMS

Absstract of: US20260212082A1

0000 An apparatus in an illustrative embodiment comprises a processing platform that includes at least one processing device, the processing device comprising a processor coupled to a memory. The processing platform is configured to implement a machine learning based framework for at least one of simulation and analysis relating to multi-modal mobility, the machine learning based framework comprising a plurality of stages, including at least a population synthesis stage and a trip generation stage. The processing platform is further configured to execute a first machine learning model in the population synthesis stage, and to execute a second machine learning model in the trip generation stage, the second machine learning model being different than the first machine learning model. The first machine learning model illustratively comprises a traveler cluster classifier, and the second machine learning model illustratively comprises a mode-duration choice model.

CONTROLLING POWER TO AN IMPLANTABLE DEVICE

Absstract of: US20260213010A1

0000 Presented herein are techniques for predicting and controlling power for an implantable device. The power prediction for the implantable device may be performed independent of real-time power information from the implantable device, and may utilize artificial intelligence (AI) or machine learning.

PREDICTION OF THE DUCTILE-TO-BRITTLE TRANSITION TEMPERATURE OF POLYMER COMPOSITIONS BASED ON IMPACT CURVES

Absstract of: US20260212964A1

The present invention relates to a computer implemented method for predicting the ductile-to-brittle transition temperature of a test polymer composition based on impact curves. Furthermore, a non-transitory computer readable storage medium is provided for tangibly storing computer program instructions capable of being executed by a processor, the computer program instructions defining the steps the aforementioned computer implemented method. Furthermore, the invention is directed to the use of impact curves and values indicative of the ductile-to-brittle transition temperatures of polymer compositions for training machine learning algorithms.

SYSTEM AND METHOD FOR INTERACTIVE ENGAGEMENT OF USERS USING MACHINE LEARNING BASED ON ENVIRONMENTAL FACTORS

Absstract of: WO2026154499A1

An integrated system (106) for interactive engagement using machine learning based on environmental factors. The system (106) receives data associated with one or more users, where the data comprises information associated with interaction among the one or more users during one or more activities for a predefined schedule. The system (106) analyzes, the data to determine one or more environmental factors associated with the interaction of a plurality of users among the one or more users during the one or more activities. The system (106) dynamically configures, via a machine learning engine (214), the one or more activities for the predefined schedule based on the determined one or more environmental factors. The system (106) enables, interaction among the plurality of users using the dynamically configured one or more activities for the predefined schedule.

COMPUTING SYSTEM AND METHOD FOR CREATING MACHINE-LEARNING MODELS UTILIZING A MANAGED CREW OF AI AGENTS

Absstract of: US20260212212A1

0000 A computing system is configured to: (i) utilize an orchestrator AI agent that is configured to use a generative AI model to decompose a prompt for generating a machine-learning model into modelling tasks, generate instructions for a set of modelling AI agents, pass the instructions for the set of modelling AI agents to the modelling AI agents, and receive modelling output from the modelling AI agents, (iii) provide the instructions for the modelling AI agents to the modelling AI agents that are each configured to utilize a generative AI model to, based on the set of instructions, perform a modelling function corresponding to a modelling task and thereby generate modelling output corresponding to the modelling task, and pass the modelling output to the orchestrator AI agent, and (iv) create the machine-learning model based on the modelling output received by the orchestrator AI agent.

MULTI-AGENT FRAMEWORK FOR SYNTHETIC DATA GENERATION AND VERIFICATION

Absstract of: WO2026155825A2

The disclosure includes a computer-implemented method including generating a dataset of questions that mimic user queries by a multi-agent framework, wherein the multi-agent framework includes deployment of a plurality of artificial intelligence (AI) agents, running a subset of the dataset of questions through a first agent of the plurality of AI agents where each running of a question produces a trajectory which is a sequence of tools that the first agent calls, selecting a most common trajectory (MCT) for each question across all runnings, wherein the MCT for each question is annotated to indicate a correctness of the MCT, reverse engineering alternate questions from the responses for trajectory verification, extracting features from the MCT for each question that has been annotated and the alternate questions, and training a discriminative machine learning model on the features.

METHOD AND SYSTEM FOR IMPROVING OVERALL QUALITY AND SCOPE OF DISTINCT ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING MODLES

Absstract of: US20260212163A1

A system is presented that optimally integrates a plurality of distinct artificial intelligence and machine learning (AI/ML) models into a cohesive AI/ML paradigm that improves an overall quality and scope of the plurality of distinct AI/ML models. The system may be configured to receive an input via an interface of at least one processor; instantiate a tokenized output of the at least one processor; establish, with the at least one processor, a plurality of connections that respectively correspond to and communicate with each AI/ML model from among the plurality of distinct AI/ML models; perform, by the at least one processor, an iterative sequence that utilizes the plurality of connections to populate the tokenized output based on the input; and transmit the tokenized output to the interface.

METHOD FOR PRODUCING AN EVALUATION MODEL FOR DETECTING A SEAT OCCUPANCY STATE OF A SEAT ARRANGEMENT

Absstract of: US20260208743A1

Described is a method for producing a machine learning model for automated detection of a seat occupancy state of a seat arrangement. Parameters are assigned to possible seat occupancy states, and hyperparameters are configured to be adjusted on the basis of a metric. A detection accuracy is determined indicating a discrepancy between the seat occupancy state assigned to the parameters and an evaluation result supplied by the evaluation model with the provided hyperparameters. A metric is evaluated which takes into account a difference between the detection accuracy and a target value to output a value for the determined detection accuracy. Hyperparameters are adjusted appropriately, where the metric is optimized in order to obtain an optimum from the output value and to adjust the hyperparameters in such a way for which the metric reaches the optimum. The evaluation model is produced with the adjusted hyperparameters for further training.

AI ALGORITHM HARDWARE ACCELERATOR FOR ADAPTIVE SLEEP MONITORING IN SMART MATTRESSES

Absstract of: US20260207126A1

An artificial intelligence algorithm hardware accelerator designed for adaptive sleep monitoring in smart mattresses is disclosed. The system includes a data input module that receives and preprocesses physiological data from various sensors, utilizing an analog-to-digital converter and signal normalization circuit. A deep learning acceleration module based on FPGA or ASIC architecture is employed to expedite the computation of deep learning algorithms, such as CNNs and LSTMs, through parallel computing and high-throughput data processing. The strategy generation acceleration module uses reinforcement learning algorithms to optimize mattress adjustment strategies, focusing on parameters like height, softness, hardness, and temperature. A control and adjustment module transmits optimized instructions to mattress adjustment devices with low-latency communication, while a low-power optimization module dynamically manages power supply based on computing load, incorporating a voltage regulator, frequency governor, and power monitor.

SCENARIO-BASED CYBER NETWORK TOPOLOGY GENERATION SYSTEM AND METHOD

Absstract of: WO2026156353A1

A method, computer program product, and computer system for processing, using a machine learning model, an input description of a target virtual network topology. A plurality of network topology parameters are generated using the input description by processing, on the machine learning model, a plurality of prompts defining the plurality of network topology parameters from the input description. A plurality of network nodes are generated, using the machine learning model, from a plurality of network schemas. A virtual network configuration file is generated based upon, at least in part, the input description, the plurality of network topology parameters, and the plurality of network nodes.

HYBRID DEEP LEARNING FRAMEWORK FOR FLOOD SUSCEPTIBILITY MAPPING

Absstract of: US20260212194A1

0000 A method for training deep neural network for flood susceptibility prediction includes receiving training data including flood conditioning factor data for spatial units of geographic region together with ground truth data. The method includes inputting training data into deep neural network. The method includes generating plurality of candidate solutions, each candidate solution including parameter vector of weights and biases encoding relationships between flood conditioning factors and flood susceptibility outcomes. Further, the method includes refining each retained candidate solution by applying local search optimization process to minimize cost function computed relative to ground truth data. The method further includes selecting trained set of weights and biases for deep neural network from among refined solutions. Further, the method includes applying trained set of weights and biases to deep neural network to generate trained deep neural network configured to produce flood susceptibility probabilities for input flood conditioning factor data.

DEEP LEARNING USING A READ SKIP MASK COMMAND

Absstract of: US20260211580A1

0000 A method for operating a model on a storage array includes receiving, from a host, a registration for the model to cause the model to be stored in the storage array. The method also includes receiving a command and, in response to receiving the command, activating the model. The method further includes transmitting a tag to notify the host that the model is activated. In addition, the method includes receiving a prompt from the host that includes an identifier for relevant data stored on the storage array and instructions to form an inference based on the relevant data. Moreover, the method includes forming, based on the prompt, the inference using the model, where the model is executed on a computing resource of the storage array and the relevant data is used as an input for the model. Further, the method includes transmitting the inference to the host.

MULTI-AGENT SIMULATOR PLATFORM

Absstract of: WO2026156238A1

A computing platform can generate a virtual world and a set of virtual agents having agent traits and interaction rules. For example, by automatically analyzing experimenter queries, the platform can bind the agents to machine learning models that perform tasks responsive to the questions. The platform can generate additional hypotheticals to enable chains of simulations. The platform can execute a first simulation session by causing the set of agents to generate a first output set. Using an additional data source to generate new world characteristics, such as environmental and/or temporal conditions, the platform can generate and apply agent evolution logic for subsequent simulations, such as by performing mutation of agent traits, selection of agents, crossover of agent characteristics, and/or deletion of agents that do not meet a fitness threshold. The platform can persist agent states and outputs to generate user-interactive visualizations, aggregations, and training data sets for training agent models.

SELECTIVE EDGE AUGMENTED GENERATION FOR DISTRIBUTED SYSTEMS

Absstract of: US20260212214A1

0000 Methods and systems for managing operation of a distributed system are disclosed. To do so, a prompt may be serviced when submitted to a first generative trained machine learning model hosted by a management system. The management system may obtain a plurality of second prompts, identify at least one topic present in one of the plurality of second prompts, and filter edge devices in the distributed system based on the at least one topic to identify a portion of the edge devices. A Plurality of first responses from the portion of the edge devices may be obtained based on a first retrieval augmented generation (RAG) processing using locally available data hosted by the portion of the edge devices. Using at least the plurality of first responses as knowledge sources during a second RAG processing, a final response may be obtained and used to manage operation of the distributed system.

USE OF MACHINE VISION AND ML MODEL TO INTERPRET MAP DATA

Absstract of: US20260210731A1

0000 A method includes a server computer obtaining location data and time data associated with a transporter user device of a transporter that travels from a first location to a second location during a journey. The server computer determines points along the journey using the location data and the time data. The server computer visually differentiates the points along the journey according to a predetermined criteria. The server computer creates a map showing the journey and the visually differentiated points. The server computer can input, into a machine learning model, the map showing the journey and the visually differentiated points to classify the journey.

COMPUTER-IMPLEMENTED METHOD, DATA PROCESSING ARRANGEMENT AND COMPUTER PROGRAM PRODUCT FOR ENABLING AN AUTOMATED ACTION

Absstract of: WO2026153657A1

The present invention relates to a computer-implemented method (50) for enabling an automated action, comprising: providing a rule data set having enabling rules in at least one block of a blockchain (52), and retrieving the rule data set by means of a blockchain read-out device (54), characterized by evaluation of the rule data set by means of an agent device (56), the enabling rules and a machine learning method being taken into account for an enabling decision (58) for the automated action. The present invention further relates to a corresponding data processing arrangement and to a corresponding computer program product.

METHODS AND SYSTEMS FOR ENCODER PARAMETER SETTING OPTIMIZATION

Absstract of: US20260214134A1

0000 Methods and systems are described for encoder parameter setting optimization. A media item to be provided to users of a platform is identified. A request for content is received from a client device, and a media item associated with the content is identified. An indication of the media item is provided as input to a machine learning model. Outputs are obtained from the model identifying one or more sets of encoder parameter settings and, for each set, a confidence level that the settings satisfy a performance criterion based on the media item's media class. Based on the model outputs, at least one set of encoder parameter settings having a confidence level satisfying a confidence criterion is identified. The media item is encoded using the identified encoder parameter settings and provided for presentation via the client device.

METHOD AND APPARATUS FOR ADAPTING MACHINE LEARNING TO CHANGES IN USER INTEREST

Absstract of: US20260212225A1

0000 Aspects of the subject disclosure may include, for example, assigning a first interest measure associated with a first input to a learning machine at a first cycle, determining a first intelligence level according to a first product of the first interest measure and a first performance level based on the first input, and responsive to receiving a subsequent input at a subsequent cycle, reducing the first interest measure associated with the first input at the first cycle of the learning machine according to a total number of cycles that have occurred since the first cycle, assigning a new interest measure to the subsequent input at the subsequent cycle, generating a subsequent performance level according to the subsequent input, and determining a subsequent intelligence level according to a second product of the subsequent interest level and a subsequent performance level. Other embodiments are disclosed.

SYSTEM AND METHOD FOR INTERACTIVE ENGAGEMENT OF USERS USING MACHINE LEARNING

Absstract of: WO2026154497A1

An integrated system (106) for user engagement. The system (106) receives data associated with one or more users. The system (106) extracts a set of attributes from the received data to authenticate the users. The system (106) analyze via a machine 5 learning engine (214), the set of attributes to classify the authenticated users into one or more categories. The system (106) generates a predefined schedule associated with the categories to enable participation of the authenticated users from each of the categories in at least an activity for a predetermined period. The system (106) enables communication between a plurality of users of the authenticated users 10 classified into each of the categories during the predetermined period. The system (106) records information associated with the communication between the plurality of users of each of the categories during the predetermined period.

EDGE AUGMENTED GENERATION FOR DISTRIBUTED SYSTEMS

Absstract of: US20260211951A1

0000 Methods and systems for managing operation of a distributed system are disclosed. A prompt may be submitted to a first generative trained machine learning model that may be hosted by a management system. To service the prompt, a plurality of second prompts may be obtained by the management system. A first retrieval augmented generation (RAG) may be initiated by the management system for processing of the plurality of the second prompts by the edge devices using locally available data hosted by the edge devices as knowledge sources for the first RAG processing to obtain a plurality of first responses from the edge devices. A second RAG may be performed for processing of the prompt using the plurality of first responses as the knowledge source for the second RAG processing to obtain a final response. Computer implemented services may then be provided by the management system using the final response.

Wearable-Based Platform for Collaborative Crowd-Sourced Media Insights and Predictions

Absstract of: US20260212614A1

0000 A system for crowd-sourced media insights and predictive analysis is disclosed. Electronic devices capture media representing a real-world environment and associate it with user-generated insights. A processing unit converts the media into unique identifiers to store and retrieve these insights from a database, enabling users to access shared knowledge by capturing similar content. Additionally, the system analyzes live-streamed video using machine learning to predict subsequent events and transmit alerts or predictive insights to the user in real-time.

MANAGING ASSIGNMENT OF TASKS FOR UPDATING SOFTWARE BASED ON CHANGE MAGNITUDES

Nº publicación: US20260211657A1 23/07/2026

Applicant:

DELL PRODUCTS LP [US]
Dell Products L.P.

Absstract of: US20260211657A1

Methods and systems for providing computer-implemented services using a data processing system are disclosed. To provide the computer-implemented services, a set of tasks to implement a modification to a codebase for a portion of software may be obtained. The set of tasks and at least one trained machine learning model may be used to estimate change magnitudes of changes to the codebase required to complete the modification. The estimated change magnitudes and a schema for assigning tasks to entities based on degrees of relevance of knowledge bases of the entities may be used to identify entities to which the tasks are to be assigned. The tasks may be assigned to the identified entities to initiate an update process for the software and to obtain an updated version of the software. The updated version of the software may be deployed to initiate provisioning of desired computer-implemented services.