2939results about "Biological neural network models" patented technology

Systems and methods for integrating generative artificial intelligence (AI) capabilities within Software as a Service (SaaS) platforms. One of the computer-implemented methods are for querying a generative AI model about structured data in a SaaS environment, enabling users to interact with and manipulate data through AI-assisted interfaces. One of the systems maintains a generative AI agent configured to interact with SaaS platform data as a virtual team member, capable of understanding context and nuances of project data. Also described are methods for color-context aware data analysis, generation of interactive elements in messaging sessions, and cross-application generative AI agent interactions triggered by user mentions. Also described is facilitating the creation of custom SaaS platform products by combining functionalities from existing products using generative AI. The systems and methods represent advancement in AI-driven SaaS customization and data analysis.

The invention relates to methods and systems for integrating generative artificial intelligence (AI) capabilities into Software as a Service (SaaS) platforms. It comprises maintaining AI agents with varying credentials, enabling their interaction with alphanumeric data in table structures, and implementing a hierarchical access control scheme. The system displays table structures, provides interfaces for user inputs, and allows AI agents to be added as platform users. The generative AI agents can analyze data, identify actions, and perform tasks autonomously. The invention also includes methods for proactive information gathering, interactive analysis of AI outputs, and management of AI resources as limited assets. This approach enhances SaaS functionality by enabling AI-driven task completion, data analysis, and decision-making while maintaining data security and user-specific access controls.

A data-driven dynamic energy management system for the adaptive coordination of renewable and conventional energy sources, consisting of: a processing unit configured to perform real-time calculations to optimize the generation, storage, and distribution of electrical energy by continuously analyzing operational data, forecasting future energy demand, and generating control instructions to match available generation resources with forecasted consumption demand; a storage unit connected to the processing unit, configured to store records of historical energy production and consumption, environmental data, operating thresholds and learned model parameters, and to provide said data as input for the forecasting and optimization routines performed by the processing unit; a multitude of IoT-based monitoring units, each comprising at least one sensor configured to measure instantaneous parameters of generation, storage level, consumption rate and environmental conditions, with each monitoring unit being configured to periodically transmit measurement packets to the processing unit via a secure communication network; a forecasting unit implemented in the processing unit, configured to process historical and real-time data to create forecast curves for demand and generation using statistical and probabilistic forecasting techniques, and to dynamically update the weights of the forecasting model in response to observed deviations between forecasted and actual output; an optimization control unit implemented in the processing unit and configured to evaluate the outputs of the forecasting unit together with current operational data to determine a set of optimized control variables representing the target generation contribution of each energy source, and to pass these targets to a lower-level controller for execution; a controller that is communicatively connected to the processing unit and the multiple energy generation sources and is configured to regulate the operation of each source by adjusting the activation state, output level and operating priority based on the control signals received from the processing unit; an energy storage management unit comprising at least one battery array and a power conditioning circuit, configured to receive control instructions from the processing unit, store excess generated energy, release stored energy when forecasted demand exceeds available generation, and report charging and discharging characteristics in real time to the processing unit for continuous recalibration; an alarm and notification control unit connected to the processing unit, configured to continuously compare storage levels and generation reserves with stored operating thresholds, trigger predefined responses when critical or abnormal conditions are detected, and transmit acoustic, visual, and digital remote alerts to designated operators; a user interface terminal connected to the processing unit, configured to display real-time generation statistics, demand forecasts, energy storage status, and system alerts, and to accept operator-defined parameter inputs that are transmitted to the processing unit for recalibration of forecast or optimization parameters; and a secure server interface configured to synchronize operational logs, learning data, and performance indicators with a remote monitoring or analysis server for centralized monitoring, long-term data analysis, and distributed decision support.

A method for monitoring an infrastructure of interest, the method comprising: providing a distributed acoustic sensor, an analysis module, and optical fibre cable adjacent to the infrastructure wherein the optical fibre cable is divided into one or more sections, and the distributed acoustic sensor generates DAS sensor data from the one or more sections; providing a processing module comprising one or more processors and a memory, wherein the memory comprises at least two or more processing algorithms, wherein the one or more processors are configured to execute the processing algorithms with the DAS sensor data as input, wherein each processing algorithm is configured to provide an output state for each section of the optical fibre cable adjacent to the infrastructure of interest.

The invention provides a page structure optimization method and system for a presentation file. The method comprises the following steps: identifying visual elements in a presentation page, and carrying out logical relationship analysis on the visual elements to obtain logical relationship information of the presentation page; mapping the visual structure information into a page three-dimensional implicit field to obtain three-dimensional voxels corresponding to the visual elements; calculating a visual focus thermodynamic diagram in the page of the presentation file, and distributing a three-dimensional voxel corresponding to the target visual element to a visual sensitive area; on the basis of the position relation of the visual elements and the first typesetting information, combining an information density energy function to predict a global optimal layout, and optimizing a narrative path in the presentation file according to logic relation information and user narrative preference to obtain third typesetting information, so that the spatial arrangement relation of the visual elements conforms to the narrative logic of the user preference; and executing secondary typesetting on the PowerPoint page based on the third typesetting information to optimize the page structure of the PowerPoint page and improve the page typesetting efficiency.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a task scheduling execution method, device, equipment and medium based on intention recognition, and the method comprises the steps: fusing an interaction request and a dialogue history library to generate a session context, and carrying out the analysis to obtain an intention recognition result; task planning is carried out based on the intention recognition result and a knowledge base, and a task subitem set and an execution sequence are generated; matching the task subitem set with a resource directory to generate an execution resource list; scheduling the task subitem set by using the execution sequence to generate an ordered task sequence; generating an execution plan in combination with the ordered task sequence and the execution resource list; and triggering execution based on the execution plan, and aggregating output to generate an execution result and a process tracking record. According to the method, automatic connection of session analysis, task planning, resource matching, sequential scheduling and execution triggering is achieved, manual operation splitting among multiple systems is eliminated, and task processing efficiency and execution accuracy are improved.

The invention relates to an intelligent short message template generation method and system based on a large language AI model, and the method comprises the steps: extracting cross-cultural communication frequency, sentence pattern preference analysis results and screen size adaptation data from social media interaction data and equipment use logs of target audiences, and generating a cultural background distribution matrix; identifying regional culture features in the culture background distribution matrix to obtain a taboo matching result; generating an expression habit preference matrix according to the updated culture background distribution matrix; generating a preliminary short message template according to the expression habit preference matrix and the symbol semantic mapping table, and outputting an adjusted template set conforming to message length control; analyzing the acceptability of the adjusted template set on the language conciseness requirement and the interactive button layout, and generating a new template set; evaluating a text alignment mode of the new template set; and screening candidate short message templates according to the information definition score. According to the invention, the accuracy and acceptability of cross-culture short message communication can be effectively improved, and the participation degree of audiences is enhanced.

A method of identifying coronary artery disease comprising: receiving contrast cardiac CT data indicative of a contrast cardiac CT scan carried out on a patient; analysing the contrast cardiac CT data using a vessel seed detector and machine learning to identify a plurality of centreline seed points in the contrast cardiac CT data predicted to correspond to locations on centrelines of the cardiac arteries of the patient; determining whether the analysis to identify centreline seed points of patient coronary arteries is incorrect or incomplete; if the analysis to identify centreline seed points of patient coronary arteries is incorrect or incomplete, reconfiguring the vessel seed detector; producing data indicative of transverse image slices of the cardiac arteries of the patient using the contrast cardiac CT data and the centreline seed points identified using the reconfigured vessel seed detector;and identifying presence of coronary artery disease using the transverse image slices.

Aspects of the disclosure provide for an accelerator capable of accelerating data dependent, irregular, and / or memory-bound operations. An accelerator as described herein includes a programmable engine for efficiently executing computations on-chip that are dynamic, irregular, and / or memory-bound, in conjunction with a co-processor configured to accelerate operations that are predictable in computational load and behavior on the co-processor during design and fabrication.

The invention discloses a large-span continuous rigid frame construction line shape intelligent monitoring method, and relates to the technical field of large-span continuous rigid frame construction, and the method comprises the following steps: S1, data collection; s2, data processing; s3, performing multi-stage intelligent early warning; s4, construction regulation and control; and S5, outputting a result. According to the method, multi-source equipment is used for synchronously collecting data, the real-time performance and integrity of the data are guaranteed, deep learning is used for eliminating anomalies, a neural network is used for constructing an environment influence model and correcting the data, the influence of environment factors on linear data is corrected, digital twinning is combined with a finite element and a neural network to predict a deformation trend, and the prediction precision is improved. A linear deviation trend is pre-judged in advance, a construction scheme is optimized, multi-level intelligent early warning dynamically adjusts a threshold value according to multiple factors, early warning information is accurately issued, risk response timeliness is enhanced, decision suggestions can be automatically generated through construction regulation and control, an evaluation effect is fed back, and the construction process is optimized.

The invention relates to the technical field of computing resource scheduling, and discloses a heterogeneous multi-task computing power dynamic scheduling method and system.The method comprises the steps that in response to multiple computing task requirements of a user, the priority of computing tasks is analyzed and sorted, and a task priority sequence is obtained; classifying the heterogeneous computing resources according to the resource states of the heterogeneous computing resources, and respectively placing different categories of computing resources in corresponding resource pools to obtain multiple categories of resource pools; in combination with the task priority sequence and the multi-class resource pool, allocating computing resources of a corresponding class to each task through a preset resource matching mechanism; in the task calculation process, when the calculation resources of the first task are insufficient, the calculation resources of the second task are preempted through a preset resource preemption mechanism, the priority of the second task is improved after resource preemption, and the calculation resources are redistributed; according to the method and the device, the corresponding computing resources can be allocated in combination with the task priorities, and the resources are dynamically scheduled when the computing resource gap appears.

A method and apparatus for acquiring aerodynamic noise of a compressor, a medium, and a product are provided. The method includes: performing inverse Fourier transform on a known frequency-domain noise spectrum to obtain known time-domain noise data, then predicting unknown time-domain noise using a time series neural network, obtaining finer time-frequency noise data in combination with the known time-frequency noise data, and finally, performing Fourier transform on the finer time-frequency noise data to obtain new frequency-domain noise data. Limitations of a limited time step and a total simulation time on the acquisition of aerodynamic noise data in traditional numerical calculation of aerodynamic noise can be overcome. Finer aerodynamic noise data can be acquired rapidly and accurately. The frequency resolution of the noise spectrum can be increased, and reducing the consumption of computing resources and saving manpower and material resources can be achieved.

Embodiments of the present disclosure relate to multi-view LIDAR perception with motion cues for autonomous and semi-autonomous machines and applications. A DNN may be used to detect objects, a navigable space, weather or surface conditions, artifacts, and / or other parts or features of an environment based on multiple views of LIDAR data from multiple time slices. The DNN may include multiple input channels for processing multiple views of sensor data from multiple time slices to provide motion cues, and the extracted features from the different time slices may be geometrically projected from a first 2D view to a second 2D view, combined with features that were extracted from the second 2D view, and applied to a subsequent stage of the DNN. The data generated by the DNN may be provided to the drive stack of an autonomous vehicle or other ego-machine to enable safe planning and control of the vehicle.

A publishing system with components, including:a system configured to receive at least one document including text that defines a base alphabet in one or more formats;a system configured to provide additional data for a reader to better understand the document which includes:a method of encoding or marking up non-phonetic words in the document to enable the reader to decode sounds of each non-phonetic word; anda system configured to output an encoded document with the text and the additional data in one or more formats,wherein the method of automatically encoding the non-phonetic words to make the encoded words phonetic:for at least one character (“spelling character”) in the non-phonetic word, using a compound character that includes the spelling character and a sound character, wherein the sound characters:are human-readable characters in the base alphabet and / or in one or more secondary alphabets,are added to the spelling characters to indicate that each spelling character makes the usual sound of the sound character,are added so that spelling characters can be visually discriminated from sound characters,are added such that a reader can recognize the non-phonetic word by sight because the spelling of the word is unchanged, andare added to the spelling characters such that the spelling characters and the sound characters remain human-readable such that the spelling character and the sound character of each compound character are within one visual field; andautomatically outputting the encoded words in a human-readable form / format such that the compound characters in the encoded word visually indicate which of the spelling characters have a sound other than their usual sound and what sound each character makes in the non-phonetic word when it does not make its usual sound.

The invention relates to the technical field of photovoltaic energy storage, and discloses a photovoltaic energy storage intelligent management system based on a micro-service architecture, which comprises the following modules: an environmental parameter acquisition module in which G is irradiance, [delta] T is temperature difference, RH is humidity, [alpha], [beta] and [gamma] are attenuation coefficients, dynamic influence of environment on power generation efficiency is quantified, and the environmental parameter acquisition module is used for acquiring environmental parameters; outputting the power correction factor to the cooperative control module; a power generation efficiency correction module; an edge calculation optimization micro-service module; an optical-storage-network cooperative control micro-service module; a self-adaptive prediction micro-service module; the dynamic influence of environmental factors on the power generation efficiency is accurately quantified through a nonlinear coupling model; a power correction factor can be calculated in real time, the influence of complex working conditions such as environment humidity sudden rise and irradiance instantaneous change in cloudy weather on photovoltaic efficiency is effectively captured, a scientific basis is provided for system decision making, and it is ensured that anomaly detection and energy efficiency optimization are close to actual working conditions.

Aspects of the disclosed technology include computer-implemented systems and methods for integrating machine-learned generative models with code editing tools. A code editor is configured to execute computer-executable code within code cells of a code editor interface including a first interface portion and a second interface portion. The interface portion is configured to receive user input for defining and editing a set of code cells within the first interface portion. Each code cell of the set of code cells is independently executable by the code editor application. The second interface portion is configured to receive user input for defining and submitting user queries to a machine-learned generative model. The code editor is configured to modify at least one code cell of the set of cells based at least in part on an output of the machine-learned generative model in response to a user query.

Provided in the embodiments of the present disclosure are a task processing method, a text processing method, an automatic question-answering method, a task processing model training method, an information processing method based on a task processing model, and a cloud training platform, which are applied to the technical field of computers. The task processing method comprises: acquiring task data of a target task; and inputting the task data into a task processing model, so as to obtain a task processing result of the target task, wherein the task processing model is obtained by means of performing training on the basis of a plurality of pieces of target sample data and target sample results of the plurality of pieces of target sample data, the target sample data is obtained by means of performing screening on the basis of detection results of repeated content of a plurality of pieces of sample data, and the content of the target sample results is not repeated. Target sample data is obtained by means of performing screening on the basis of detection results of repeated content, and the content of target sample results is not repeated, such that the repetition illusion of a model is reduced without damaging the processing capability of the model itself and relying on external knowledge, thereby improving the accuracy and integrity of a task processing result.

The invention discloses a brand marketing planning scheme making system and method based on a knowledge graph, and relates to the technical field of brand marketing and cognitive analysis, and the system comprises a data preprocessing module, a drift monitoring and labeling module, and an evaluation regulation and control module. According to the system, multi-source heterogeneous text data is converted into a structured and semantic brand emotional relation graph, scattered user emotional expressions and public opinion concern points are converted into computable nodes and relations supporting time sequence analysis, the data organization is improved, the semantic hierarchy of a knowledge graph is enriched, and the user experience is improved. The cognitive evolution modeling availability is enhanced, the evolution direction and the risk signal are insighted, the time sensitivity and the evolution expression ability of the atlas are improved, the effect of converting from passive monitoring to active intervention is achieved, real-time response and high-credibility labeling of enterprise brand cognitive fluctuation are achieved, and the method is suitable for popularization and application. The timeliness, the accuracy and the intelligent response capability of an enterprise brand marketing strategy are improved, and a closed-loop adaptive regulation and control system is formed.

The invention discloses a building equipment energy consumption statistics and optimization method, and relates to the technical field of building energy consumption optimization, and the method comprises the steps: collecting the energy consumption data of building equipment, and carrying out the preprocessing, and the energy consumption data comprises the equipment power consumption, the operation time, and environment parameters; inputting the preprocessed energy consumption data into the constructed energy consumption reference model to obtain an energy efficiency deviation index; calculating an equipment energy efficiency score based on the energy efficiency deviation index in combination with the environmental parameters; and performing hierarchical management on the building equipment according to the equipment energy efficiency score, and when the equipment energy efficiency score is lower than a preset threshold value, triggering equipment maintenance early warning and generating an equipment optimization scheme. According to the invention, by establishing a dynamic grading threshold and a differentiated management and control mechanism, intelligent grading management of equipment is realized, and management standards can be adaptively adjusted according to historical data distribution characteristics.

Systems, methods, and apparatuses for providing dynamic, prioritized spectrum utilization management. The system includes at least one monitoring sensor, at least one data analysis engine, at least one application, a semantic engine, a programmable rules and policy editor, a tip and cue server, and / or a control panel. The tip and cue server is operable utilize the environmental awareness from the data processed by the at least one data analysis engine in combination with additional information to create actionable data.

The invention discloses a full-closed-loop automatic spraying method, system and device based on geometrical characteristic driving and multi-physical quantity feedback and a storage medium. The method comprises the following steps: performing three-dimensional scanning on a workpiece and extracting geometric features; generating a spraying track and a spraying posture for keeping a topological structure based on the geometrical characteristics; motion, process and coating quality parameters are collected in real time through a sensor, and are processed and fused to form a state vector; inputting the state vector into an intelligent control module, performing optimization by combining model predictive control, fuzzy logic and a neural network, and outputting trajectory and process parameters; a control strategy is updated according to the quality deviation of online detection, and closed-loop correction is carried out; spraying is automatically executed in the closed spraying cabin, and unmanned operation of the whole process is achieved. According to the method, complex curved surface spraying can be completed without manual programming, and the method has the advantages of high flexibility, multi-parameter closed loop, self-adaptive optimization and the like.

A method for separation of non-stationary quasi-periodic signals when limited data is available is described. The method utilizes prior knowledge of time-frequency patterns in the signals to mask and in-paint spectrograms. In one implementation this is achieved through an application-inspired deep harmonic neural network coupled with an integrated pattern alignment component. The network's structure embeds the implicit harmonic priors within the time-frequency domain, while the pattern-alignment method transforms the sensed signal, ensuring a strong alignment with the network.

The invention relates to the technical field of network communication, in particular to a tourist group non-inductive cross-AP roaming method based on multi-stream graph neural network optimization hierarchical reinforcement learning. Comprising the steps of group identification and Bluetooth sideband control establishment, multi-stream scene graph modeling and foresight anchor point labeling, opportunity window inference and group-level constraint projection, HRL high-level sub-target planning, HRL low-level sideband collaborative execution, steady-state shaping, congestion suppression, rollback bundling and the like. Group movement and session states are sensed and represented in a unified manner at a network side, a cross-AP switching plan with a controllable time sequence is generated for groups rather than individuals, and resources and time slots of Bluetooth and Wi-Fi are coordinated, so that low-interruption, low-delay and consistent non-inductive roaming experience is realized in a process that the groups pass through boundaries of adjacent cells.

The invention discloses a method for predicting plant photosynthetic ability by using plant leaf traits, and relates to the technical field of agricultural science, and the method comprises the following steps: obtaining image data of a target plant leaf in a natural state through a high-resolution imaging device, and synchronously recording environment parameter information during collection; extracting multi-dimensional feature parameters of the leaf based on the image data, wherein the multi-dimensional feature parameters comprise a color index, a texture feature, an area, a thickness estimation value and an edge contour form; and carrying out stress feature identification on the character data by utilizing an association rule of the environmental parameters and the leaf characters. According to the invention, by introducing a stress identification and data correction mechanism, the accuracy of photosynthetic capability prediction is improved; the method has good environmental adaptability and crop universality, and is suitable for various agricultural scenes; and a prediction feedback closed-loop system is constructed, continuous optimization and intelligent adjustment of the model are realized, and the stability, reliability and practical application value of the system are remarkably enhanced.

Aspects relate to a compact and low-cost gas analyzer that can be used for different types of gas analysis, such as air quality analysis. The gas analyzer can include a light source, a gas cell configured to receive a sample (e.g., a gas under test), a spectral sensor including a spectrometer and a detector, and an artificial intelligence (AI) engine. Light can enter the gas cell and interact with the sample to produce output light that may be measured by the spectral sensor. The resulting spectrum produced by the spectral sensor may be analyzed by the AI engine to produce a result. The gas analyzer further includes a self-calibration component configured to enable calibration of the sample spectrum to compensate for spectral drift of the spectral sensor.

The invention relates to a construction safety management technology, and discloses a construction potential safety hazard automatic identification method and system based on deep learning to improve the efficiency and accuracy of construction potential safety hazard identification. The method comprises the steps of performing feature extraction on historical construction anomaly information to obtain time period features, personnel behavior features, environment parameter features and equipment operation features, and training a preset construction anomaly hidden danger model to obtain a construction anomaly hidden danger model. And processing the personnel behavior information, the equipment operation information and the environment parameter information to obtain construction precursor data. Inputting the construction precursor data into the construction abnormal hidden danger model for identification to obtain a hidden danger identification result, if at least one abnormity exists in the hidden danger identification result, determining an area position with the hidden danger abnormity, performing hidden danger grade analysis on the area position to obtain a hidden danger abnormity grade, and if the hidden danger abnormity grade meets a preset hidden danger grade, determining that the hidden danger abnormity exists. And if yes, generating hidden danger abnormal early warning.

Systems and methods for tracking inventory items in an area of real space are disclosed. The method includes receiving a signal generated in dependence on sensors. The signal indicates a change to a portion of an image of an area of real space. The method includes, in response to receiving the signal, implementing a trained location detection model to determine, based on inputs, whether an inventory item identified in the portion of the image has changed a position in the area of real space. The method includes implementing a trained item classification model to determine a classification of the inventory item. The method includes updating an inventory database with inventory item data determined in dependence on the classification of the inventory item to provide an updated map of the area of real space as a result of the received signal indicating the change to the portion of the image.

A system and method of encoding a tensor. The tensor includes a first set of feature maps and a second set of feature maps, a feature map in the first set having a first size, and a feature map in the second set having a second size lager than the first size. The method comprises upsampling the first set of feature maps; deriving basis vectors by performing a predetermined process on a tensor including the upsampled first set of feature maps and the second set of feature maps, and deriving coefficients for the tensor using the derived basis vectors to encode the tensor.

A computer-implemented method for performing searches in a document database is disclosed. The method comprises automatically detecting a line of business associated with a user, receiving a text query from the user, and generating a query embedding from the text query. The method further comprises scoring entries in a reverse index using a hybrid scoring function. The reverse index comprises titles, title embeddings, sentences, sentence embeddings, and entity tags corresponding to documents in the document database. The hybrid scoring function is used to generate a score based both on a keyword match score between the text query and the reverse index and on a cosine similarity score calculated from embeddings in the text query and in the reverse index. The method also comprises ranking scores for sentences in the document database, and displaying a sentence associated with a top score to the user.

The present disclosure provides a fault prediction method, a fault prediction model training method, a computing device, a storage medium, and a computer program product. The fault prediction method comprises: obtaining abnormal log data and unit attribute information of a service processing unit; determining an abnormal event sequence on the basis of the abnormal log data; and inputting the unit attribute information and the abnormal event sequence into a fault prediction model to obtain a fault prediction result of the service processing unit, wherein the fault prediction model is obtained by performing training on the basis of a positive sample, a first sample label corresponding to the positive sample, a negative sample, and a second sample label corresponding to the negative sample, the positive sample comprises a positive sample abnormal event sequence and sample unit attribute information, and the negative sample comprises a negative sample abnormal event sequence and sample unit attribute information. The training data of the fault prediction model is richer, so that the accuracy of prediction results of the fault prediction model during applications is improved.