39 results about "Multiple indicator" patented technology

The invention requests to protect an emotion analysis method based on a text-guided comparative learning and gating fusion mechanism, which is used for solving the problems of insufficient modal feature representation discrimination and insufficient modal information fusion in multi-modal emotion analysis. Firstly, features of three modes are extracted in a feature extraction module, then a text mode is selected as a dominant mode, a text-guided weighted comparison learning strategy is designed, the strategy screens positive and negative sample pairs through the text mode, and weights are dynamically distributed for hard pairs and easy pairs, so that the model keeps universal features while focusing on the hard pairs, and the model is optimized. Therefore, the discrimination capability of modal feature representation is improved. Secondly, in a feature fusion module, a text-guided gating fusion mechanism is provided, weights of text modals and cross-modal fusion features are adaptively adjusted through a gating unit, and a self-attention mechanism is used to suppress fusion noise, so that the modals are fused more fully. Experiments on public data sets CMU-MOSI and CMU-MOSEI show that compared with a reference method, the method provided by the invention has advanced performance on multiple indexes such as F1, ACC-2 and the like.

According to the xerophthalmia core index multi-task AI detection system, method and terminal provided by the invention, the xerophthalmia multi-modal image is collected and exclusive preprocessing is completed, and core index detection is carried out on the preprocessed image based on the xerophthalmia multi-index detection model, so that the xerophthalmia core index detection result is obtained. Outputting detection results such as meibomian gland deletion rate, tear film rupture time, staining classification, tear river height and blepharitis marginal inflammation classification; and dynamically adjusting the index result according to a calibration coefficient corresponding to the type of the data acquisition equipment, and finally generating a standardized detection report. A xerophthalmia multi-index detection model is constructed based on an improved U-Net3 + architecture, a dry eye feature enhancement encoder and a cross-scale feature fusion decoder are adopted, a customized multi-task detection head is matched, an inter-task attention collaboration mechanism is established, meanwhile, a preprocessing algorithm and a dynamic calibration mechanism matched with a dry eye image are designed, and the detection accuracy of the xerophthalmia is improved. And multi-index synchronous, high-precision and high-efficiency detection is effectively realized.

The invention relates to a safety monitoring method for an irrigation and drainage project operation state and an intelligent system, and belongs to the technical field of artificial intelligence. The method comprises the following steps: collecting and marking irrigation and drainage project operation monitoring data through a sensor, and constructing a training data set; completing data normalization by combining quantile and median robust scaling with adaptive nonlinear transformation, and mining and screening high-order interaction features by combining a mutual information theory and a gradient boosting decision tree; constructing a deep classification network fusing physical prior and adaptive feature interaction, introducing physical constraint and multi-scale feature fusion, and optimizing a model through adaptive marginal classification loss and physical feature manifold alignment loss; real-time data is preprocessed and then input into the model, and operation state grade classification and graded alarm are achieved. According to the method, data noise can be inhibited, a multi-index coupling relationship can be mined, the interpretability and robustness of the model can be improved by integrating a physical rule, irrigation and drainage project abnormity can be accurately identified and early warned, and the method is suitable for intelligent safety monitoring of an irrigation area.

The invention discloses an intensive care state real-time prediction method and system based on mask dynamic composition. The method directly faces the challenge that clinical time sequence data is extremely sparse, and a monitoring missing mode is creatively converted into valuable prediction features. A monitoring mode dynamic relation graph is constructed to encode the clinical cooperative monitoring intensity between the indexes; comprehensively modeling time sequence dependence in combination with a time sequence attention mechanism and multi-level local feature extraction; and finally, through adaptive multi-channel fusion, accurate and end-to-end prediction of the future multi-index clinical trajectory of the patient is realized. Compared with a traditional filling or special architecture method, the method has the advantages that knowledge hidden in clinical monitoring behaviors can be more effectively mined, more excellent prediction performance can be achieved on an intensive care state data set with extremely sparse data, and a new technical path is provided for development of an intelligent clinical early warning system.

This invention discloses a method and system for implementing role-playing dimensions, relating to the field of data processing technology. It includes: creating a meta-model of dimension models related to role-playing dimensions, the meta-model comprising five types of data tables; selecting role-playing dimension names from their sources according to requirements; when a role-playing dimension name differs from a physical dimension table name, the dimension corresponding to the role-playing dimension name is considered a role-playing dimension; when displaying the related dimensions of an indicator, using the role-playing dimension name to replace the physical dimension name to demonstrate the actual business meaning of the dimension; when summarizing the indicator using selected dimensions, constructing dynamic SQL using the role-playing dimension name; when analyzing composite indicators, if the facts of the multi-atomic indicators constituting the composite indicator are all associated with the same physical dimension, then the composite indicator is summarized using that physical dimension; when performing parallel analysis of multiple indicators, if the facts of each indicator involved in the analysis are all associated with the same physical dimension, then the multiple indicators are summarized using the physical dimension; and displaying the list of related dimensions using the physical dimension name to avoid confusion caused by different role names.

The invention relates to a fund multi-dimensional quantitative ranking method and system, and the method comprises the steps: collecting the real-time position data of a target fund, and dividing the fund types based on all ranking framework dimensions; calculating each quantitative index value of the target fund in each time window, and obtaining the similar ranking, ranking level and similar mean value of the target fund according to the quantitative index values; obtaining structured input data of the target fund based on the structured cue word, and submitting the structured input data to the large language model to generate a natural language report, the natural language report including fund analysis, risk cue and adaptation suggestion; and repeating the steps based on each ranking framework dimension, generating a natural language report of the same target fund under different fund categories, and constructing a multi-dimensional ranking system of the target fund. A fund multi-dimensional quantitative ranking system is constructed by covering a plurality of classification dimensions, a plurality of time periods and a plurality of indexes and giving appropriate comments to each index data.

The invention relates to the technical field of physiological index data processing, in particular to a user feedback driven physiological index adaptive health assessment method and device. The method comprises the following steps: acquiring a BCG signal of a user based on a non-contact sensor, and acquiring index monitoring data of a plurality of indexes based on the BCG signal; the user inputs a user feedback text representing the abnormal state of the body, matching is conducted on the basis of the user feedback text and the disease words in the lexicon, and if the disease words are matched, indexes corresponding to the disease words serve as final monitoring indexes; if the disease word cannot be matched, screening out an initial monitoring index based on the user feedback text, and determining a final monitoring index based on the initial monitoring index; and obtaining daily index monitoring data based on the final monitoring index. In this way, mapping from coarse-grained subjective perception to refined objective indexes can be realized, and monitoring blind areas and misjudgment risks are remarkably reduced while the early-stage anomaly detection rate is increased.

Systems and methods are provided for assessing cyber attack preparedness associated with organizations. One example computer-implemented method includes accessing data indicative of multiple controls of an organization, where the organization includes an information network, which includes the controls and where the controls are associated with securing one or more information assets. The data is indicative of the controls including multiple indicators. The method also includes aggregating one or more ratings for at least one of the indicators to a criteria, aggregating the aggregate rating for the criteria to a category, and aggregating the aggregate rating for the category to one of the controls as a control maturity score for said one of the controls. The method then includes displaying the control maturity score for said one of the controls to a first user associated with the organization.

This invention relates to the field of psychological assessment technology, specifically a VR-based mental health assessment method, comprising the following steps: analyzing event types, interaction states, and priorities based on virtual scene event nodes; selecting suitable nodes; adjusting the acquisition window; judging changes in EEG amplitude; identifying segmented time points; combining extreme values ​​and response intervals; matching high-risk benchmarks; and classifying psychological risk indicators. In this invention, user interaction behavior and virtual situation events are dynamically mapped to achieve closed-loop management of the entire data flow. The acquisition window is adjusted in conjunction with multi-source parameters, signal segments are defined in real-time based on actual EEG fluctuations and scene cognitive states, EEG electrode values ​​and behavioral response timing are extracted synchronously, multi-dimensional features are coupled and analyzed, and the dynamic grouping of psychological risks is based on the joint setting of multiple indicator intervals. Data is guaranteed to be timely in high-frequency interaction scenarios, and the assessment results can reflect the continuous evolution of the user's psychological state in complex environments.

When searching based on the similarity of multiple indicators, it is possible to check the similarity of each indicator. [Solution] The drawing search system includes a search unit that searches for similar drawing information, which is existing drawing information similar to drawing information, from among multiple existing drawing information registered in an existing drawing database based on a similarity calculated from a detailed similarity, which is the similarity between drawing information and similar drawing information for each of multiple indicators, and a similarity display unit that displays the detailed similarity used by the search unit to search for similar drawing information together with the similar drawing information.

This invention provides a parameter-driven, intuitionistic fuzzy logic method for evaluating the suitability of mega-span cavern sites, relating to the field of geotechnical engineering evaluation. This invention, for the first time, constructs a dedicated tunneling design index system for mega-span caverns integrating "rock mass quality, environmental response, shape effect, and support status." It separates evaluation criteria indicators from engineering response indicators, establishing a mapping relationship between the two through numerical simulation, making the evaluation process more consistent with engineering cognitive principles. Intuitionistic fuzzy numbers are used to quantify the merits of evaluation index values, achieving an organic fusion of objective data and fuzzy expression. The final output is a single, quantifiable site suitability index, enabling intuitive comparison of multiple sites and multiple indicators, greatly facilitating engineering decision-making. By replacing expert scoring with a pre-set mapping function, subjective interference is completely eliminated, ensuring that the evaluation results are entirely determined by the input parameters. Different personnel can obtain consistent conclusions, greatly improving the repeatability and engineering applicability of the method.

Systems and methods are provided for assessing cyber attack preparedness associated with organizations. One example computer-implemented method includes accessing data indicative of multiple controls of an organization, where the organization includes an information network, which includes the controls and where the controls are associated with securing one or more information assets. The data is indicative of the controls including multiple indicators. The method also includes aggregating one or more ratings for at least one of the indicators to a criteria, aggregating the aggregate rating for the criteria to a category, and aggregating the aggregate rating for the category to one of the controls as a control maturity score for said one of the controls. The method then includes displaying the control maturity score for said one of the controls to a first user associated with the organization.

This invention discloses a method, system, electronic device, and computer-readable storage medium for evaluating the sidetracking potential of different types of reservoirs in complex fault blocks. The method includes: determining the reservoir type; selecting evaluation indicators affecting the sidetracking potential of different reservoir types; establishing the weights of each evaluation indicator; determining the distribution of evaluation levels and the corresponding evaluation classification standards for each evaluation indicator in each evaluation level; establishing an unknown single-indicator measurement function; statistically analyzing the actual values ​​of each evaluation indicator for the target reservoir; determining the quantitative value of a single indicator; classifying the sidetracking potential levels based on the comprehensive evaluation of multiple indicators; and calculating the comprehensive score based on the comprehensive evaluation vector of multiple indicators and the assignment of evaluation space. This invention proposes the concept of evaluating sidetracking potential by reservoir type, accurately characterizing the size of the sidetracking potential of the target area, and compensating for the inaccurate potential evaluation when using the same parameters for reservoirs with different conditions; it also establishes an unknown single-indicator measurement function, realizing the hierarchical and quantitative analysis of geological parameters in the target area.

Visualize key performance indicators (KPIs) from various structural perspectives. [Solution] The Key Performance Indicator Tree Display Device 1 includes: a Key Performance Indicator Tree Database 11 having an organizational tree structure and a task tree structure; an indicator selection unit 12 that displays the organizational tree structure and the task tree structure of the Key Performance Indicator Tree Database 11 and accepts the selection of multiple indicators by the user from among the organization and task; an aggregation unit 13 that aggregates performance information related to the multiple indicators; a graph selection unit 14 that accepts the selection of the type of graph to show the performance information; and a graph visualization unit 15 that visualizes the performance information aggregated by the aggregation unit 13 in a graph of the type accepted by the graph selection unit 14.

This invention discloses a reservoir water leakage detection method based on interval membership. The lower limit c of the leakage point with respect to the i-th index interval is calculated respectively. i L and upper limit c i U Regarding the membership degree r of the j-th evaluation level ij L and r ij U Substitute the interval numbers of the hydrochemical indicators at the seepage point into the corresponding membership function formula to calculate the membership degree interval evaluation level vector for each hydrochemical indicator at the seepage point. Combine the membership interval evaluation level vectors of multiple indicators to form an interval number evaluation matrix R. Based on matrix R, calculate the integrated relationship membership degree evaluation result of the seepage point with respect to groundwater, reservoir water, and mixed water levels. When the result is -1... j U When b ≤ 0, the seepage point and the grade show an opposing relationship, and the seepage point is judged not to belong to this grade; when b j L <0 j U At that time, the seepage point and the level of that grade have both similarities and differences; when b j L When the value is greater than 0, the seepage point is determined to belong to that category. This invention is highly reliable, yields accurate conclusions, and saves manpower and resources.​​

One method includes determining (101) a sensing input and, based on the sensing input, determining (103) the probability of the presence of a person or animal. The sensing input reflects changes in radio frequency signals received by one or more devices. The method also includes controlling (105) a lighting device to produce a light effect when the probability is determined to exceed a presence detection threshold, and to continue producing the light effect for a subsequent predetermined period of time, even if the probability changes beyond a predetermined value within a predetermined period of time. The method also includes instructing (107) the probability to a user via an indication selected from a plurality of indications, for example, by presenting the indication on the lighting device. When it is determined that the probability has changed beyond a predetermined value, a different indication from the plurality of indications is selected.

The embodiment of the invention provides an evaluation method and device of an intelligent question answering system, computer equipment, a computer readable storage medium and a computer program product, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring an ith theme type test case belonging to the same target theme, wherein the theme type test case comprises a test question and an expected answer corresponding to the test question; inputting a test question in the ith theme type test case into the intelligent question-answering system, and obtaining an output answer of the intelligent question-answering system; calling a plurality of judgment models by taking the expected answer and the output answer as input so as to obtain a plurality of groups of evaluation conclusions; each group of evaluation conclusions comprises a plurality of evaluation conclusions output by the judgment model corresponding to the group for a plurality of preset index dimensions; and for each index dimension, determining a final evaluation conclusion of each index dimension based on the evaluation conclusions of all theme class test cases under the target theme. According to the technical scheme, the evaluation accuracy and comprehensiveness can be improved.

The application provides a bearing fault diagnosis method and device based on domain adversarial feature selection across working conditions, and relates to the technical field of fault prediction and diagnosis.The original data of different working conditions are collected and preprocessed; then, multi-dimensional features are extracted to obtain a source domain sample feature matrix and a target domain sample feature matrix; then, task correlations between the features and task labels in multiple indexes are calculated; the source domain sample feature matrix and the target domain sample feature matrix are combined, and the domain importance of each feature is calculated in combination with the task importance of each feature; then, a comprehensive scoring function of the features is calculated in combination with a game adversarial optimization mechanism to obtain a feature subset; the feature subset is input into a classification model to obtain a recognition result, and bearing fault recognition under cross-working conditions is completed.The application comprehensively measures the contribution degree of the features in the two dimensions of task and domain, avoids feature selection deviation caused by a single index, and significantly improves the fault diagnosis precision and generalization ability under cross-working conditions.

The invention provides an index query method and device based on a large model, and the method comprises the steps: obtaining and processing index information from an index platform, so as to construct a knowledge base; receiving a query problem of a user based on a natural language, and determining a query time parameter and at least one candidate index based on the query problem; using the large model and combining user interaction to confirm the candidate indexes to obtain to-be-queried indexes; extracting index information of the to-be-queried index from the knowledge base; traversing the dimension list to obtain a dimension value list related to the query problem; obtaining an index real-time query interface parameter based on the index ID, the dimension value list and the query time parameter; and obtaining an index value from the index platform by using the index real-time query interface parameter and feeding back the index value to a user. According to the method, the intelligence, the accuracy and the user experience of index query are greatly improved. In addition, the method solves the problem that the intelligent reconstruction cost of a traditional index platform is high, and can adapt to various index platforms in a non-invasive manner.

This invention discloses a machine learning-based method and system for comprehensively evaluating the environmental remediation performance and safety of micro / nano iron-based reducing materials. The method includes: acquiring environmental remediation performance and safety scores predicted using machine learning; introducing a bottleneck penalty mechanism—when each score is below a corresponding basic compliance threshold, a linear penalty function is used to significantly amplify the weight of low-scoring indicators; introducing a high-zone attenuation mechanism—when each score is above a corresponding high-zone boundary, a saturation function is used to reduce the marginal contribution rate of ultra-high-scoring indicators; calculating the dynamic weight corresponding to each score based on the outputs of the bottleneck penalty and high-zone attenuation mechanisms; and obtaining a comprehensive evaluation score by weighted summation of the environmental remediation performance and safety scores using the dynamic weights. This significantly improves R&D efficiency and safety, and its dynamic fusion and evaluation methodology is highly universal, extending to other functional material evaluation fields requiring the balance of multiple indicators, such as catalysis and pharmaceuticals.

The embodiment of the invention provides a satellite task scheduling algorithm testing method and device, equipment, a medium and a product. The method comprises the following steps: obtaining a plurality of indexes corresponding to a target satellite task, carrying out subjective weight calculation on the plurality of indexes based on a preset subjective weight algorithm to obtain the subjective weight of each index, and carrying out objective weight calculation on the plurality of indexes based on a preset objective weight algorithm to obtain the objective weight of each index, and correcting the subjective weight of any index according to the objective weight of each index to obtain a target weight after index correction. A plurality of to-be-tested satellite task scheduling algorithms are obtained, each satellite task scheduling algorithm comprises a plurality of indexes, the plurality of satellite task scheduling algorithms are tested based on the corrected target weight of each index, and thus a test result corresponding to each satellite task scheduling algorithm is obtained. According to the method, the accuracy of the test result of the satellite task scheduling algorithm is improved.

The application discloses a kind of intensive care state real-time prediction method and system based on mask dynamic composition.It directly faces the challenge of extreme sparsity of clinical time series data, and creatively converts the monitoring missing pattern into valuable prediction features. By constructing a monitoring pattern dynamic relationship diagram, the clinical collaborative monitoring strength between indicators is encoded; combined with time series attention mechanism and multi-level local feature extraction, time series dependence is modeled comprehensively; finally, through adaptive multi-channel fusion, the future clinical trajectory of multiple indicators of patients is accurately and end-to-end predicted. Compared with traditional filling or special architecture method, the present application can more effectively mine the knowledge hidden in clinical monitoring behavior, and can achieve superior prediction performance on the intensive care state data set with extremely sparse data, providing a new technical path for the development of intelligent clinical early warning system.

1. Name of the product in this design: Multi-dimensional monitoring graphical user interface for electronic devices. 2. Intended use of this design: for use in an electronic device. 3. The key design feature of this product is its graphical user interface. 4. The picture or photo that best illustrates the key design points: Design 1 front view. 5. Electronic devices are designed in a conventional way, so other views are omitted. 6. Design 1 is designated as the basic design. 7. Purpose of the graphical user interface: It is used to integrate multi-dimensional information of liquid-cooled servers and present the various indicators of liquid-cooled servers to users in a comprehensive manner through a combination of graphics and text and multiple indicator switching. It also supports customizable dashboards to meet users' personalized data analysis needs. 8. Human-computer interaction methods of graphical user interface: Select the corresponding button in the main view interface of Design 1 to perform the corresponding operation; select the second item "XXXX" in the navigation bar at the top of the main view interface of Design 2 to display the interface change status diagram of Design 2; select the fourth item "XXXX" in the navigation bar at the top of the main view interface of Design 3 to display the interface change status diagram of Design 3; select the fifth item "XXXX" in the navigation bar at the top of the main view interface of Design 4 to display the interface change status diagram of Design 4.

This invention discloses a method and system for anomaly root cause analysis based on a container management platform, belonging to the field of data processing technology. The method includes: acquiring an indicator set, which includes multiple indicators; obtaining a correlation coefficient between two indicators based on the Pearson correlation coefficient algorithm; acquiring an abnormal event and a first indicator associated with the abnormal event; obtaining one or more second indicators associated with the first indicator based on the correlation coefficient; determining whether the second indicator is abnormal; if so, the second indicator is the root cause of the anomaly. By using the Pearson correlation coefficient algorithm, the second indicator associated with the abnormal event is calculated, and the root cause / indicator of the anomaly is quickly located within the range of the second indicator, improving the monitoring efficiency of the root cause of the anomaly and the troubleshooting efficiency.

This invention discloses a prognostic method and system for maintenance hemodialysis patients based on time-series analysis. The method includes: predicting patient data using a prediction model to obtain a prognostic plan; a feature embedding layer of the prediction model encoding input features into a unified embedding representation; a Transformer encoder capturing temporal features and long-term trends from the unified embedding representation to obtain a second feature; a variational autoencoder module for compressing and denoising the second feature to obtain a latent representation; and a prediction head for making predictions based on the latent representation to obtain a prognostic plan. Based on an improved network structure using time-series analysis, a prediction model is trained to predict the prognosis of maintenance hemodialysis patients. Multiple prediction heads can predict prognostic risks, indicator types, or indicator regression values; accurate prediction of multiple indicators can meet the multi-dimensional decision-making needs of clinical practice, providing effective data support for physicians' diagnostic / intervention plans.

The invention relates to the field of water resource evaluation, and particularly discloses a regional water resource toughness evaluation method based on multiple factors, and the method comprises the following steps: S1, collecting meteorological, hydrological, social economic, ecological and engineering data of a region, and carrying out the classification and processing of the data, and constructing a comprehensive database; s2, based on the comprehensive database, selecting a plurality of indexes from five dimensions of meteorology, hydrology, social economy, ecology and engineering under each dimension, and constructing an evaluation index system; s3, calculating the factor importance of each index, and selecting a main influence factor under each dimension; s4, calculating the influence degree of each main influence factor on the water resource toughness, and selecting the factor with the maximum influence degree under each dimension as a main driving factor; and S5, representing the dimension by the main driving factor under each dimension, and respectively calculating the coupling degrees among the dual systems, the three systems, the four systems and the five systems to analyze the interaction degree among the systems so as to complete the water resource toughness evaluation.

This invention relates to the field of deep foundation pit assessment technology, specifically to a method, apparatus, equipment, and storage medium for assessing the toughness of deep foundation pit construction. The invention calculates the characteristic value of each assessment indicator based on multiple indicator values, and obtains a comprehensive characteristic value for all assessment indicators based on the weight value and characteristic value of each indicator. Finally, the comprehensive characteristic value and the benchmark characteristic value are combined to obtain the toughness assessment result of the deep foundation pit. From the above analysis, this application comprehensively considers multiple indicator values ​​and characteristic values, thereby improving the accuracy of the toughness assessment results.

According to the heterogeneous computing network resource abnormity monitoring and early warning method, unified modeling, unified acquisition processing and unified monitoring are carried out on heterogeneous computing network resources, so that mutual separation of monitoring systems can be avoided, root cause positioning can be assisted, abnormity checking efficiency can be improved, abnormity influence range deduction is carried out based on root causes and topology, and the method has the advantages of being high in practicability and the like. The method is favorable for defining the influence range of exception on tenant services, and comprises the following steps: step 1, carrying out unified collection and processing on heterogeneous resource monitoring data based on a unified monitoring model and an index processing rule; 2, monitoring index anomaly detection is carried out based on an index monitoring threshold value; 3, searching a multi-index anomaly joint detection rule, performing joint detection analysis on a single anomaly monitoring index in combination with the resource topology, and if a matching rule does not exist, directly outputting an anomaly index; 4, auditing resource topology, inferring an abnormal influence range and outputting an inference result; and step 5, performing unified abnormity early warning according to the inference result.

This invention relates to the field of data processing technology, specifically to a method and system for monitoring aquatic ecological environment. The method includes: first, collecting monitoring datasets containing multiple indicators from various monitoring points; then, calculating weight correction coefficients by assessing the degree of indicator fluctuation; next, calculating the clustering degree of each monitoring point based on the distance between them, and determining the weight of each monitoring point accordingly; then, using the monitoring point weights, weighted summing the values ​​of each indicator at different monitoring points to obtain a comprehensive value for each indicator; finally, calculating a comprehensive evaluation index for aquatic ecological environment quality by combining the initial indicator weights and the weight correction coefficients; and finally, monitoring the aquatic ecological environment based on the magnitude of this evaluation index. This method avoids the errors caused by fixed weights and discontinuous sampling in traditional monitoring methods, thereby improving the accuracy and scientific rigor of aquatic ecological environment monitoring.