17 results about "Derivation method" patented technology

The invention discloses an agent cue word deduction method and system and a storage medium, and is used for the technical field of language processing. The method comprises the following steps: performing intention analysis on user input information through a main model to obtain an initial task abstract; judging whether the logic of the initial task abstract is complete or not through an auxiliary model; if not, re-correcting the initial task abstract through the main model to obtain a target task abstract; converting the target task abstract into a structured cue word through an auxiliary model; evaluating and optimizing the structured cue word based on a domain knowledge database to obtain a preferred cue word; performing compliance filtering on the preferable cue words to obtain compliance cue words; the compliance cue word complexity is evaluated through a lightweight evaluation model, and an evaluation result is obtained; if the evaluation result is simple, calling an expert model to generate text content according to the compliance cue word; and if the evaluation result is complex, calling a large language model to generate text content according to the compliance cue word.

The invention discloses a discrete manufacturing multi-view BOM dynamic derivation method and system based on a tree-shaped network structure, and belongs to the technical field of intelligent manufacturing and product data management. The method comprises the following steps: constructing a tree body as a single product data source, and storing a hierarchical relationship, attribute definition and engineering semantics of a product in a tree structure; based on the tree body, dynamically matching derivation rules through a mesh mapping relation library; executing rule matching, semantic verification, structure reconstruction and dictionary verification by utilizing a dynamic view derivation engine to generate a target BOM view; semantic constraint and verification are carried out on the derivation process through a BOM data dictionary module; and realizing access control and audit based on role, scene and data sensitivity through the security and authority control module. According to the method, a new generation of xBOM dynamic derivation architecture which takes a tree form as an ontology, takes a net form as mapping, is configurable in rule and is strong in data specification is constructed; the fundamental problem of discrete manufacturing multi-view BOM management is solved, and engineering rationality and service flexibility are both achieved.

The invention discloses a feature derivation method and computing equipment, the method comprises feature generation performed by multiple rounds of iteration, the feature generation of any round specifically comprises: determining an initial feature data set and an effect feedback cue word of the current round, the initial feature data set comprising a plurality of initial features, each initial feature comprising a feature description, utilizing a large language model, taking the initial feature data set as input data, generating a plurality of to-be-screened features according to the effect feedback prompt word, screening in the plurality of to-be-screened features according to the sample data set, determining a plurality of derivative features obtained in the current round, taking the plurality of derivative features as initial features, and extracting the initial features; determining an initial feature data set of the next round; and the large language model is utilized to determine the quality evaluation for the generation process according to the generation process of the current round, and the quality evaluation is used as the effect feedback prompt word of the next round, so that the feature generation process of the next round can be optimized according to the quality evaluation, and the quality of the generated new features is improved.

The invention belongs to the technical field of intention recognition and task derivation, and particularly discloses and provides a group session-oriented intention recognition and task automatic derivation method, which can effectively filter redundant information by deeply understanding a causal relationship and focusing a key path, so that the task derivation efficiency is improved. Implicit and variable intentions in the group session are accurately captured and analyzed, the accuracy and depth of intention recognition are improved, session target understanding is closer to reality, and a solid foundation is provided for subsequent decision making; by simulating the influence of task missing on a core intention implementation path and accurately guiding task supplementation or adjustment according to a verification result, the one-sidedness and redundancy of a task set are avoided, high alignment of the final task set and the core intention is ensured, the completeness and effectiveness of task derivation and the success rate of execution are enhanced, and the task implementation efficiency is improved. And task planning is promoted from static rule driving to a dynamic adaptive optimization level.

To provide a technique for appropriately detecting a marker image in a captured image.SOLUTION: A captured image acquisition unit 212 obtains image data that is read upside down from an image sensor. A first extraction processing unit 234 extracts a connected component of consecutive pixels, from the image data that is read upside down from the image sensor. A representative coordinate derivation unit 238 derives representative coordinates of a marker image, based on the pixels of the connected component extracted by the first extraction processing unit 234.SELECTED DRAWING: Figure 8

The invention discloses a credit investigation variable derivation method and device based on double paths, and belongs to the field of credit investigation variable derivation, and the method comprises the steps: obtaining each to-be-derived credit investigation variable and the corresponding JAVA structured data and derivative variable configuration table, and classifying the to-be-derived variables into JAVA derivative variables and SQL derivative variables according to the configuration table; and processing the JAVA derivative variable through a corresponding JAVA code to obtain a result, obtaining an SQL derivative variable result from an Oracle transaction level temporary table of the optimal database subjected to health degree screening according to the corresponding SQL code, and finally combining the two types of results to output a final derivative result. Therefore, by implementing the method and the device, the problem that it is difficult to ensure that the credit investigation variable derivation result is generated in time when the credit investigation data volume reaches the ten million level in the prior art can be solved.

The invention discloses a feature derivation method, device and equipment and a computer program product, and relates to the technical field of big data, and the method comprises the steps: carrying out the slice aggregation processing of target data to generate basic features, carrying out the derivation operation based on the basic features, and constructing an extended feature library; performing feature similarity calculation based on an extended feature expression and the extended features in the extended feature library to obtain a similarity score between the extended feature expression and the extended features; and determining corresponding target feature derivation processing based on the similarity score, and generating a derived feature corresponding to the extended feature expression. According to the method, judgment is carried out by calculating the similarity between the extended feature expression and the extended feature, the corresponding feature derivation processing is further determined, the derived feature is generated, and the feature derivation efficiency is improved.

The invention provides a worst execution time path derivation method for an ARINC653 operating system. Only tasks to be tested are reserved in a partition to be tested; from the entrance of the to-be-tested task to top down, performing pile pitching on each branch possibly influencing the program trend in the to-be-tested task; running the to-be-tested task under the target machine, injecting a test case set into the to-be-tested task, and collecting time information of each branch in all pile points; subtracting the execution time of the pile point from the execution time of the to-be-tested task; and obtaining a theoretical worst path and an actual worst path of the to-be-tested project, fitting the bifurcation points of the two paths to obtain a fitting path, running the fitting path under the target machine again, and obtaining a final WCET path. The method solves the problems that a static analysis method is difficult to meet the time margin resource requirement of embedded software operation, a dynamic measurement method is difficult to prove that the obtained result is a real WCET value, and the algorithm threshold is too high.

The invention provides an off-line and on-line integrated feature derivation method and a related device. The off-line and on-line integrated feature derivation method is applied to a feature derivation system. The method comprises the following steps: determining a target derivation mode, wherein the target derivation mode comprises an offline mode and an online mode; determining a feature package and a derivation task according to the target derivation mode; the feature packet comprises one or more of original data, derivative data and custom data; performing derivation calculation on the feature packet according to the derivation task to generate a feature wide table; the feature wide table comprises a plurality of derivative features and is used for model training or data analysis; wherein the derivative calculation modes of the offline mode and the online mode are the same. Therefore, the same set of calculation scheme can be used for both off-line derivation and on-line derivation, so that off-line and on-line data can be directly converted, and the off-line and on-line data conversion efficiency is improved.

The application discloses an agent prompt word derivation method and system and a storage medium, and belongs to the technical field of language processing. The method comprises the following steps: performing intention analysis on user input information through a main model to obtain an initial task abstract; determining whether the logic of the initial task abstract is complete through an auxiliary model; if not, revising the initial task abstract through the main model to obtain a target task abstract; converting the target task abstract into a structured prompt word through the auxiliary model; evaluating and optimizing the structured prompt word based on a domain knowledge database to obtain an optimized prompt word; performing compliance filtering on the optimized prompt word to obtain a compliant prompt word; evaluating the complexity of the compliant prompt word through a lightweight evaluation model to obtain an evaluation result; if the evaluation result is simple, calling an expert model to generate text content according to the compliant prompt word; and if the evaluation result is complex, calling a large language model to generate text content according to the compliant prompt word.

The application discloses a kind of derivation methods and systems of maximum sum rate of two-user Gaussian access channel based on rule LDPC code, method is as follows step: S1, the sum node under two-user Gaussian access channel is analyzed, with the sum rate maximization of two-user rule LDPC code as criterion, with two-user system decoding success as premise, construced target function with constraint condition;S2, the degree of LDPC code variable node is fixed, deduce the theoretical analytic expression of check node degree, solve the inverse function of target function in non-convex optimization problem;S3, optimize the degree distribution of two-user LDPC code of two-user Gaussian access channel under system successful decoding, make system sum rate maximization.The present application will provide important help for the design and optimization of LDPC code, to ensure that high performance and reliability are realized in 6G communication system.

The present application relates to an extraction method, an extraction device, an extraction system, and a program, the extraction method including: an acquisition step of acquiring time-series data including a physical quantity generated at a predetermined observation point in a structure as a response caused by movement of a grouping mobile body, which is a grouping of one or more mobile bodies, on the structure; an environmental information acquisition step of acquiring, as environmental information, a structure length, a mobile body length, and information on a setting position of a contact site of the mobile body with the structure; a fundamental frequency extraction step of extracting a fundamental frequency of the time-series data based on the time-series data; a passage period extraction step of extracting a passage period of the grouping mobile body through the structure based on the time-series data; and a number extraction step of extracting a number of the mobile bodies included in the grouping mobile body based on the environmental information, the fundamental frequency, and the passage period.

In one method, a similarity value is calculated based on samples from at least one candidate template among a plurality of candidate templates for the current block and samples from the current block that correspond to samples from at least one candidate template among the plurality of candidate templates. The plurality of candidate templates includes at least one of a left-hand template, an upper-hand template, and an L-shaped template adjacent to the current block. The template of the current block is determined as one of the plurality of candidate templates based on whether the similarity value is greater than a threshold or whether it corresponds to one of the candidate templates. A prediction model is determined based on the template of the current block and a template of a reference block of the current block. The current block is encoded based on the prediction model and the reference block.

The invention discloses a diffusion time step self-adaptive derivation method based on a potential space PSNR (Potential Signal to Noise Ratio), which comprises the following steps of: under an extremely low bit rate condition, taking compression potential as a condition, adaptively determining a starting time step of diffusion back-stepping by utilizing the potential domain PSNR, and introducing noise to count matching loss in a training stage, so as to obtain the diffusion time step of the diffusion back-stepping. And empirical statistics of sample differences are aligned with theoretical diffusion noise statistics, so that the perception quality and stability of reconstruction are improved while the code rate bpp is ensured. A diffusion time step matched with the PSNR is deduced in a self-adaptive mode according to the PSNR of potential space calculation, noise characteristic matching loss is constructed through theoretical noise statistics corresponding to the time step, joint optimization is carried out on the noise characteristic matching loss and main loss, and therefore the statistical characteristics of compression / condition noise and diffusion noise are aligned in the training process; reconstruction degradation caused by mismatch is relieved, and the reconstruction quality and stability under the extremely low bit rate are improved.

The invention provides a characteristic derivation method based on clinker components, and belongs to the technical field of process data processing. The method comprises the following steps of: 1, acquiring time sequence data of process parameters on a cement production line through a quality control system, a quality management system and a sintering system DSC, and acquiring a clinker strength measured value of a corresponding batch through a clinker strength application system; meanwhile, relevant data of the clinker are aligned and integrated according to company names, sampling positions on a production line and date information. 2, abnormal values in the data are filtered through a z-score method; and 3, constructing a mineral phase dynamic statistical feature data set based on daily details and a derivative feature data set based on a time sequence for the data. And 4, combining the mineral phase dynamic statistical feature data set and the derivative feature data set based on the time sequence with unprocessed original data according to a company name, a sampling position on a production line and a date, and outputting an enhanced feature data set. According to the method, the data expression effect is improved, so that the data availability is enhanced.

The present disclosure provides an asynchronous export method and system, the method comprising: in response to a user submitting an export request, creating a corresponding export task; preprocessing the export task based on a preset strategy set, and storing the preprocessed export task in a task queue, the preset strategy set including a task deduplication strategy and a task decomposition strategy; based on the dynamic priority of each export task in the task queue, executing each export task in a multi-thread concurrent manner to obtain an execution result; calculating the execution progress and the remaining time of each export task and feeding back to the user end for display, and when any export task is abnormal, combining a differentiated fault tolerance strategy to retry the export task; generating an export file for the execution result of each export task and uploading the export file to a file storage system. The present disclosure realizes high-reliability, high-efficiency and perceptible asynchronous export through task decomposition, dynamic scheduling and differentiated fault tolerance.

The invention provides a multi-dimensional feature derivation method and device based on a master model, which can describe multi-dimensional derivation features represented by time sequence data in a preset time period from multiple feature dimensions by taking the master model as an anchor for the time sequence data corresponding to a user in a target business scene, and can improve the user experience. According to the method, different characteristics of the time sequence data can be reflected more comprehensively, automatic derivation of the characteristics is achieved, parameter configuration is more flexible, actual requirements of different service scenes can be supported, and meanwhile robustness of characteristic calculation is enhanced.