47 results about "Scoring rule" patented technology

The invention relates to a multi-competency intelligent scoring method and system, and belongs to the technical field of intelligent evaluation and talent evaluation. The method comprises the following steps: acquiring multi-source response data of a target evaluation object, executing evidence sufficiency and consistency analysis for each competency dimension, generating a judgment state identifier, and constructing a structural description; performing responsibility distribution on the evaluation evidence, mapping the evaluation evidence into a corresponding evidence model and generating a priority constraint relationship; performing time sequence consistency and stability analysis on the evidence model, and constructing a time sequence constraint rule to prohibit unconstrained score backtracking; and carrying out resolution processing on the continuous undetermined dimension, generating a scoring result through a preset risk scoring rule and a conservative determination strategy, and synchronously generating a scoring basis path and confidence information. According to the method, standard management and control of multi-source evidences and accurate constraint of the whole scoring process are realized, the scoring accuracy and traceability are effectively improved, and reliable support is provided for multi-scene talent evaluation decision.

The invention discloses a heterogeneous workflow task scheduling method and device and a medium. The method comprises the steps of obtaining original data of sub-tasks under multiple feature dimensions according to a preset period; according to a preset scoring rule of each feature dimension, converting the original data into a quantitative score under the corresponding feature dimension; based on the original data, adopting an entropy weight method and an analytic hierarchy process to calculate the weight of the subtask under each feature dimension; calculating a total feature score of the sub-tasks based on the quantitative scores and weights of the sub-tasks under each feature dimension; and obtaining a scheduling scheme of the heterogeneous workflow task through the total feature score of each subtask. According to the method, the timeliness and adaptability of the scheduling decision are improved.

The invention provides an intelligent reviewing method, and belongs to the technical field of computers. In the method, firstly, a target image is processed to obtain a target answer text corresponding to a student answer, secondly, in response to a first input of a first user, a scoring detailed rule corresponding to a target test question is obtained, and the scoring detailed rule can only comprise a first scoring condition and can also comprise the first scoring condition and a second scoring condition at the same time; the second scoring condition is a logical scoring condition, an integrity scoring condition and / or a relevance scoring condition; and finally, based on the scoring detailed rule, scoring the target answer text to obtain a first score corresponding to the student answer, thereby realizing structuralization, standardization and automation of the subjective question reviewing process. According to the method and the device, through a mode of combining image recognition, rule configuration and semantic scoring, not only is the efficiency of reviewing subjective questions significantly improved, but also the dependence of the scoring result on subjective judgment of teachers is effectively reduced.

The invention discloses an enterprise low-carbon evaluation method and system based on retrieval enhancement generation, and the method comprises the steps: carrying out the format recognition and table semantic reconstruction of an unstructured report through a preprocessing module, carrying out the data cleaning and unit standardization in combination with a carbon accounting standard, and constructing a vector index database; in the retrieval stage, the evaluation indexes are converted into query vectors, and related evidences are accurately positioned through mixed retrieval; constructing an enhanced cue word containing unit standardization logic to guide large model reasoning, and generating a structured intermediate result; an anti-fact consistency verification mechanism is introduced, the generated content and the original context are compared, and a low-confidence result is removed; and finally, carrying out weighted calculation in combination with a scoring rule and evidence confidence to obtain an enterprise low-carbon level score. According to the method, high-precision and traceable automatic low-carbon evaluation is realized by fusing table semantic perception based on physical coordinates, unit normalized reasoning of an embedded industry standard and an anti-fact consistency verification mechanism.

The invention discloses a batch composition scoring system, method and device based on multi-dimensional quantization and a storage medium. The system comprises an input processing module, a scoring module and a scoring module, wherein the input processing module extracts basic information of each composition from a batch composition picture based on multiple threads; the large model module extracts composition text feature information from the corresponding composition pictures according to the basic information of each composition based on the message queue, analyzes the composition text feature information according to a preset special constraint rule, a penalty factor rule and a basic score rule, and outputs a multi-dimensional score of each composition picture; and the multi-dimensional scoring engine performs multi-dimensional quantitative scoring on each composition picture through a multi-dimensional quantitative scoring formula according to the multi-dimensional scores. According to the method, the multi-dimensional quantification rule is fused to the large model, the objectivity of traditional rule quantification is reserved, a dimension self-defining mechanism is introduced, a user is allowed to expand the scoring dimension and adjust the quantification standard or the weight proportion according to the requirement, and fair and accurate automatic batch scoring is achieved.

The embodiment of the invention relates to a document score generation method and system based on a large model, and the method comprises the steps: generating a score dimension list and a score constraint condition according to a user-defined score statement, and defining a score logic according to the score dimension list and the score constraint condition; constructing a structured scoring rule, and inputting the structured scoring rule into the plurality of basic large models; analyzing the submitted document by utilizing the basic large model to obtain a plurality of semantic unit vectors, performing similarity matching on the plurality of semantic unit vectors and each scoring dimension in the structured scoring rule to screen out a target semantic unit vector, and according to the target semantic unit vector and the structured scoring rule, obtaining the target semantic unit vector; generating a preliminary scoring result of each scoring dimension; and inputting the preliminary scoring result into a high-level large model, and performing weighted aggregation on the preliminary scoring result of each scoring dimension generated by each basic large model by using the high-level large model so as to generate a document score with higher accuracy.

The invention relates to the field of artificial intelligence, and particularly provides a 3D model scoring method and system based on a neural network model, a terminal and a medium, and the method comprises the steps: employing a pre-trained neural network model to generate standard fusion features for a 3D model uploaded by a teacher user, and employing a local geometric feature extraction model to extract local geometric features, extracting global topological relation characteristics by using a global topological relation modeling model, and performing weighted fusion on the local geometric characteristics and the global topological relation characteristics by using an attention mechanism to obtain fusion characteristics; using a pre-trained neural network model to generate a to-be-scored fusion feature for the 3D model uploaded by the student user; comparing the fusion feature to be scored with a standard fusion feature, and obtaining a 3D model score based on a pre-configured scoring rule according to a comparison result; and feeding back the 3D model score of the student user to the teacher user. According to the invention, two-channel feature extraction of CNN and GNN is combined, so that the scoring efficiency and accuracy are improved.

The invention provides a virtual practical training scoring method and system for the whole process of ship power operation, and belongs to the technical field of virtual practical training evaluation and assessment. Comprising the following steps: S1, dividing a virtual practical training whole process of ship power device operation into different links, and enabling a student to reserve a corresponding virtual practical training after completing a preposed theoretical course of one link; s2, respectively constructing scoring content based on a student virtual practical training operation process for each link, and drawing up a specific scoring rule; s3, performing virtual practical training on the successfully reserved trainees in sequence, obtaining virtual practical training scores according to a scoring rule, and sending different suggestions to the trainees according to the score range of the virtual practical training scores; and S4, extracting a link of error-prone operation of the current semester, constructing a multi-dimensional screening parameter, searching associated knowledge, and reconstructing the content of the preposed theoretical course, thereby reducing faults during fault identification and fault repair operation errors after fault identification.

This invention belongs to the field of artificial intelligence technology and discloses a multi-tool collaborative LLM intelligent agent analysis method and system for high-tech enterprise certification. The method includes identifying the assessment intent for high-tech enterprise certification, identifying data-gathering and inference-type tools, and constructing a binary classification directed dependency graph; asynchronously and in parallel calling tools for business registration, intellectual property, tax and finance, judicial risk, and R&D expense aggregation according to predecessor dependencies; performing format normalization, field mapping, subject alignment, conflict retention, deduplication, dimension aggregation, and scoring caliber mapping on the returned data; performing cross-dimensional cross-validation before comprehensive inference, and inputting the unified intermediate dataset into the LLM to perform a single comprehensive inference. This scheme can reduce serial waiting and invalid inference calls, and reduce the impact of subject confusion, caliber conflicts, and misuse of scoring rules on the certification conclusion.

The embodiment of the invention provides a quantifiable fall risk determination method, electronic equipment, a storage medium and a product, and relates to the technical field of data processing, and the method comprises the steps: obtaining a preliminary screening result of a fall risk of a target object, and obtaining a body function evaluation parameter, collected by timing equipment, of the target object; based on a preset quantitative scoring rule, calculating a body function scoring result of the target object; obtaining maximum angle range data, limb function quantitative data and gravity center displacement quantitative data of a specific body part of the target object in a preset motion test action; based on a preset weight coefficient and a quantitative fusion model, obtaining a quantifiable fall risk value of the target object; and determining the fall risk level of the target object according to a preset risk level division rule and the quantifiable fall risk value of the target object. The embodiment of the invention is used for providing a technical scheme capable of observably and accurately quantifying the falling risk of the old people.

The application provides an enterprise credit intelligent analysis method and system fusing a knowledge graph and a dialogue engine, and relates to the technical field of enterprise credit risk analysis.The application firstly acquires multi-source credit records of a target enterprise and associates source information fields to construct an enterprise credit knowledge graph; analyzes a credit query request to obtain a credit conclusion unit and a graph query condition; searches a candidate fact set and forms an evidence chain data package; groups according to a preset conflict rule and calculates a confidence score based on a preset scoring rule to determine a main conclusion and an alternative conclusion and generate a conflict explanation, which is output to a dialogue engine to generate a dialogue reply.The application associates the credit conclusion with the corresponding evidence chain, introduces a conflict grouping and confidence sorting mode, and improves the structured expression ability and conclusion reliability of the enterprise credit analysis result.

The application provides an industrial mold structure parameter optimization method and system based on multi-source data, relates to the field of industrial mold design, and has the technical scheme as follows: original parameters are extracted from multi-source mold data and converted into parameter records to ensure the consistency of data sources; a local object set is obtained through extension and a correlation between structure objects is established to determine a parameter allowable range and lay a foundation for subsequent optimization; when a parting boundary is adjusted, the alignment deviation of a guide part, the fitting length of a mold locking fitting boundary, and the connection state of an exhaust passage can be calculated according to the correlation to realize linkage analysis; a candidate scheme is generated by adjusting parameters within the parameter allowable range under the condition of meeting a preset hard constraint condition, a target scheme is determined through a preset scoring rule, and intelligent optimization is realized; finally, the landing point of the target scheme in multi-source mold data is checked reversely and written back synchronously to ensure the consistency of data and the executability of the scheme.

The invention discloses a normal distribution-based standard quantitative score optimization system, method and device and a storage medium, and the system comprises the steps: when a large model module receives batch composition pictures transmitted by an input processing module, carrying out the multi-dimensional feature analysis of each composition picture, and outputting the multi-dimensional score of each composition picture; a multi-dimensional scoring engine calculates a multi-dimensional quantitative initial total score of each composition picture according to the multi-dimensional scores; and the score optimization module performs composition ranking according to a preset ranking rule according to the multi-dimensional quantized initial total score, and optimizes the multi-dimensional quantized initial total score of each composition picture according to a composition ranking result based on normal distribution. According to the method, through a coupling mechanism of AI ranking generation and normal distribution scoring, the problem of scoring stiffness caused by excessive dependence on a fixed scoring rule in a traditional system is solved, and fair and flexible automatic scoring conforming to an education evaluation rule is realized.

The elevator operation risk scoring method based on time sequence anomaly detection and correlation analysis comprises the steps that multi-source data of the Internet of Things, the Internet and the like of an elevator are obtained, a plurality of initial risk indexes are constructed, and good and bad samples are divided by calculating risk initial values; calculating the correlation between each index and the good and bad attributes of the sample, and adopting chi-square test and V value quantification and screening out effective risk indexes; determining weights and scores of the indexes in multiple dimensions such as identity features and operation capability according to the correlation, forming a scoring rule, and calculating the score of each dimension; and carrying out weighted summation on the dimension scores to obtain a total risk score, and dividing the elevators into different risk grades according to the total elevator total score distribution. According to the elevator risk assessment method, objective, dynamic and efficient assessment of elevator risks is achieved through a data-driven full-process automatic model, and a scientific basis is provided for precise supervision and preventive maintenance.

The invention discloses an information processing method and device, and relates to the technical field of data processing. The method comprises the steps that multi-source investment information is collected and preprocessed, and an investment information set is obtained; calculating an authoritative score of each piece of investment information according to an authoritative scoring rule based on the source identifier included in each piece of investment information to obtain a to-be-processed information set; extracting the structured investment data of each piece of to-be-processed information through a pre-trained large language model to obtain an investment event; based on a preset classification tag system, matching the structured investment data through keywords to obtain a classification tag of each investment event; calculating an importance score according to an importance scoring rule based on the structured investment data and the classification label of each investment event; and outputting all investment events in sequence. The information acquisition efficiency can be effectively improved, the importance of the investment event is quantitatively evaluated through the authority scoring rule and the importance scoring rule, and the efficiency and accuracy of investment and research decision making are remarkably improved.

The invention belongs to the technical field of artificial intelligence, and relates to a search keyword generation method, which comprises the following steps: constructing a prompt text and establishing a scoring rule set according to user basic information and historical behavior data, and constructing a fine tuning training sample set based on the prompt text and the scoring rule set, inputting the fine tuning training sample set into a pre-trained large language model to generate a candidate search keyword set, scoring the candidate search keyword set by adopting a scoring rule set to obtain a keyword scoring result, and performing fine tuning on the large language model through a preset reinforcement learning algorithm until an iteration stop condition is met, a final keyword generation model is obtained, and personalized search keywords are generated according to the target prompt text through the keyword generation model. The invention further provides a search keyword generation device, computer equipment and a storage medium. The keyword searching method and device can be applied to business management program systems such as financial science and technology, it is guaranteed that the generated keywords have high adaptability, and the keyword searching accuracy is improved.

The invention discloses a runoff probability forecasting method based on an improved deep integration strategy, belongs to the field of runoff probability forecasting, and solves the problem that an existing deep integration strategy can only integrate a single network and arithmetic average distribution weight. The method comprises the following steps: 1, selecting LSTM, GRU and SWM as an integrated sub-network according to an inclusive strategy, and constructing a prediction framework of an input step length 4, an output step length 1 and a hidden layer number 3; 2, a sigma unit is added to an output layer of the sub-network, and softplus is used for activating function constraint, so that uncertainty can be predicted in a quantized mode; 3, taking a negative logarithm likelihood function based on normal distribution as a training scoring rule; 4, after the sub-networks are trained in parallel, a training set is split through three-fold cross validation, an integrated weight is dynamically optimized through a genetic algorithm, and an MNDE model is constructed; the model is superior to a single deep learning model and a traditional machine learning model in certainty, interval, probability prediction and reliability, and is suitable for runoff probability prediction.

This disclosure provides a method and system for intelligent scoring of product display images based on dynamic rule configuration. The method includes recognizing and performing basic structured modeling on the collected shelf images to generate a structured data set; identifying a cross-category and spatially continuous set of products to construct a combined display semantic unit with business semantics; semantically reconstructing the original scoring rules through a dynamic rule engine, transforming the single-product rule system into a two-layer rule structure consisting of combination priority rules and single-product supplementary rules; calculating the overall combined score of the combined display semantic unit according to the mapping relationship, performing single-product supplementary scoring on products not covered by the combination priority rules, and merging to generate the shelf scoring result; and re-executing the dynamic rule reconstruction and scoring calculation based on the archived combined display semantic units and structured data set in response to changes in scoring rules. This method solves the problems of slow rule adjustment response, single scoring dimension, and difficulty in reusing historical data.

The invention discloses a metal hydride safety risk multi-level fuzzy evaluation method based on game equilibrium clustering neutralization. The method comprises the following steps: firstly, constructing a natural environment, production process and production personnel management criterion layer system according to a multi-process flow risk, and subdividing specific risk indexes to construct a hierarchical structure; thirdly, organizing experts to compare every two level factors according to a scoring rule to construct a fuzzy judgment matrix, then calculating index weights by using a game theory equilibrium utility function, a clustering neutralization algorithm and a fuzzy analytic hierarchy process, performing normalization, defuzzification by using a gravity center method, solving an average value to obtain a weight vector, and solving a maximum characteristic value; and the weight is ensured to be reasonable through consistency check. And finally, risk index coefficients are calculated and summed to obtain a safety risk score by combining the index weights of all levels and expert scores, and the risk level is determined according to a score interval. The method is systematic, comprehensive, scientific and objective, risk assessment accuracy is effectively improved, and powerful guarantee is provided for industrial safety production.

The invention discloses a scoring rule formula generation method, system and device and a storage medium. The scoring rule formula generation method comprises the steps that S1, index scoring rule content in a natural language form input by a user is received; s2, analyzing and processing the content of the index scoring rule by adopting a natural language processing algorithm and a model, and extracting key information; s3, according to the key information and a preset rule, generating a scoring rule formula which can be analyzed by a computer; s4, performing grammar check and optimization on the scoring rule formula; and S5, outputting the optimized scoring rule formula to a back-end system, and applying the scoring rule formula to a scoring calculation process. By adopting the method, automatic, efficient and accurate generation of the scoring rule formula is realized, manual errors are reduced, the formula format is unified and standardized, and the method is suitable for making the multi-scene scoring rule.

The invention discloses a reinforcement learning training method and system for a gauge-driven reward model, a medium and equipment. The method comprises the following steps: constructing a dynamic gauge library containing a plurality of definable scoring dimensions, detailed rules and weights; performing fine-grained analysis and multi-dimensional scoring on a task response text generated by the strategy model by using the gauge library, and performing weighted fusion to generate a comprehensive reward signal; based on the reward signal, training by adopting a grouping near-end strategy optimization algorithm: dividing a sample into a high reward group and a low reward group, and configuring differentiated dominant function cutting threshold values and strategy gradient updating step lengths for different groups so as to realize refined strategy updating; in the training process, the training state is monitored in real time, and when triggering conditions are met, score weights, compliance boundaries or scoring rules of related gauge items in the gauge library are adjusted in a self-adaptive mode. According to the method, the efficiency and stability of reinforcement learning training and the controllability of model behaviors are effectively improved.

The application relates to the technical field of architectural design, and provides a three-star green building cost optimization method, device, equipment and medium, which can construct a full-life cycle incremental cost prediction model of a green scheme based on the interaction between technical measures, so as to accurately perform building cost prediction; a plurality of to-be-optimized technical measures are selected from each technical measure included in the green scheme through cost sensitivity analysis, so as to lock the technical measures with high conversion rates for optimization; a target function is constructed according to annual total energy consumption, annual carbon emission and the full-life cycle incremental cost prediction model, and a constraint condition is constructed according to a three-star green building scoring rule, so that multiple optimization targets can be considered; a Pareto front solution set of the optimization model is solved based on an orthogonal table, so that the solving efficiency can be improved; and a target scheme is selected from a plurality of candidate schemes according to a value coefficient and a dynamic benefit balance point, so that an optimal three-star green building cost optimization scheme can be reasonably selected.

The invention discloses a Nand particle testing method, electronic equipment and a testing system. The test method comprises the following steps of: according to a preset execution sequence of test items, executing the following steps on Nand particles in sequence: executing a current test item, and obtaining a test result index; according to the test result index, calculating the score of the current test item by adopting a scoring rule corresponding to the current test item; under the condition that the test result indexes meet test adjustment triggering conditions, re-determining scoring rules of subsequent test items, and / or re-determining scoring weights of all the test items; and calculating the overall test score of the Nand particles according to the score of each test item and the corresponding score weight. By adopting the automatic multi-scene test and test data analysis scheme provided by the invention, the test scoring rule and / or the test item weight are dynamically adjusted, and more comprehensive, more efficient and more accurate reliability test of Nand particles is realized.

The invention relates to the technical field of water conservancy and hydropower engineering, and provides a waste slag site selection method and platform, a computer readable storage medium and equipment, and the waste slag site selection method comprises the steps: obtaining a waste slag site primary selection point location of a target water conservancy and hydropower engineering; automatically scoring the primary selection point locations of the waste slag yard based on a first scoring system, and sorting according to a first scoring result to generate a first point location sequence; the first scoring system is constructed based on a plurality of site selection influence factors, and each site selection influence factor corresponds to a weight coefficient and a quantitative scoring rule; according to the first point location sequence, in combination with auxiliary site selection decision information input by a user through interactive operation, determining a waste slag site final selection point location from the waste slag site primary selection point locations; and carrying out visual presentation on the waste slag field final selection point location and the engineering design information thereof. According to the method and the device, automatic screening, man-machine collaborative decision and three-dimensional visual presentation of waste slag site selection can be realized, and site selection scientificity is remarkably improved.

The application discloses an intelligent recruitment analysis method and system based on evidence chains, and particularly relates to the technical field of computer data processing.The method comprises the following steps: receiving a post requirement text, post metadata and candidate multi-source input data, establishing an evaluation transaction and generating an evaluation tracking number; associating a post portrait version number, a capability dimension model version number, a scoring rule version number, a model service version number and a prompt parameter version number, forming a version anchoring record and determining a credible historical version baseline; structurally processing the multi-source input data, extracting evidence units, generating input snapshots and initial evidence chain maps; executing evaluation calculation under the constraint of the version anchoring record, outputting candidate evaluation results and generating output snapshots; and performing same-version recalculation or differential recalculation according to a recalculation request, and generating recalculation audit results. The version anchoring, evidence tracing and result review of the recruitment evaluation process are realized, and the interpretability and review reliability of the recruitment analysis results are improved.

The invention relates to the technical field of educational evaluation and artificial intelligence, in particular to a subjective question scoring detailed rule conversion method based on LLM and scoring key point extraction, and the method comprises the steps: obtaining scoring detailed rules compiled by a teacher and question stems, question types and total scores of target test questions, classifying the scoring detailed rules according to question identifiers and question types through a large language model, and obtaining scoring detailed rules; identifying score key point names, corresponding original text sentences and score weight information through a character-level score key point extraction model, and completing combination and splitting of score key points and score or weight grade labeling; matching the scoring key points with the question type scoring template, distributing each key point score according to the recommendation proportion and the total score of the question, and generating a structured automatic scoring rule on the premise of meeting the minimum scoring unit and the total score constraint for the automatic scoring system of the subjective question to call. Therefore, the workload of manual configuration is reduced, and the score consistency and interpretability are improved.

This invention discloses an intelligent collaborative execution method, system, and medium for multimodal rule fusion datasets. The method includes: obtaining an execution plan containing rule structure hashes and quality fingerprints; obtaining the current structure hash of the target rule from a knowledge base and performing version verification with the hash in the execution plan; if inconsistent, loading the latest rule and triggering dynamic dependency reconstruction; calculating the data source priority score based on the quality fingerprint and employing a dynamic weight allocation mechanism; selecting the highest-scoring data source as the primary data source and initiating a four-level degradation mechanism; when rule conclusions conflict, calculating a score by comprehensively considering priority, quality score, and visual confidence, and outputting the highest-scoring rule. This invention achieves rule version consistency assurance, quality-aware scheduling, and multimodal rule fusion reasoning, improving execution accuracy and system robustness.

The invention discloses a machine learning and expert rule fused anti-fraud method and system for a public account. The method comprises the steps of obtaining multi-source heterogeneous data and performing feature extraction; based on the risk feature wide table, performing feature optimization by adopting a dual-stage feature screening algorithm; a mixed risk assessment framework is constructed, the optimized features are converted into a scoring system rule and a triggering system rule, the scoring system rule calculates a dynamic risk score through multi-dimensional risk feature weighting, and the triggering system rule sets an instant early warning threshold value for the high-risk features; and dynamically adjusting rule parameters in the hybrid risk assessment architecture according to a regional risk portrait, matching a regional risk mode through a localized feature library, automatically updating a score weight and a trigger condition, and forming a closed-loop iteration rule system adapted to risk features of different regions or industries. According to the method, the precision and the real-time performance of anti-fraud detection of the public account can be remarkably improved, rapid evolution of fraud means can be effectively dealt with, the false alarm rate is reduced, and the risk scene coverage range is expanded.

This invention relates to a method for microservice anomaly root cause localization based on multi-source data fusion and a large language model, comprising the following steps: parsing a set of anomaly cases input by the user, extracting indicator data, log data, and call chain data corresponding to the anomaly time period; performing multi-dimensional anomaly detection and data compression based on the anomaly time period to obtain indicator anomaly detection results, log anomaly detection results, and call chain anomaly detection results; constructing prompt words; performing reasoning based on preset scoring rules to output structured analysis results; validating the format of the structured analysis results, and generating a structured localization result file. The method, system, apparatus, processor, and computer-readable storage medium of this invention for microservice anomaly root cause localization based on multi-source data fusion and a large language model solve the problems of complex dependencies and heterogeneous data in microservice system fault localization, achieving high-precision, interpretable, and automated root cause analysis.

This invention provides a system and method for analyzing experimental results from multiple samples of the same case, comprising: an intelligent sample receiving end, used to receive sample information from multiple different types of samples submitted at the same time for the same case, and send the sample information to a sample quality assessment module and an intelligent decision-making module; a sample quality assessment module, used to assess the quality of each sample based on the sample information, obtain the quality control result of each sample, and send the quality control result to the intelligent decision-making module; and an intelligent decision-making module, used to acquire the experimental results of each sample and the clinical diagnosis result of the same case, and calculate the score of each sample according to a preset scoring rule based on the sample information, experimental results, quality control results, and clinical diagnosis result, and generate auxiliary decision-making suggestions based on the score of each sample. This invention can automatically acquire and analyze the experimental results of multiple samples submitted at the same time for the same case, improving analysis efficiency.