12 results about "Weighted score" patented technology

The invention provides a subjective question scoring system and method based on large language model combination weighting, and the system comprises a key point matching scoring module which is used for recognizing an answer to be scored based on a keyword, and obtaining a key point matching score; the similarity comparison scoring module is used for scoring the similarity between the to-be-scored answer and the standard answer semantics to obtain a semantic similarity comparison score; the language model scoring module is used for obtaining subjective question scores based on a large language model on the basis of a large language model fusion prompt word engineering technology; and the weighted scoring module is used for weighting each scoring result to obtain a final subjective question scoring result. According to the technical scheme, scoring errors caused by limitation of a single method can be effectively avoided, so that the scoring accuracy and credibility are remarkably improved. The integrated design of the method not only enhances the robustness of the scoring system, but also brings a new technical breakthrough for the subjective question scoring field.

The invention provides a user portrait method and device for driving training learning, electronic equipment and a medium, and belongs to the technical field of information recommendation, and the method comprises the steps: collecting multi-source heterogeneous data of a student, carrying out the data preprocessing, and obtaining a student data file, the multi-source heterogeneous data comprises learning stage data, capability evaluation data and basic attribute data; based on the trainee data file, extracting two-dimensional features of trainees, including obtaining learning stage features according to division of passing states and learning progress completion rates of driving test subjects, and quantifying the mastering degree of subdivided test points under weak subjects based on a weighted scoring algorithm to obtain weak subject features; and fusing the two-dimensional features and the basic attribute data, and constructing a student portrait including a learning stage label and a weak subject label. According to the invention, the accuracy of the user portrait is improved.

The invention is suitable for the technical field of data processing, and provides a news credibility evaluation method based on a multivariate fusion weighting and nonlinear punishment mechanism, and the method comprises the following steps: carrying out the multi-dimensional scoring of a news article through a large language model; determining a subjective weight matrix of the news article through the AHP expert matrix; objective weights are calculated based on an entropy method and a plurality of pieces of historical news of the same type, and weight fusion is carried out; penalty factors are applied to the key indexes, and weighted scores are adjusted; and summing the weighted scores and carrying out nonlinear normalization to output a total score. According to the method, the news credibility analysis process is divided into a plurality of detailed evaluation dimensions, and the score of each dimension can be independently output, so that a solid technical basis is provided for manual rechecking, intelligent auditing and transparent responsibility asking. And a weight distribution mechanism gives consideration to subjective expert experience and historical data statistics, and through fusion of AHP and an entropy weight method, the problem that single subjective configuration is easily influenced by personal experience and prejudice is overcome, and the accuracy is improved.

The invention discloses an automatic test selection method based on a utility function weighted scoring model, and the method comprises the following steps: S1, constructing a comprehensive scoring function which is used for calculating the priority score of a test case; s2, sorting the test cases based on the scores, determining the test cases with the scores higher than a threshold value T as a necessary test group, and determining the test cases with the scores lower than T as a spot check group; and S3, executing the test cases of the necessary test group by 100%, and executing the spot check group by taking a Score (i) normalized value as a probability. According to the method, quantitative evaluation is carried out by introducing three factors of risk, importance and cost, high-value test cases are executed preferentially, and repeated execution of test cases with stable long-term passing rate and low risk is avoided, so that the test time and resource overhead are effectively reduced; besides, factors such as module change frequency, historical defect distribution and function key degree are comprehensively considered, so that the test process is more focused on a potential high-risk path, and exposure of more key defects in an early test stage is facilitated.

Methods, apparatus, and processor-readable storage media for automated request processing using an ensemble machine learning framework are provided herein. An example computer-implemented method includes aggregating interaction data associated with a request; computing a weighted score for the request, wherein the weighted score comprises a first component that is based at least in part on a comparison of the aggregated interaction data to a set of keywords and a second component corresponding to a sentiment predicted by a first machine learning model for at least a portion of the aggregated interaction data; using a second machine learning model to determine whether the request is anomalous based at least in part on the weighted score; and in response to determining that the request is anomalous, initiating one or more automated actions for the request.

A small object detection method based on weighted score label assignment is described, involving computer vision. Preprocessed images to be detected and their corresponding categories are fed into a neural network. Image features are extracted and fused, and sample frames are filtered based on these features and fed into a weighted score label assignment module. The weighted score labels are assigned to sample frames, with positive and negative labels assigned based on the weighted scores. During model training, the probability of each category corresponding to the detected object in the image is output. During testing, the coordinates, category, and score of the predicted box for the image to be detected are output. The label assignment process occurs before model training, and the weighted score introduces two completely decoupled scores based on the original Intersection over Union (IoU) score. This effectively alleviates the IoU score's skewness towards large objects, enabling the model to better detect small objects while also ensuring the quality of the sample frames. Experiments on the DOTA and HRSC2016 datasets demonstrate significant and consistent performance improvements.

The invention relates to a nuclear emergency path and opportunity decision-making method based on reinforcement learning and deep learning, and the method comprises the steps: taking minimization of exposure dose and evacuation time as targets, employing a deep reinforcement learning A3C algorithm to carry out the decision-making of the evacuation of a to-be-evacuated agent, and obtaining a plurality of evacuation paths and corresponding action opportunity schemes as candidate schemes; establishing a training set based on a decision sample formed by the candidate scheme, the corresponding decision variable and the multi-source environment information, training a deep learning model which is formed by Transform and a random forest and is from an environment state tensor to the candidate scheme, and outputting a prior score; and obtaining weighted scores of the candidate schemes by adopting linear weighting of score weights, obtaining a comprehensive score by fusing the weighted scores and the prior scores, and determining an executable scheme from the candidate schemes based on the comprehensive score. According to the method, the defects in the aspects of self-adaptability, real-time performance and benefit cost analysis capability in nuclear accident emergency decision-making and path planning in the prior art are overcome.

The invention relates to the technical field of railway construction, in particular to a railway construction advanced intervention problem classification method based on business feature fusion and a gradient boosting tree, and the method comprises the steps: S1, data collection and data storage; step S2, data preprocessing; s3, carrying out feature engineering; s4, performing model training; and S5, performing model performance evaluation and packaging. According to the method, the time feature, the category feature and the long text feature are mined from the advanced intervention problem data, the weighted score and other business features are fused, the hidden feature information in the data is mined, the business mode in the data can be more intelligently captured, and the pertinence and the accuracy of model learning are improved; by optimizing a tree structure and a search strategy, the classification accuracy is improved while the calculation complexity is controlled; the problems that in the prior art, the data statistical analysis capacity is poor, hidden service features in data are difficult to mine, and problem classification lacks are solved.

A competition evaluation system and an evaluation method that provide a user unit for a plurality of players to respectively create a competition member list. The competition member list is generated by selecting at least one of a plurality of competition members by the player. The competition members respectively have a competition result and an appearance information, and each of the competition members is respectively given a performance score and a contribution degree based on the competition result and the appearance information. The invention uses a statistics unit to positively correlatedly correct the performance score of the competition member based on the contribution degree of each of the competition members to generate a weighted score for the competition member, thereby generating a total score for each of the competition member lists, and performing ranking according to the total score, so that the associated players respectively obtain a competition ranking.

A text difficulty grading and assessment method, apparatus, device, and storage medium, the method comprising: obtaining a text to be graded; preprocessing the text to be graded to obtain a feature set, including multiple features related to text granularity, wherein the feature types in the feature set include at least characters and words; obtaining difficulty assessment values ​​corresponding to each feature, including character difficulty values ​​and word difficulty values; obtaining the sum of the character difficulty values ​​of all characters in the feature set of the text to be graded as a Chinese character weighted score, and obtaining the sum of the word difficulty values ​​of all words in the feature set of the text to be graded as a word segmentation weighted score; using the Chinese character weighted score and the word segmentation weighted score to obtain a predicted difficulty level of the text to be graded; judging the predicted difficulty level and the difficulty assessment value, and adjusting the predicted difficulty level when the predicted difficulty level difficulty assessment value meets a threshold condition. The present invention is advantageous in improving the accuracy of text difficulty grading while improving assessment efficiency.

The invention discloses a multi-dimensional quantitative analysis method and device based on performance test data quality, equipment and a storage medium, and relates to the technical field of performance test. The influence factor data comprises a data volume scale, key field distribution, a data space scale, a table association proportion, a table writing rate, a security information check result and a failure rate; analyzing the influence factor data according to a preset weight distribution strategy to generate an original weight matrix; performing quantitative scoring on the influence factor data based on a preset scoring rule and the original weight matrix to obtain a weighted score; and performing quality analysis on the target data set according to the weighted score to obtain an optimization suggestion, and adjusting a construction strategy of the target data set based on the optimization suggestion, thereby improving accuracy and reliability of test result evaluation and optimization.

The invention relates to the technical field of scientific research information management and data processing, and discloses a scoring method for a scientific innovation grade, and the method comprises the steps: obtaining metadata; generating a to-be-processed data set according to a preset mapping rule; on the basis of a field mapping result, constructing a preprocessing table matched with the to-be-processed data set by utilizing a dynamic table structure generation mechanism; according to a preset time interval, writing the to-be-processed data set into a preprocessing table in a batch mode; calculating an average value and a standard deviation of the index data of each first-level innovation dimension according to a statistical year; performing normalization processing on the normalized score, and weighting the normalized score according to a preset weight to obtain a weighted score; and accumulating the weighted scores of all the first-level innovation dimensions, and outputting a scientific innovation comprehensive grade score of the target entity object. According to the method, the problems that a traditional evaluation method is not uniform in quantization scale and high in subjectivity are solved, and the gap in multi-source data fusion and dynamic index weighting in the prior art is made up.