131 results about "Ranking" patented technology

Disclosed is a chat group management method, including: detecting a message receiving mode corresponding to a chat group; obtaining a degree of interest of a user for chat group messages and an activity degree of the user in the chat group in accordance with a determination that the message receiving mode corresponding to the chat group is a mute-notification receiving mode; determining an importance ranking for the chat group according to the degree of interest and the activity degree; and updating the chat group's position among a plurality of chat groups in accordance with the importance ranking.

A system and method for extracting data from large language model (LLM) outputs, including: training a model using labeled data items to assign rankings to LLMs; selecting, by the trained model, one or more of the LLMs based on the rankings; sending an LLM prompt to selected models; and outputting, by the model, a refined response to the prompt based on responses to the prompt by the LLMs. Some LLM prompts according to some embodiments may include different sets input parameters of different types—such as, e.g., a set of block parameters and a set of editorial parameters. In some embodiments, a model or LLM may be updated or retrained using a reinforcement learning approach and based output items or refined responses generated by that model or LLM—which may for example be scored or ranked and used in combination with reward or cost functions to update model parameters.

Existing techniques for automated generation of test data for testing web applications need detailed requirement documents. The present disclosure receives a plurality of textual documents to extract context. Rephrasing the extracted context by implementing a plurality of rules and passing extracted context along with a first set of prompts to Large Language Model (LLM). Generating program, validator and first set of constraints for extracted context and generating test data by running the generated program. Assigning ranking to test data and selecting the test data with highest ranking. Statically refining the generated program by calling a mathematical library function on the highest ranked test data to generate structural information and modifying language of the second set of prompts passed to the LLM. Dynamically refining the generated program by passing feedback generated by executing the highest ranked test data on a web application and refining the response obtained.

In some implementations, a device may determine a set of features for the machine learning model. The device may determine, using off-policy evaluation, a performance level of the machine learning model associated with the set of features. The device may determine, based on the performance level, a ranking of the set of features. The device may generate, based on the ranking, one or more subsets of features from the set of features. The device may determine, using the off-policy evaluation, performance levels of the machine learning model for respective subsets of features from the one or more subsets of features. The device may select, based on the performance levels, a subset of features from the one or more subsets of features. The device may obtain, via the machine learning model, one or more recommendations that are based on input data, wherein the machine learning model uses the subset of features.

Disclosed are techniques for using machine learning models to more reliably predict likelihoods of application failure. A model is trained to identify and display events that may cause high severity application failures. Logistic regression may be used to fit the model such that application features are mapped to a high severity event flag. Significant features of applications that relate to the high severity flag may be selected using stepwise regression. The identified applications may be displayed on a graphical user interface for review and reprioritization. Information may be ranked and displayed according to multiple different ranking criteria, such as one ranking generated by a first model, and another determined by one or more users. The multiple ranking criteria may be used to inform steps taken, and / or to retrain or tune the parameters of the model for subsequent predictions or classifications.

A computing system comprising a memory and processing circuitry may perform the techniques. The memory may store time series data comprising measurements of one or more performance indicators. The processing circuitry may determine, based on the time series data, an anomaly in the performance of the network system, and create, based on the time series data, a knowledge graph. The processing circuitry may determine, in response to detecting the anomaly, and based on the knowledge graph and a machine learning (ML) model trained with previous time series data, a causality graph. The processing circuitry may determine a weighting for each edge in the causality graph, determine, based on the edges in the causality graph, a candidate root cause associated with the anomalies, and determine a ranking of the candidate root cause based on the weighting. The analysis framework system may output at least a portion of the ranking.

The invention relates to the technical field of incremental training, and particularly discloses a method for large model incremental training, comprising the following steps: S1, obtaining a target data set based on an industry corpus data set, dividing the target data set into sub-data sets, obtaining quality indexes of the sub-data sets, and determining quality scores of the sub-data sets; s2, acquiring a first sequence, and determining a performance score of the universal model from the first sub-data set in the first sequence; calculating a performance score difference value, and determining whether the state is an abnormal state or not according to a performance score difference value threshold value; s3, coordinate points are generated and fitted to obtain a fitted curve, and abnormal sorting positions are obtained according to the fitted curve; and determining a target model according to the abnormal sorting position, and performing incremental training on the target model through the to-be-determined data set. According to the method, the negative influence of low-quality data on the model can be reduced, the adaptability of the model to new domain knowledge is improved, and the problem of performance degradation is avoided.

A method includes: generating a set of tasks; determining, by a machine learning model and based on multiple data types from multiple sources, that an overall risk score exceeds a first failure threshold due to a risk score of a task exceeding a second threshold; selecting a replacement task for the task, the selecting including: receiving, replacement candidates, each replacement candidate including a candidate offset potential and one or more candidate failure mechanisms; assigning, by the machine learning model and to each of the replacement candidates, a replacement score for the replacement candidate based on a failure correlation of the replacement candidate with respect to each other sets of the set of tasks; ranking the replacement candidates based on the replacement scores; and selecting, based on the ranking, the replacement task; and generating, an updated set of tasks including the replacement task.

The invention discloses a retrieval enhancement generation method and system based on large model multi-level sorting, and the method comprises the steps: carrying out the multi-round rewriting and disassembly operation of a question inputted by a user, and carrying out the analysis and summarization of the disassembled data, so as to obtain the disassembled data; performing intention analysis by utilizing an intention recognition optimization model on the basis of the disassembly data, and obtaining weight data corresponding to each intention; based on the weight data corresponding to each intention and the constructed knowledge base, performing multi-source recall by using a retrieval enhancement generation technology to obtain a multi-source recall result; and performing multi-level sorting on the multi-source recall result to obtain a retrieval result. According to the method, the user intention recognition model, the satisfaction sorting based on the thinking chain technology and the availability sorting based on the large language model are introduced, so that the real requirements of the user are accurately understood, the accuracy and quality of the generated content are improved, the complex and professional problems are solved, and the actual application requirements are met.

The invention discloses a tensor recovery method fusing non-convex low-rank minimization and depth prior. The method comprises the following steps: obtaining an observation tensor, and an index set, a prior set and reversible transformation corresponding to the observation tensor; inputting the acquired data into a constructed tensor recovery model, and outputting an initial recovery tensor; the construction of the tensor recovery model comprises the following steps: carrying out differential processing on a multi-order tensor to obtain a gradient tensor; constructing a TCTV regular term according to a tensor non-convex norm corresponding to the gradient tensor; building a tensor recovery model based on a TCTV regular term; and inputting the initial recovery tensor into the convolutional neural network model, and outputting a final recovery tensor. According to the method, the image recovery capability is enhanced, the recovery performance in an extreme environment is more remarkable, and the method plays a key role in coping with loss or damage of high-dimensional data; and the accuracy of downstream tasks based on image processing, such as face recognition, target tracking and the like, can be ensured.

Systems and methods for automatically generating search list rankings of code snippets are provided. An exemplary method includes searching source code repositories to identify code snippets in response to a search query, assigning weight values to ranking parameters, and processing the code snippets using machine learning models to generate rating scores for each of the code snippets, where each rating score applies to a corresponding ranking parameter. The method includes generating a combined score for each of the code snippets by combining the rating scores for the code snippet according to the weight values assigned to the corresponding ranking parameters and generating and presenting a user interface including an ordered list of the code snippets based on the combined scores for the code snippets.

Machine learning is used for pre-submit test selection. A commit including a new code change is obtained and a feature vector is generated based on the code change and historical test information. A ranking pre-submit tests is determined by providing the feature vector to a learning algorithm. One or more pre-submit tests are selected based on the ranking and the new code change is tested using selected tests. An assessment of the new code change is determined based on the test results, which indicate whether the new code change is accepted or rejected for submission to the source code repository. Later, new submitted source code stored in the repository is tested using a set of post-submit tests to obtain post-submit test results. Then the learning algorithm is updated based on the assessment of the new code change and the post-submit test results to obtain an updated learning algorithm.

A method includes accessing a plurality of weight matrices of a machine learning model. The method also includes, for each weight matrix, decomposing the weight matrix into a U matrix, an S matrix, and a V matrix using singular value decomposition. The S matrix is a diagonal matrix, and a singular group corresponds to each element in the S matrix. The method further includes, for each weight matrix, determining an importance score of each singular group. The importance score of the singular group represents a change in loss if the singular group is removed from the machine learning model. The method also includes, for each weight matrix, ranking the singular groups across the plurality of weight matrices based on the importance scores. In addition, the method includes, for each weight matrix, identifying one or more of the singular groups to prune based on the ranking of the singular groups.

In various examples, a ranking is generated for a set of computing instances based on predicted metrics associated with computing instances. For example, a prediction model estimates various system performance metrics based on information associated with a workload and configuration information associated with computing instances. The system performance metrics estimated by the prediction model are used to rank the set of computing instances.

Various embodiments of the present disclosure provide computer forecasting techniques for initiating presentation of an interactive user interface. The techniques may include receiving one or more candidate prediction-based actions and generating a plurality of causal risk-based impact scores with respect to a candidate prediction-based action. The techniques include generating a plurality of causal quality-based impact scores and an action sequence for a plurality of evaluation entities and generating a causal net impact score based on (i) an aggregation of the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence. The techniques include generating a sequence ranking for the action sequence and initiating a presentation of an interactive user interface reflective of the action sequence and the sequence ranking.

The invention discloses a model-based task instruction management method and device, a storage medium and equipment, and belongs to the technical field of deep learning. Obtaining text data of each task to be executed, wherein the text data comprises description information, demand information and supplementary information of the tasks and log information of historical tasks; extracting key information of each task from the text data by utilizing a large language model, wherein the key information comprises an emergency parameter, a target parameter, a resource demand parameter and a time limit parameter of the task; generating a corresponding task instruction based on the key information of each task by utilizing a generative model; combining similar task instructions in the plurality of task instructions; and performing priority ranking on the combined task instructions, and performing execution scheduling on the ranked task instructions. According to the method, the generation requirement of the high-complexity task instruction can be met, the execution efficiency of the instruction and the resource utilization rate are improved, and execution scheduling of the task instruction is adjusted in real time according to resources.

A method can include receiving log data for different types of logs; identifying a portion of the log data that corresponds to a type of formation; defining combinations of the portion of the log data that correspond to the type of formation; implementing a machine learning model that generates scores for the combinations, where each of the scores indicates an ability of each of the combinations to predict one or more target logs therein as selected from the different types of logs; and outputting, based on a ranking of the scores, at least a top ranked one of the combinations that corresponds to the type of formation.

A system and method for extracting data from large language model (LLM) outputs, including: training a model using labeled data items to assign rankings to LLMs; selecting, by the trained model, one or more of the LLMs based on the rankings; sending an LLM prompt to selected models; and outputting, by the model, a refined response to the prompt based on responses to the prompt by the LLMs. Some LLM prompts according to some embodiments may include different sets input parameters of different types-such as, e.g., a set of block parameters and a set of editorial parameters. In some embodiments, a model or LLM may be updated or retrained using a reinforcement learning approach and based output items or refined responses generated by that model or LLM-which may for example be scored or ranked and used in combination with reward or cost functions to update model parameters.

This application is directed to systems and methods for ranking items in response to a query. In some embodiments, a disclosed method includes identifying a plurality of items to be provided in response to a first query and determining a plurality of messages. A first item of the plurality of items was previously engaged by former users in response to a plurality of second queries, and corresponds to a first message determined based on the second queries. The method further includes determining a query feature vector of the first query based on the messages including the first message of the first item, ranking the items associated with the first query into an ordered item list based on the query feature vector, and in response to receiving the first query from a next user, presenting information of the items based on the ordered item list on an electronic device.

The invention relates to the technical field of computer software, in particular to a vertical search engine fine arrangement method and system based on multi-dimensional features. The method comprises the steps of obtaining recalled resource information, and performing feature extraction based on the obtained resource information, including extracting document features, resource popularity features, column features and user behavior features of the resource information; performing classification model weighting according to the extracted resource information features, and performing resource popularity weighting according to the extracted resource information features, including setting a resource popularity hyper-parameter for adjusting the weight of the resource popularity; and carrying out model integration on the classification model weighting result and the resource popularity weighting result, carrying out fine arrangement scoring by utilizing the weight after model integration, and sorting the recalled resources according to the final scoring result. According to the method, multi-dimensional features are comprehensively considered, and the search results are reasonably weighted, so that accurate search result display is realized.

Disclosed are techniques for using machine learning models to more reliably predict likelihoods of application failure. A model is trained to identify and display events that may cause high severity application failures. Logistic regression may be used to fit the model such that application features are mapped to a high severity event flag. Significant features of applications that relate to the high severity flag may be selected using stepwise regression. The identified applications may be displayed on a graphical user interface for review and reprioritization. Information may be ranked and displayed according to multiple different ranking criteria, such as one ranking generated by a first model, and another determined by one or more users. The multiple ranking criteria may be used to inform steps taken, and / or to retrain or tune the parameters of the model for subsequent predictions or classifications.

Systems and methods for automatically generating search list rankings of code snippets are provided. An exemplary method includes searching source code repositories to identify code snippets in response to a search query, assigning weight values to ranking parameters, and processing the code snippets using machine learning models to generate rating scores for each of the code snippets, where each rating score applies to a corresponding ranking parameter. The method includes generating a combined score for each of the code snippets by combining the rating scores for the code snippet according to the weight values assigned to the corresponding ranking parameters and generating and presenting a user interface including an ordered list of the code snippets based on the combined scores for the code snippets.

Systems and methods for ranking user interface elements using explainability vectors. The system receives training data for a ranking machine learning model. The training data comprises values for a first set of features. The system trains the ranking machine learning model using the training data. The system processes the ranking model to extract an explainability vector. Based on the explainability vector, the system processes the first set of features to generate a second set of features. The system processes the second set of features and the output of the predictive machine learning model to generate an explanative factor and trains a ranking model using the explanative factor and a third set of features. The system receives as output from the ranking model a vector indicating display positions and rankings of user interface elements.

The invention relates to the technical field of artificial intelligence and the field of financial science and technology, and discloses a task solution strategy generation method, which comprises the following steps: acquiring a task demand, constructing preliminary prompt information containing roles, task requirements and scheme generation guidance based on the task demand, inputting the preliminary prompt information into a preset language model to generate a preliminary solution, and generating a task solution strategy; analyzing the preliminary solution to extract a plurality of solution strategies, constructing review prompt information for each solution strategy and inputting the review prompt information into a preset language model to generate a score, determining a target solution strategy of which the score exceeds a preset score threshold, generating specific execution content for each target solution strategy, analyzing the execution content and determining an execution score, and performing priority ranking on the target solution strategies based on the execution scores to generate a priority list. According to the method, a better solution is generated through analysis of automatic processing task requirements, generation of solution strategies and priority ranking, customer requirements are quickly responded, and the development cost is reduced.

A system and method for extracting data from large language model (LLM) outputs, including: training a model using labeled data items to assign rankings to LLMs; selecting, by the trained model, one or more of the LLMs based on the rankings; sending an LLM prompt to selected models; and outputting, by the model, a refined response to the prompt based on responses to the prompt by the LLMs. Some LLM prompts according to some embodiments may include different sets input parameters of different types—such as, e.g., a set of block parameters and a set of editorial parameters. In some embodiments, a model or LLM may be updated or retrained using a reinforcement learning approach and based output items or refined responses generated by that model or LLM-which may for example be scored or ranked and used in combination with reward or cost functions to update model parameters.

The invention discloses a ranking list determination method and device which can be applied to the field of data processing, and the method comprises the steps that ranking list data are efficiently stored and ranked through an ordered set of Redis, each user serves as an element, and the invitation amount of each user serves as a score. When each preset updating time is satisfied, whether the invitation amount information of the target user is changed or not is checked. If the current ranking information changes, the ranking of the target user is recalculated according to the new invitation quantity information, and the offline ranking list is dynamically updated to ensure that the updated ranking list can accurately reflect the current ranking information of the user. Namely, the ranking list is stored through redis, and before the next timed task is executed, the ranking list is kept unchanged, so that the write-in frequency of the database can be reduced, the storage cost caused by frequent write-in is reduced, and resource waste is avoided.

The application discloses a kind of estimation ranking method, device, computer equipment and storage medium, belong to artificial intelligence technical field.The application is based on the exposure event carrying first behavior and the exposure event not carrying first behavior to construct first sample set and based on the exposure event carrying second behavior and the exposure event not carrying second behavior to construct second sample set, based on first sample set and second sample set fitting training to multi-task learning model, obtain estimation ranking model, import user data into estimation ranking model and obtain information estimation ranking result, complete information push based on information estimation ranking result.In addition, the application also relates to blockchain technology, and user data can be stored in blockchain.The application is fitted by first behavior feature and second behavior feature in multi-task learning model, solves sample selection bias problem, improves model generalization ability, improves prediction accuracy and user experience.

This invention discloses a method for ranking reserve projects for new power grid infrastructure, comprising: constructing a ranking indicator system from five dimensions: power grid development, security and supply, service quality, efficiency and effectiveness, and green and low-carbon development; collecting raw indicator data and distinguishing indicator types; obtaining objective weights for the indicators through a combination of two objective weighting methods; constructing a matrix by standardization transformation and weight combination; calculating relative proximity to form an objective ranking; subsequently integrating expert rankings; obtaining a subjective consensus ranking by constructing a pairwise preference matrix and an integer programming model; and finally, calculating standardized scores and weighted fusion of the subjective and objective ranking results using a weighted aggregation method to determine the ranking position of the reserve projects. This method balances data objectivity and expert experience, covers the core requirements of projects from multiple dimensions, has a rigorous ranking logic, and effectively improves the scientificity and accuracy of ranking reserve projects for new power grid infrastructure.

The invention discloses an automatic project evaluation method. The method comprises the steps of obtaining a demand file and a target project file of each associated object; performing classification processing on each target project file according to a machine learning algorithm to obtain structural data and non-structural data; constructing a quantization rule base according to the demand file, and performing matching processing on the quantization rule base and the structure data to obtain quantifiable data; performing semantic similarity analysis according to the quantifiable data and the demand file, and obtaining a first sorting result of each target project file according to an analysis result; inputting the non-structural data into a pre-trained sorting model to obtain a second sorting result; and generating a log file of each associated object according to the first sorting result and the second sorting result, and obtaining an evaluation result of each associated object according to the log file. According to the automatic project evaluation method, the project evaluation efficiency is improved.

Various embodiments of the present disclosure provide computer forecasting techniques for initiating presentation of an interactive user interface. The techniques may include receiving one or more candidate prediction-based actions and generating a plurality of causal risk-based impact scores with respect to a candidate prediction-based action. The techniques include generating a plurality of causal quality-based impact scores and an action sequence for a plurality of evaluation entities and generating a causal net impact score based on (i) an aggregation of the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence. The techniques include generating a sequence ranking for the action sequence and initiating a presentation of an interactive user interface reflective of the action sequence and the sequence ranking.