11 results about "Programming knowledge" patented technology

The application discloses a student programming answer prediction method combining semantic understanding and structured modeling, collects student programming homework data and pre-processes the data to obtain data samples, translates the data samples to obtain an English feature set; uses low-rank adaptation to fine-tune a pre-trained semantic understanding model, constructs input features, and predicts an answer correctness probability; performs graph neural network structured modeling based on the association relationship between problems and concepts, performs message propagation and feature updating to obtain an embedding set, inputs the embedding set and student embedding into a prediction layer to obtain a probability of answering a question correctly; and inputs the probability of answering a question correctly and the probability of answering a question correctly after fusion into a multilayer perceptron for nonlinear mapping to predict the probability of answering a question correctly. The application solves the problems of difficulty in effectively processing non-standardized codes submitted by students, insufficient semantic understanding of question texts and low prediction accuracy in programming knowledge tracking, and provides a scientific basis for personalized teaching and learning resource scheduling.

This application provides a no-code orchestration system based on intelligent agents, belonging to the field of no-code programming and multi-agent collaboration. It addresses the problem that building multi-agent collaborative systems in related technologies requires specialized programming knowledge and has a high technical threshold. The system dynamically organizes intelligent agents through a group collaboration container building module that responds to no-code operations such as visual drag-and-drop. A dialogue-driven intelligent agent collaboration network module parses natural language commands and generates task orchestration plans. A visual rule-protocol conversion engine module converts graphically configured interaction rules into underlying protocols to schedule intelligent agent collaboration. This allows non-technical personnel to quickly build and command complex multi-agent collaborative systems in an intuitive no-code manner, effectively reducing the technical threshold and improving the development and deployment efficiency of intelligent collaborative applications.

The invention provides a programming learning scene-oriented knowledge tracking method and system, and belongs to the field of programming education, and the method comprises the steps: S1, constructing a hierarchical multi-label code snippet classification network, carrying out the knowledge point labeling of an input code through the hierarchical multi-label code snippet classification network, and generating a correction Q matrix; s2, based on a code representation learning method and code embedding of the students, calculating the similarity between codes continuously submitted by the students to obtain a constructed code similarity matrix; and S3, constructing a heterogeneous graph based on the corrected Q matrix, constructing a heterogeneous graph network comprising a spatial feature extraction module and a time feature extraction module, and tracking the student programming knowledge state and predicting the next submission performance by using the heterogeneous graph network. According to the method, the dynamic tracking and performance prediction of the knowledge concept mastering state of the student programming are realized.

The system according to this embodiment aims to automate tasks by generating scripts based on voice commands, even without programming knowledge. [Solution] The system according to the embodiment comprises a reception unit, a generation unit, an execution unit, and a provision unit. The reception unit receives voice instructions. The generation unit analyzes the voice instructions received by the reception unit and generates a script. The execution unit executes the script generated by the generation unit. The provision unit provides the results executed by the execution unit.

The application discloses a kind of numerical control programming field knowledge graph construction method, steps are as follows: numerical control programming knowledge collection and classification;Numerical control programming field knowledge graph mode layer construction;Numerical control programming field knowledge graph data layer construction;Numerical control programming knowledge knowledge fusion and reasoning;Numerical control programming knowledge graph knowledge storage;Numerical control programming field knowledge graph visualization.The application analyzes the representation form of numerical control programming field knowledge, proposes a variety of strategies Knowledge extraction method, effectively covers the entire numerical control programming field.Therefore, the numerical control programming field knowledge graph constructed based on the application is larger in scale, and the knowledge is more complete.The application uses the knowledge reasoning method based on ontology and semantic rules to mine the semantic relationship between the constructed numerical control programming knowledge graph, which ensures the knowledge quality of the constructed knowledge graph and further expands the scale of the numerical control programming field knowledge graph.

The invention provides an intelligent programming teaching-assistant recommendation method and device based on a large language model. The method comprises the steps of obtaining historical data of programming knowledge; generating a macroscopic path of programming learning based on historical data; obtaining a pre-trained programming teaching-aiding recommendation model, wherein the programming teaching-aiding recommendation model is obtained based on macroscopic path training of programming learning; and based on the programming teaching-assistant recommendation model, generating a programming learning scheme corresponding to the knowledge progress of the current student. According to the method, the macroscopic path is generated through the historical data, and programming path planning is achieved; according to the method, the programming teaching-aiding recommendation model is trained, and the corresponding programming learning scheme is generated based on the personal progress of the student, so that the stiff programming teaching-aiding recommendation is avoided, and the personalized learning requirements of the student are met.

The embodiment of the invention provides a script generation method for liquid processing operation and an automatic liquid processing method.The script generation method comprises the steps that based on an experiment description statement input by a user, an operation matching result matched with target liquid processing operation is determined from a database, the database comprises code templates of various types of liquid processing operations, and the experiment description statement comprises a target liquid processing operation; and inputting the operation matching result into a script generation model to generate an operation script which is of the target liquid processing operation and conforms to the preset grammar. Therefore, the use threshold of the liquid work station is greatly reduced, and a user can conveniently operate the liquid work station without mastering professional programming knowledge to complete related liquid treatment operation.

The application belongs to the technical field of computers and specifically discloses a programming knowledge tracking method based on code prediction enhancement, which comprises the following steps: based on historical programming questions and corresponding historical programming answers, extracting embedded representations, obtaining historical question embedded representations and historical answer embedded representations; based on the historical question embedded representations, the historical answer embedded representations and the embedded representations of target programming questions, predicting the embedded representations of the source code input by the measured object for the target programming questions through a multi-head attention mechanism model, and taking the predicted embedded representations of the source code as programming answer prediction results; and based on the embedded representations of the target programming questions, the programming answer prediction results and the current knowledge state of the measured object, predicting the answer correctness probability of the measured object for the target programming questions, wherein the current knowledge state is obtained by analyzing the time sequence evolution of the knowledge state based on the historical programming answers of the measured object. The application can accurately predict the programming performance of learners.

To solve the problem that it is difficult to construct an original business system when there is no staff skilled in knowledge and experience of programming.SOLUTION: A programming support method comprising: a step of storing a source code of an action set for each of a plurality of existing screens of a plurality of existing business applications; a grouping step of grouping the plurality of existing screens into a plurality of groups based on images of the plurality of existing screens and attributes of the plurality of existing screens; The method includes a group specifying step of specifying the group corresponding to the screen, and a providing step of providing a source code of the action set in the existing screen belonging to the specified group.SELECTED DRAWING: Figure 9

The invention provides a group game strategy generation method and a group game method, and relates to the technical field of artificial intelligence, a game strategy is explicitly defined as an executable strategy code with a clear structure, so that people can directly read decision logic of an intelligent agent, strategy interpretability is achieved, a user can debug and correct logic vulnerabilities conveniently, and user experience is improved. And the method can also be applied to the field with high requirements on safety and compliance. Moreover, according to the method, programming knowledge of a large language model is utilized, so that the generated strategy population has natural diversity in algorithm structure, strategy codes with high-level competitiveness can be directly written in the initial stage of the game, zero-sample cold start is realized, and computing power consumption and time cost are greatly reduced. The method can be suitable for application scenes with high requirements for strategy interpretability, logic complexity and diversity.

The invention discloses a semantic understanding and structured modeling combined student programming answer prediction method, which comprises the following steps: collecting student programming homework data, preprocessing to obtain a data sample, and translating to obtain an English feature set; performing fine tuning on the pre-trained semantic understanding model by using low-rank adaptation, constructing input features, and predicting an answer correct probability; performing graph neural network structured modeling based on the association relationship between the question and the concept, performing message propagation and feature updating to obtain an embedded set, fusing the embedded set and student embedding, and inputting the fused embedded set and student embedding into a prediction layer to obtain the probability of answering the question by the student; and fusing the question answering correct probability and the question answering probability of the student, inputting the fused result into a multi-layer perceptron for nonlinear mapping, and predicting the question answering correct probability. The problems that in programming knowledge tracking, non-standardized codes submitted by students are difficult to effectively process, topic text semantic understanding is insufficient, and prediction precision is low are solved, and a scientific basis is provided for personalized teaching and learning resource scheduling.