265 results about "Tree node" patented technology

The invention discloses a policy service recommendation method and system based on a large language model, and relates to the field of policy information, and the method comprises the steps: analyzing a policy file, extracting policy terms with a numbering structure, building a unique identifier, and forming a term node set; identifying explicit or implicit reference behaviors in terms, constructing a reference path, and generating a reference directed graph; constructing a clause mapping tree based on the target clauses; mapping tree nodes are extracted to form structured prompt information, and the structured prompt information is embedded into large language model input; generating a policy report paragraph containing clause reference content; and outputting the structured policy text with clause reference chain closure and recommending the structured policy text to the user. According to the method, the clear reference context combined with the user information can be provided for the large language model, the large language model is guided to generate a text with a reasonable structure and complete terms for recommendation, and the problem of inaccurate generation when an existing model processes a complex reference structure is solved.

The invention relates to the technical field of routing context connection, in particular to a zero-code visualization and automatic routing method for large-model servitization orchestration, which comprises the following steps of: analyzing a user session text, extracting a core entity and an intention dependency relationship, constructing an entity intention graph, compressing to generate a state snapshot, and distributing a global session ID (Identity); the method comprises the steps that a snapshot is bound to a distributed routing decision tree node and stored, cross-node state continuity is established, when a new request is received, the snapshot and new session semantics are fused, a routing weight offset correction decision table is calculated, and a routing decision logic flow is generated by dragging an interface configuration rule and strategy. The method improves the service matching degree, reduces the development threshold, and is suitable for agile construction of large-model collaborative application.

The invention provides an authentication document form generation method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining an authentication document; converting the authentication document into an XML (Extensible Markup Language) file so as to map authentication document data to an XML tree node of the XML file; standardizing the XML file through the global unique identifier and the tag system to obtain a standardized XML file; in combination with the standardized XML file, template candidate items conforming to different manageable data templates in the authentication document are extracted respectively; in combination with the template candidate item, mapping the authentication document to a target form; and establishing a bidirectional mapping relationship between the target form and the authentication document to realize bidirectional conversion between the target form and the authentication document and complete the form processing of the authentication document. Through establishment of the bidirectional mapping relation, synchronous updating of the form and the document content can be achieved, errors caused by manual operation are avoided, and automation, accuracy and practicability of document conversion are remarkably improved.

The invention relates to the technical field of natural language processing, and discloses a question retrieval method and system, electronic equipment and a storage medium. The method comprises the steps of performing semantic understanding and decomposition processing on an original question to obtain a plurality of candidate sub-questions; constructing an initial problem tree structure by taking the original problem as a root node and the candidate sub-problems as sub-nodes; rationality scores of the child nodes are determined, and the rationality scores represent rationality of candidate sub-problems corresponding to the child nodes; determining a relevance score between the nodes, the relevance score representing a semantic repetition condition between the nodes; based on the rationality score and the relevance score, optimizing the initial problem tree structure to obtain a target problem tree structure; and performing retrieval based on the target question tree structure to obtain answer data for the original question. By means of the method, rationality and relevance evaluation can be conducted on the problem tree nodes, and architectural disassembly and semantic hierarchical management of complex problems are achieved.

The invention discloses a dual-path three-dimensional point cloud compression reconstruction method fusing multi-scale geometric features. The method comprises the following steps: extracting representative skeleton anchor points from original point clouds by adopting farthest point sampling; predicting an octree node occupancy state by using an anchor encoder, and performing entropy coding to generate a skeleton bit stream; a skeleton condition feature encoder is introduced, probability distribution of local features is predicted by using skeleton context priori, cross-path context association is constructed, and a feature bit stream is generated; in the decoding stage, the sparse skeleton point set is restored from the skeleton bit stream, local features are recovered from the feature bit stream in combination with the skeleton structure context, fine-grained reconstruction is carried out on the point cloud through a geometric fusion point cloud up-sampling module, and finally the high-fidelity point cloud is obtained. According to the invention, the point cloud compression effect and reconstruction quality can be improved. The method can be widely applied to the point cloud field.

The invention relates to a database Schema version management method and device, computer equipment and a storage medium. The method comprises the following steps: responding to received target Schema tree structures sent by each user terminal, generating a combined target Schema tree structure according to each target Schema tree structure, and performing format conversion on the combined target Schema tree structure to obtain a modified Schema file; performing tree node traversal comparison according to the combined target Schema tree structure and the to-be-modified Schema tree structure to obtain corresponding distinguishing tree nodes; the distinguishing tree nodes are tree nodes on the combined target Schema tree structure which are different from the Schema tree structure to be modified; and determining a version modification type of the modified Schema file according to each distinguishing tree node, generating a semantic version identity identification number of the modified Schema file according to the version modification type, and storing the modified Schema file into a database according to the semantic version identity identification number. By adopting the method, the efficiency can be improved and version management chaos can be avoided.

The invention discloses a real-time analysis and view editing method and system based on streaming data, and the method comprises the steps: receiving streaming data which comes from a large language model and is organized in an editing protocol format, carrying out the real-time lexical analysis of the streaming data, and decomposing the streaming data into a semantic unit sequence through a semantic pattern recognition technology; performing grammar analysis on the semantic unit sequence, and converting the semantic unit into a structured editing instruction for modifying an existing rendering tree node according to an instruction pattern recognition mechanism and a target object analysis mechanism; converting the structured editing instruction into an atomic instruction sequence executable by a rendering tree editing interface; and modifying the rendering tree in real time according to the atomic instruction sequence and triggering update rendering of the view. According to the method, the content generated by the AI is analyzed in real time by adopting a streaming output mechanism, so that a user can observe the generation process of the view in real time, local redrawing of the view is realized by using an editing protocol, and the view editing efficiency is greatly improved.

The invention discloses a modeling analysis and calculation optimization method for operator performance on an intelligent chip. The modeling analysis and calculation optimization method comprises the following steps: carrying out graph representation modeling on an intelligent chip hardware architecture and an interconnection mode; designing a link mechanism of operator loop tree nodes, so that operator loops are described and mapped in a hardware architecture model represented by a graph; and performing delay performance analysis on the single operators and the fusion operators in various combination forms by using a pipeline analysis method to obtain the performance of the operators under a hardware architecture, and further performing calculation optimization acceleration. By the adoption of the technical scheme, sufficient guidance can be provided for the initial stage of optimization of the hardware architecture, guidance is provided for optimization space exploration of operators, and therefore the calculation performance of intelligent chip hardware is improved.

The invention discloses a dynamic octree mapping method and system based on probability semantic fusion and a storage medium, and mainly solves the problem that a traditional octree map lacks semantic information and cannot support an advanced cognitive task. According to the implementation scheme, the method comprises the following steps: acquiring an RGB image, depth information and pose data through a sensor; extracting pixel-level semantic probability distribution by using a pre-trained semantic segmentation network; geometric information in the perception data is projected to a three-dimensional space based on the pose, and a three-dimensional observation point with semantic information is formed; performing ray projection on octree nodes, respectively updating geometric occupancy states of penetrated non-end-point nodes, and synchronously updating geometric occupancy states and semantic belief distribution of end-point surface nodes by using three-dimensional observation point data; semantic information is continuously optimized through Bayesian probability fusion, and dynamic pruning and map optimization are realized based on semantic entropy. According to the method, the map storage efficiency and the calculation speed can be remarkably improved while refined and real-time updated geometric and semantic information is given. The method can be used for map construction and storage of visual SLAM positioning, and fine three-dimensional object modeling and environment information description.

A tree-structured world state storage method, a recovery method, and a computer device. The storage method comprises: with respect to generated / updated tree nodes, according to a logical page structure obtained by division, generating corresponding memory pages, the memory pages being used for representing content of all corresponding tree nodes in the logical pages, wherein the logical pages are results obtained by aggregating several upper and lower adjacent tree nodes according to a node association relationship of the tree structure, each logical page contains at least one tree node, and the tree nodes in different logical pages are mutually exclusive; on the basis of changes in the contents of all the tree nodes in consecutive versions of the logical pages, generating base pages and delta pages; and persisting the base pages and the delta pages in one database without distinguishing between a currentDB and a historyDB.

The invention discloses a medical drug knowledge RAG optimization method based on a Trie tree, and relates to the field of artificial intelligence and medical drug information processing. The method comprises the following steps: constructing a dual Trie tree knowledge base in the field of medical drugs (a term Trie tree: storing drug / disease / medication attribute terms and ATC / ICD-11 codes; the content Trie tree is used for storing medicine specification hierarchical content and efficacy weight), carrying out preprocessing segmentation on a medical medicine ultra-long text and associating Trie node attributes, and adopting multi-stage retrieval of Trie prefix matching and BioBERT semantic retrieval to recall knowledge blocks; the context (father node-chapter overview, child node-refining rule and brother node-security association information) is expanded based on the hierarchical relationship of the nodes of the Trie tree, finally a response with accurate reference (drug name + ATC code + specification chapter) is generated, and drug conflict is avoided through consistency check. The medical drug knowledge retrieval accuracy and response safety are improved, and the method is suitable for scenes such as intelligent medication consultation and clinical medication aid decision making.

The invention provides a rapid demand customization method based on scene unit identification and context fusion, which comprises the following steps: S1, carrying out coarse-grained intention identification to obtain a first-level coarse-grained intention label; s2, acquiring multi-source context data in a corresponding time range to form a standardized context data set; s3, processing the standardized context data set to obtain structured scene data and a locked demand path; s4, complementing missing parameters through intention tree node mapping and automatic filling logic, and generating a complete structured demand configuration template; s5, generating an optimal component combination and a low-code configuration template; and S6, generating an interpretable description and a recommendation reason based on the optimal component combination, and displaying the interpretable description and the recommendation reason to a user in combination with a low-code configuration template. According to the method, intelligent tracking, dynamic response and automatic recommendation of the whole process of the demand are realized in an industrial MOM scene, so that production management is more refined and efficient.

The present disclosure relates to a mesh reconstruction method, a device and a storage medium. The mesh reconstruction method includes constructing a data structure according to three-dimensional scanning data and a preset mesh side length. Node information of a target tree node among a plurality of tree nodes included in the data structure is determined. A level of a mesh to be extracted of the target tree node is determined according to the target data. Once a scalar field of the target tree node is calculated according to the node information of a neighboring tree node corresponding to the target tree node and the node information of the target tree node, a reconstructed mesh model is obtained according to the scalar field.

Provided in the present invention is a multi-level equipment cross-platform fault knowledge recommendation and adaptive transfer method, comprising: constructing an association relationship between product structure tree nodes and a fault knowledge base; calculating the similarity between first-level nodes of a start-domain product structure tree and a target-domain product structure tree, so as to obtain a similar first-level node pair set; calculating the similarity between child node sets of a similar parent node pair, and outputting a similar second-level node pair set; repeating the previous steps on a next-level node until a terminal node is reached, and outputting a final similar node pair set; calculating the fault case knowledge similarity between every two nodes in the similar node pair set, and excluding, from the similar node pair set, node pairs having relatively low similarities; and selecting a fault knowledge type that needs to be transferred, acquiring and copying, from node pairs having relatively high similarities, fault knowledge data of a start domain, modifying into a target domain a node to which related data is bound, and writing the related data into a database. The present invention can solve the problem of rapid transfer of fault knowledge between different items in avionics equipment.

The invention discloses a data processing method and device, equipment, a storage medium and a product, the method is applied to integrated circuit electronic design automation software, and the method comprises the following steps: obtaining a digital circuit design file; analyzing the digital circuit design file to generate a corresponding netlist data structure; responding to an object query request sent by a user, accessing the netlist data structure, and searching a data item matched with a target query object specified by the user in the netlist data structure; based on the data item matched with the target query object specified by the user and found in the netlist data structure, the object query result is generated, the netlist data structure is adopted as a data processing basis for object query, hierarchical recursive traversal does not need to be carried out, branch relations among multi-way tree nodes do not need to be processed, and the data query efficiency is improved. And the query efficiency is obviously improved.

The invention discloses a data burying point and analysis method and system for cross-platform dynamic configuration, and relates to the technical field of data burying point, a server determines a candidate parameter set of burying point configuration based on an event model, performs optimization through a swarm intelligence algorithm, generates dynamic burying point configuration including a target control topological neighborhood feature vector, and transmits the dynamic burying point configuration to a server; issuing to a terminal; the terminal calculates the geometric structure similarity between view tree nodes and topological neighborhood feature vectors, and binds the nodes exceeding a threshold value as a buried point trigger source; when a burying point trigger source is activated, obtaining an operation sequence of a current context and sampling strategy generation event data, injecting a logic precursor identifier and a statistical weight factor, and sending the data to a server through local caching and batch reporting; and the server performs logic causal rearrangement, constructs a statistical analysis model to identify a statistical weight factor, executes inverse probability weighted calculation and outputs an unbiased analysis result, so that the problem of burying point failure caused by version iteration is solved, and the robustness and adaptability of burying point configuration are improved.

The invention provides a longitudinal federated learning method based on a fully homomorphic ScureBoost model, which comprises the following steps that: an active node calculates a first derivative g and a second derivative h for each sample, encrypts and packs the first derivative g and the second derivative h by using a fully homomorphic encryption algorithm CKKS and sends the first derivative g and the second derivative h to a passive node; the passive node uses a security weighted quantile algorithm to find candidate segmentation points, and gradient aggregation is carried out on the encrypted derivative according to a result; the active node determines an optimal split point by utilizing a bucket subtraction method between tree nodes based on the aggregation gradient sent by the passive node and the own aggregation gradient; the passive node divides sample data according to the optimal splitting point; the active node updates the node information and the sample distribution condition of the split tree model; repeating the steps to the maximum depth of the tree to obtain a complete tree model; and after the model training is finished, carrying out PSI-based model federated reasoning. According to the method, the model training efficiency can be improved, the side channel attack resistance and security are improved, the communication traffic is reduced, and the practical application value of the fully homomorphic encryption algorithm is proved.

This invention discloses a lightweight BIM data transfer method based on 3DE, comprising the following steps: S1, acquiring components; S2, acquiring structure tree nodes and updating the transfer; S3, acquiring components and generating hierarchical folders; S4, acquiring lightweight geometric files and material map files, and associating the lightweight geometric files and material map files; S5, acquiring lightweight attribute files, and associating the lightweight geometric files and lightweight attribute files; S6, compressing and outputting lightweight data files. This invention solves the problem of information loss during BIM lightweight transfer caused by traditional single-feature lightweight methods, and ensures the accuracy of the association between geometric information, material maps, and attribute information of the transferred model, as well as the consistency of coordinate information and structural organization of the transferred model, thereby improving the quality and efficiency of BIM lightweight transfer.

The invention provides a full life cycle carbon emission evaluation method based on a transformer substation prediction model, and the method comprises the steps: building an operation structure tree based on the distribution characteristics and function relevance of a physical entity object of a transformer substation by taking a carbon emission entity object of each full life cycle stage as a tree node; mapping physical attributes and dynamic data of each carbon emission entity object to a digital twin platform based on node levels in the operation structure tree to obtain virtual nodes; associating the dynamic data of the virtual node with the static data of each full life cycle stage based on the node identifier in the running structure tree to obtain a full life cycle data chain; based on the associated data of each tree node in the full-life-cycle data chain, calculating a carbon emission result in each full-life-cycle stage; and according to a father-child node relationship of the running structure tree, upwards aggregating carbon emission results of lower-level nodes of each tree node to obtain a total carbon emission evaluation amount. According to the invention, the accuracy of a carbon emission evaluation result is improved.

The invention provides a document chapter jumping method and device, equipment and a storage medium, and the method comprises the steps: preprocessing a target document to be opened by electronic equipment, and obtaining a structured text of the target document; performing structure recognition on the structured text, and constructing a chapter tree structure of the structured text; generating positioning index information for each tree node in the chapter tree structure; when the electronic equipment opens the target document, an interactive sidebar is generated on a display interface, and the interactive sidebar is used for displaying the chapter tree-shaped structure; and when detecting that any tree node in the chapter tree structure is triggered on the interactive sidebar, based on the positioning index information of any tree node, skipping to display chapter content under the hierarchical chapter title corresponding to any tree node on the display interface. The method can support the rapid jumping to the specified chapter content, and improves the teaching rhythm of teachers.

A mining method performed by node devices connected to a blockchain network, comprises: a step in which a specific node device acquires a transaction including a first block having a first block number and a first trie node related to specific state data of the blockchain network, and state change data for changing the state of a specific account; a step in which the specific node device generates a second block including a second trie node and a second block number, the second trie node being generated based on the transaction and the first trie node; and a step in which, if the hash value of a second node corresponding to a state path-being a path from a specific leaf node corresponding to the specific account to a specific root node-among one or more second nodes included in the second trie node satisfies a predetermined criterion.

The invention provides a heterogeneous multi-agent kinematics constrained path-attitude planning method and product, and relates to the technical field of path planning. The method comprises the following steps: acquiring steering constraints of a plurality of heterogeneous agents in a path planning task, and constructing a respective transition state topological graph of each agent based on the steering constraints; performing space-time DL-A * algorithm search on the bottom layer based on the respective transition state topological graph of each agent to obtain the respective optimal state path of each agent; constructing a constraint tree, selecting a constraint tree node with the minimum current cost from the constraint tree, detecting conflicts included in a group of transfer states planned for different agents in the node, screening out conflicts with the highest priority, and designing state-time constraints on the top layer of the involved agents; adding child nodes to the current constraint tree node; and iteratively performing conflict detection until a group of bottom layer conflict-free solutions are obtained to serve as a final state path of a plurality of heterogeneous agents. According to the method, the constraint of the steering angle can be realized, the search space is effectively reduced, and conflict-free paths can be planned for intelligent agents with different steering capabilities in a scene at the same time.

The invention relates to the technical field of computers, in particular to a test case generalization method for automobile network and data security, which comprises the following steps: collecting and preprocessing multi-source heterogeneous data, fusing the data to construct a dynamic risk feature model to extract an attack vector, and constructing a layered scene space based on real-time topology of a vehicle-mounted network; a two-channel generalization engine is utilized, unknown threat scene cases are generated by means of a structure causal model, and variation is injected into nodes of a protocol syntax tree to generate fuzzy test cases so as to enrich case types; a reinforcement learning framework is introduced to dynamically optimize a test path, and a closed-loop feedback mechanism is adopted to expand high-risk scene nodes, so that the multi-source heterogeneous data processing capability is effectively improved, the unknown threat simulation capability is enhanced, the test efficiency is improved, and the problem that a test method in the prior art is incomplete in multi-source heterogeneous data processing and poor in reliability is solved. The technical problems of influence on subsequent analysis accuracy, single test case generation type, lack of unknown threat simulation and low test efficiency in the prior art are solved.

This invention discloses a method and system for managing and automatically formatting slide content based on a tree structure. By abstracting different types of materials such as text, images, tables, code snippets, and mathematical formulas into tree nodes with parent-child hierarchical relationships, users can freely add, delete, move, or mark nodes in a visual interface to achieve rapid merging or splitting of multi-level documents. The system then maps several nodes into multiple slides based on the node merging or splitting markers; and automatically formats the pages during the output stage, including overflow detection and font scaling, adaptive image proportioning, line wrapping for tables or code snippets, and format conversion of mathematical formulas. This approach significantly reduces the burden of manually adjusting text and images, maintains consistency and aesthetics in slide layout across different resolutions, and is suitable for multi-level, text-rich, and large-scale presentation document scenarios.

The invention provides a real-time cargo loading decision-making method based on a tree structure and reinforcement learning. The real-time cargo loading decision-making method comprises the steps of constructing a buffer area module, a temporary backup space module, a DRL position planner, a tree structure moving sequence planner and a removal punishment mechanism. The remaining space and the current state are initialized, the tree structure is initialized, and the current state is a root node; recursively expanding nodes, judging whether to enter a removal stage, if not, selecting goods in the buffer area and the temporary backup as candidates, and if yes, acquiring a removable goods set to obtain candidate goods; calling a DRL position planner for candidate goods, expanding tree nodes and continuing recursion, judging whether the maximum depth is reached or not, if not, returning to recursion and expanding the nodes, and if yes, calculating accumulated rewards and future expectations of the nodes; and searching a state node with the highest space utilization rate, and calling a tree structure moving sequence planner and a removing penalty mechanism to generate an optimal action sequence. The space utilization rate of the container is increased.

A method for reading a world state of a tree structure, and a computer device. The method comprises: receiving a read state key and a version; on the basis of the state key and the version, matching a version and a page ID in a page identifier of a memory page in lexicographic order, and after matching, if the memory page comprises at least two layers of tree nodes, continuing to match a version and a node ID of a tree node below the uppermost layer of tree nodes in lexicographic order until a leaf node is found, the page ID comprising lexicographic content from a root node of the tree structure to an upper-layer tree node in the memory page, and the node ID comprising lexicographic content from the root node to the tree node; and reading a state in the leaf node.

The present invention discloses a decision tree product recommendation method based on two-party secure computing, which can be applied to data holders or model holders. The method includes: obtaining a secret sharing of the decision tree root node index and a lookup table; obtaining a secret sharing of the node information of the corresponding node on the decision tree through a selection protocol, thereby obtaining a secret sharing of the corresponding eigenvalue index, and obtaining a secret sharing of the corresponding eigenvalue through a lookup table; comparing the eigenvalue and the threshold through a conditional oblivious transfer protocol, thereby selecting a left subtree node index or a right subtree node index; based on the secret sharing of the selected node index, obtaining a secret sharing of the node information of the corresponding node on the decision tree through a selection protocol, thereby obtaining a secret sharing of the corresponding classification value, updating the secret sharing of the product recommendation result based on the secret sharing of the classification value, traversing the decision tree until a predetermined number of nodes are traversed to obtain the final product recommendation result.

This application provides a robot control method, apparatus, device, robot, medium, and product, belonging to the field of robotics technology. The method includes: generating a behavior tree based on task intent information corresponding to a user instruction; the task intent information represents the user's intention to instruct the robot to perform an operation task, the tree nodes of the behavior tree represent the subtasks of the operation task to be performed by the robot, and the node information of the tree nodes represents the execution conditions of the subtasks corresponding to the tree nodes; controlling the robot to perform the operation task based on the behavior tree; receiving task execution status information fed back by the robot during the execution of the operation task; and dynamically adjusting the behavior tree to obtain an updated behavior tree when the task execution status information represents an execution anomaly in the operation task, and controlling the robot to continue performing the operation task based on the updated behavior tree. This application enables the robot to operate efficiently in dynamically changing and complex scenarios.

The present invention relates to the technical field of intelligent voice customer service, and a method and system for accurately identifying the intention of a voice customer service based on deep learning, including: determining whether there is only one intersection tree node in the subsequence of intersection tree nodes. If not, identifying multiple groups of associated customer voice word sets containing the subsequence of intersection tree nodes, and constructing multiple groups of branched tree node sequences on the last intersection tree node. If so, constructing multiple groups of branched tree node sequences on the last intersection tree node according to multiple groups of iterative customer voice word sets to obtain a customer voice word tree, collecting the customer voice word trees corresponding to each voice customer service intention to obtain a customer voice word tree forest, and performing deep learning recognition of the voice customer service intention according to the set of overlapping tree nodes to obtain the current customer service intention. The present invention can improve the recognition accuracy and recognition speed of the current intelligent customer service in the aspect of voice customer service intention recognition.