8 results about "Structure algorithm" patented technology

The embodiment of the invention provides a supply chain optimization result visualization and feedback method and device based on artificial intelligence interaction, and relates to the technical field of supply chain optimization result visualization technologies. The method comprises the following steps: receiving language information for at least one supply chain optimization result; generating a structured query request according to the language information; based on the structured query request, obtaining optimization scheme data corresponding to the decision object entity, and generating and displaying at least one visual chart according to a visual demand; receiving multi-modal feedback information input by a user on the visual chart; performing fusion analysis on the multi-modal feedback information to generate a structured algorithm optimization instruction; and performing recalculation based on the algorithm optimization instruction to generate an updated supply chain optimization result. According to the method and the device, the problem of poor interpretation of the algorithm result in the existing supply chain optimization system is solved, and the effect of improving the interpretation of the algorithm result in the supply chain optimization system is further achieved.

The application belongs to the field of post-quantum cryptography and information security technology, and relates to a lattice polynomial operation acceleration method and system based on fusion accumulation and recursive divide-and-conquer and a storage medium. The method comprises the following steps: obtaining a to-be-operated polynomial, determining the polynomial operation type, selecting a corresponding operation method according to the determination result, generating a corresponding type of synthetic long polynomial, and continuously writing the synthetic long polynomial into a shared accumulation buffer. The synthetic long polynomial is read from the shared accumulation buffer, polynomial folding and modulus reduction are performed, a polynomial operation result component is generated, and the polynomial operation result component is output to a next-level storage unit. The application solves the technical problems of high memory read-write bandwidth consumption and redundant algorithm power consumption caused by modulus reduction and folding immediately after multiplication of each sub-polynomial in the existing traditional matrix-vector multiplication, and the strict limitation of the existing NTT acceleration scheme to a specific modulus structure and poor algorithm agility.

The invention belongs to the field of path planning, discloses a path planning method based on an adaptive dynamic weighting A star algorithm, and aims to solve the problems of low search efficiency, multiple redundant nodes, high memory consumption and the like of a traditional A star algorithm. According to the method, the path planning effect is optimized through three core improvement strategies: 1, a dynamic weighted scoring strategy: weighted scoring of direction consistency, target guidance and environment constraint is performed on eight candidate directions of a current node, the search direction and number are dynamically adjusted, and the optimal direction is preferentially expanded; secondly, a self-adaptive heuristic method based on environment scoring is adopted, direction weighted scoring is fused into heuristic cost calculation, weight factors are dynamically adjusted in combination with the distance between a current node and a target node, and efficient search is guided; and 3, a path redundant node elimination strategy is adopted, and a path structure is simplified through node duplication removal and continuous node direct connection judgment. The algorithm execution process comprises four steps of initializing an open list and a closed list, searching dynamic nodes, judging target reachability and outputting path backtracking. Experimental verification shows that compared with a traditional A star algorithm and a part of improvement schemes, the method has the advantages that the path length, the calculation time, the number of expansion nodes and the number of path points are all remarkably optimized, the path is shorter and smoother, the search efficiency and robustness are higher, and the method is suitable for large-scale popularization and application. The method is suitable for complex path planning scenes in multiple fields of logistics distribution, robot autonomous movement, unmanned aerial vehicle path planning and the like.

The invention discloses a method for dynamically configuring cooperative task processing. The method comprises the following steps: S1, dynamically configuring and managing a multi-level organization unit, a region unit and a cooperative post based on a tree structure algorithm; s2, configuring a customizable task initiation form template and an electronic approval process for the collaborative post; s3, receiving a collaboration request which contains task clue information and is initiated by the base layer unit based on the form template; s4, analyzing the cooperation request, and completing multi-dimensional feature extraction and dynamic weight calculation to generate a task model; s5, performing capability evaluation and real-time state monitoring on the cooperative post personnel to generate post portraits; s6, performing multi-stage filtering and comprehensive scoring calculation on the task model and the post portrait through an intelligent matching engine, and generating a candidate person and member sorting list; s7, selecting a target personnel allocation task from the candidate personnel sorting list according to a load balancing strategy; and S8, driving the task to enter a closed-loop process of receiving, processing, feedback and result confirmation. The task cooperative processing efficiency is high.

The invention relates to a BPMN flow chart optimization method based on a tree structure algorithm, and the method comprises the steps: obtaining a BPMN model through an Activiti flow engine, and reading all flow elements in the BPMN model to construct a directed graph; a starting node in the BPMN model serves as a root node, the directed graph is traversed, and a tree structure is constructed; the tree-shaped structure comprises a trunk and sub-trees; hierarchy division is carried out according to the dependency relationship among the nodes on the tree trunk; arranging the corresponding sub-trees according to the positions and the widths of the sub-trees; and on the basis of an optimization strategy, reducing connection edge intersection between the nodes on each sub-tree, and generating an optimized BPMN flow chart. According to the method, the nodes in the flow chart are rearranged, so that node layout optimization of the flow chart is realized, the optimized flow chart can intuitively display the flow execution logic, and the requirements of complex business scenes can be met.

In the disclosed systems and methods for categorizing medical data, a computer system obtains, in electronic form, a plurality of medical records. Each medical record includes corresponding medical data from a respective medical evaluation and corresponding metadata comprising a plurality of attributes about the respective medical evaluation. Each respective attribute comprises a corresponding string of text. The computer system determines, for each respective pair of medical records consisting of a first medical record and a second medical record, a corresponding pairwise similarity between, for each respective attribute in a set of attributes, the corresponding string of text for the first medical record and the corresponding string of text for the second medical record. The computer system identifies a first subset of the plurality of medical records. Each respective medical record in the first subset is connected to each other through pairwise similarities that each satisfies a similarity threshold.

The embodiment of the application provides a kind of structure algorithm simulation ability evaluation method, the method comprises the following steps: first, establish standard model library;Step 100: verify the geometric model processing capability of structure algorithm: different versions and different formats of geometric model are imported into structure algorithm one by one, the geometric model processing capability of structure algorithm is recorded and judged, step 200: verify the meshing capability of structure algorithm: after the geometric model is imported into structure algorithm, the geometric model is meshed, then the meshing process is recorded and judged, step 300: verify the solving capability of structure algorithm: the structure of structure algorithm is analyzed;Step 400: the final result is obtained by weighted evaluation of data sets S1, S2, S3, the application can be subdivided test category, make the evaluation of structure algorithm simulation ability more comprehensive and accurate, and then obtain the evaluation result with higher reliability.

The invention discloses a source measurement unit control method for improving a spider bee algorithm to guide reinforcement learning and optimize PID. The method comprises the following steps: setting initial parameters; calculating a system transfer function according to an actual SMU model, and building a simulation control model of the SMU system based on an MATLAB platform; an SWO algorithm is improved through multiple strategies, and an MISWO module is formed; the DDPG deep strengthening model is improved, and an IDDPG model is formed; the MISWO module and the IDDPG model are fused, an evolutionary reinforcement learning framework (ERL) is constructed, and online guidance of the MISWO algorithm on the IDDPG is formed; and the ERL is applied to the SMU system to realize PID control of the SMU. The SWO and the DDPG are improved in a targeted mode, the PID control performance is improved, the ERL structure algorithm can use a group search mechanism to make up for the limitation of a strategy gradient algorithm on global search, and therefore the control performance and robustness of an SMU system are improved.