99 results about "Result list" patented technology

Disclosed in the present invention is a knowledge graph-based unified fault search method for a specialized vehicle device. The method comprises: acquiring device maintenance related data, and taking a device as a core to establish association relationships between the device maintenance related data; acquiring a fault report, and parsing the fault report to obtain parsing result data; constructing a device fault graph on the basis of the device maintenance related data and the parsing result data; integrating the device fault graph into a knowledge graph platform; receiving a search term entered by a current user in a search box, acquiring search results corresponding to the search term, and displaying the search results on a search result page, wherein the search results comprise a result list, data classification statistics, search result sorting, a knowledge cloud picture, and associated recommendations; and providing in the search result page an operation entry for knowledge exploration, and viewing related knowledge exploration of the current knowledge by the user by triggering the operation entry for knowledge exploration. The method solves the problem in the prior art that the search results are not accurate enough.

The invention discloses an RAG question and answer optimization method and system based on fusion retrieval, and relates to the technical field of large model knowledge question and answer. The method comprises the steps of obtaining user query, executing keyword retrieval and semantic vector retrieval in parallel, and obtaining a keyword retrieval result list and a semantic vector retrieval result list respectively; and calculating the quality score of each text unit in each retrieval result list based on multi-dimensional quality evaluation, and filtering out the text units with the quality scores lower than a preset quality threshold value. And sorting the text units in all the filtered retrieval result lists by adopting a quality weighted reciprocal sorting fusion algorithm. And based on the length of each text unit and the document format to which the text unit belongs, performing context content enhancement and combination on each text unit after sorting, and generating an enhanced context block list. And constructing structured prompt information based on the enhanced context block list and user query, inputting structured prompt words into the large language model, and generating and outputting a final answer.

The invention discloses an OCR (Optical Character Recognition) optimization method based on multi-label classification and active learning, and relates to the technical field of image recognizing.The method comprises the steps that an original image-text image is obtained and preprocessed, problem type classification is performed on problems existing in the preprocessed image based on a multi-label classification model, corresponding image problem types are obtained, and meanwhile, priorities are generated; performing targeted enhancement on the image according to the priority of the image problem classification result, repairing various problems existing in the image, and identifying the repaired image to obtain a structured character identification result list; and verifying an identification result, screening out difficult sample data, entering an active learning link according to the number of the sample data, and optimizing a multi-label classification model at the same time, thereby realizing dynamic cooperation of image quality improvement and OCR performance optimization. The problem that the accuracy of OCR recognition is reduced due to the multi-source quality degradation problem of the image in the prior art is effectively solved.

The invention relates to the technical field of data integrity verification, in particular to a drug compliance detection method and system for drug supervision, and the method comprises the following steps: extracting a process change data field, eliminating a deviation value and setting a reference, marking and classifying a current batch difference field, recognizing a deviation expansion item, constructing a field mapping path and recognizing an interruption point. And judging whether the interrupt field is missing or not, and generating various result lists. According to the method, through medicine process parameter extraction and abnormal value elimination, a stable reference standard is constructed, representativeness of a comparison standard is guaranteed, classification is performed based on field differences, parameter offset is identified, difference identification precision is improved, continuous offset fields are judged in combination with trend changes, risk early warning timeliness is enhanced, and data inheritance continuity is expressed; according to the method, breakpoint fields are filled and verified, missing information is accurately captured, a closed-loop process of offset identification, trend research and judgment, path tracing and missing verification is integrally formed, and the integrity and sensitivity of data compliance inspection are improved.

The invention discloses a resource recommendation method and device based on a multi-cloud environment, equipment and a storage medium, relates to the technical field of resource management, and discloses a resource recommendation method based on the multi-cloud environment, and the method comprises the steps: obtaining resource information of a resource demand and a resource instance, and carrying out the feature extraction according to the resource information, and obtaining a resource feature vector; generating a target demand vector according to the resource demand and the resource feature vector; performing vector similarity retrieval according to the target demand vector to obtain a target candidate resource set; performing score calculation according to candidate resources in the target candidate resource set to obtain candidate resource scores; and sorting the candidate resources according to the candidate resource scores to generate a recommendation result list. The technical effects of accurately matching the user demand and the resource instance in the multi-cloud environment, efficiently recommending the resource meeting the user demand, and improving the resource utilization efficiency and the user experience are achieved.

The invention relates to the technical field of artificial intelligence, and discloses an information auditing method and device, equipment and a medium, and the method comprises the steps: obtaining inquiry sheet information of a user, and carrying out the entity recognition of the inquiry sheet information through a natural language model, and obtaining an information field set; verifying the information field combination through a preset attribution rule and a preset logic verification condition to obtain a field compliance verification result; wherein the preset attribution rule is a preset rule based on business difference; the preset logic verification condition is a condition preset based on a field consistency relationship; if the field compliance verification result is that verification fails, performing inference analysis on the information field combination through a large model, and generating corrected inquiry sheet information; and analyzing the corrected inquiry sheet information through a risk assessment model to generate an audit result list. The information auditing method and device can be applied to a financial science and technology business program system, and the information auditing efficiency and accuracy can be improved.

The invention discloses a distributed operation-oriented multi-agent collaborative recommendation method and system, and relates to the technical field of intelligent recommendation, and the method comprises the steps: configuring a private domain operation agent in each independent private domain platform, deploying a big language model-based construction demand analysis agent at a user side to receive a natural language demand description input by a user, and constructing a multi-agent collaborative recommendation system; generating recommendation task instructions of different private domain platforms; the method comprises the following steps: receiving user preference description and recommendation object description, sending the description to a corresponding private domain operation agent, calling a recommendation model to generate a recommendation result list based on the received user preference description and recommendation object description, collecting recommendation result lists returned by all private domain operation agents by a demand analysis agent, and integrating and classifying the recommendation result lists to generate a comprehensive recommendation list. According to the method, an efficient, safe and extensible distributed recommendation architecture is constructed by introducing a user demand analysis agent, a private domain operation agent and a federal learning mechanism.

As shown in Reference Image Figure 2, search conditions for search items (12) are set in the search condition setting window (10), search conditions for additional search items (15), search for processing performed on wafers or the like in the past, and a search result summary (3) and search result list chart (4) are displayed in area (2) (for example, a time series graphic display or a time axis trend plot display). The desired processing is selected from the search result list chart (4), and the data for the selected processing is displayed in area (5). Area (5) shows, for example, a wafer map (9) that displays data for each part on the wafer in different colors.

The invention discloses an open source platform task matching method based on a developer behavior map and double-layer filtering, and the method comprises the following steps: S1, dividing the behavior characteristics of a developer into three dimensions, and obtaining a developer behavior characteristic data set; s2, identifying and screening developers which do not meet the basic requirements; s3, constructing a lightweight knowledge graph Gdevtask comprising developer nodes, task nodes and associated edges of the developer nodes, the task nodes and the associated edges of the developer nodes and the task nodes in a manner of combining a relational database with graph storage; s4, task demand texts Textreq of the historical open source tasks and the open source tasks to be matched are obtained, demand disassembly is carried out on the open source tasks to be matched to obtain a plurality of sub-tasks, and corresponding vectorization expressions are generated; s5, adopting a double-layer filtering strategy to construct a primary matching layer, and screening to obtain a primary candidate developer set; and constructing a fine screening scoring layer to obtain a recommendation result list.

The invention provides an old city bungalow area resident translation and moving decision-making method based on multiple agents. The method comprises the steps of collecting plot information of an old city bungalow area and constructing a data set, establishing a resident agent and a developer agent and setting behavior rules and action spaces of the resident agent and the developer agent, setting a learning method of the developer agent as multi-target hierarchical reinforcement learning, setting multi-target hierarchical rules respectively, and performing multi-target hierarchical reinforcement learning. And setting a reward mechanism of reinforcement learning, running all agents and performing data simulation, dynamically adjusting a relocation strategy in the simulation process until a convergence condition is met, outputting data and forming a result list. Compared with the application of the prior art, the multi-agent-based translation and moving decision-making method for residents in the one-storey area of the old city innovatively introduces the agents into a new scene of stock development, helps developers to make a more scientific moving decision, and can obtain a win-win balance result of the developers and the residents.

The invention discloses a commodity recommendation method and device, equipment and a medium, and relates to the technical field of data processing, and the method comprises the steps: receiving to-be-processed image data in response to an image input instruction; preprocessing the to-be-processed image data to obtain a processed image; inputting the processed image into a target detection model for target detection to obtain a commodity image, and performing visual feature extraction on the commodity image through a feature extraction model to obtain an image feature vector; based on the image feature vector, performing retrieval in a preset commodity feature library, and determining a main search result list; according to a preset commodity recommendation mode, determining a target commodity based on the main search result list and displaying the target commodity; the commodity recommendation mode comprises a hot commodity recommendation mode and a similar commodity recommendation mode. According to the method, the target commodity can be screened and displayed in a targeted manner, the retrieval efficiency of the commodity is greatly improved, and the commodity recommendation accuracy is further improved.

A system and method for performing parallel local and remote searches, in particular to searching a local cache and remote database in parallel while accepting a search input, to obtain additional data for the local cache. The method includes, responsive to detecting a first portion of a search input, searching a local cache using the first portion to populate a search results list and initiating a remote search using the first portion; and responsive to receiving remote search results from the remote search, updating the search results list based on the remote search results and updating the local cache with data associated with items in the remote search results.

The invention discloses a method for automatically generating a maritime combat killing net based on reinforcement learning, which relates to the field of killing nets and comprises the following steps: acquiring demand data of a plurality of combat tasks and resource data of a plurality of combat platforms; generating a legal action set based on the current system state and the hard constraint condition; scoring the actions in the legal action set by using a pre-trained reinforcement learning model, and selecting the action with the highest score as a killing network resource allocation decision; executing a killing network resource allocation decision, and updating a battlefield state; and in the evaluation stage of the reinforcement learning model, if it is detected that the model weight is unavailable, automatically switching to a greedy rollback strategy based on instant income to make a decision, and outputting a distribution result list consistent with the model decision format. According to the method, the dynamic legal action set is constructed, the decomposition type reward is designed, and the greedy rollback mechanism is introduced, so that stable training and reasoning can be realized in a large-scale and strong-constraint scene, and the efficiency and reliability of planning of the killing net are remarkably improved.

The invention relates to the field of engineering structure calculation and generation, and discloses an SAP2000-based calculation book screenshot and generation method and system, and the method comprises the steps: selecting a needed model file at a client, opening the model file through SAP2000 software, and obtaining and displaying the related data information of the model file; a screenshot parameter is configured at the client, and a screenshot task is initiated to the Python local service based on the screenshot parameter; the Python local service adopts a task execution engine based on a Python decorator to realize automatic screenshot, and a screenshot result list is generated; and the Python local service initiates a screenshot backfilling request to the Java remote service, and the Java remote service realizes dynamic backfilling of the Excel template based on the Apache POI and the EasyExcel, so that the required calculation book is automatically generated. The method has the advantages that the generation efficiency of the calculation book can be remarkably improved, and meanwhile, the accuracy of the generated calculation book is ensured.

The invention provides a multi-thread risk detection method and device and electronic equipment. The method comprises the steps that key symbols indicating shared resources are extracted from risk code blocks of a to-be-detected source code file; aiming at each key symbol, searching an associated code block referencing the key symbol in the global source code, screening a write-in code block which performs write-in operation on the shared resource through the key symbol in the associated code block, and identifying whether the write-in code block uses a lock mechanism or not to provide protection for the write-in operation; determining potential risk points for target code blocks which are identified in the write code blocks and do not use the lock mechanism to protect the write operation; generating an associated file result list based on the risk information of the potential risk point of each key symbol, the file path of the to-be-detected source code file and the risk code block; and analyzing the associated file result list based on the risk large model to obtain an analysis report. According to the method, symbol association and intelligent large model analysis are combined, and efficient and accurate multi-thread risk detection is achieved.

An information processing apparatus, an information processing method, and an information processing system. The information processing apparatus is provided with one or more memories that store state information indicating a predetermined state of an object, identification information for an image or video that represents a mode of the predetermined state, and document information including a description describing the mode represented by the image or video, and circuitry that receives from the communication terminal, a search input, transmits to the communication terminal, first screen information of a search result list screen indicating an image or video identified by the identification information of the image or video stored with the state information specified based on the input, receives from the communication terminal, information for specifying an image or video selected by a user, generates description information including a description in the document information stored in association with identification information for the image or video, and transmits the description information to the communication terminal.

The invention provides an intelligent question answering method and device, equipment, a storage medium and a program product, and relates to the field of financial science and technology. The method comprises the following steps: performing retrieval in a preset bank field document library to obtain a retrieval result list corresponding to each query in a question set generated based on user input; wherein the retrieval result list comprises a plurality of documents sorted according to similarity; according to document sorting and similarity information of each retrieval result list, dynamically optimizing fusion parameters of a fusion sorting algorithm; fusing and reordering documents in each retrieval result list based on a fusion ordering algorithm by adopting the dynamically optimized fusion parameters to obtain a reordered target retrieval result list; and generating a target answer corresponding to the user input based on the target retrieval result list. According to the method, the intelligent question and answer effect can be improved.

The application discloses a Chinese human phenotype ontology recognition method and related equipment, which can be applied to the technical field of text information processing. The application can effectively improve the recall rate by expanding the human phenotype ontology concept synonym set through a large language model, generating an expanded dictionary with the initial dictionary, significantly enhancing the discrimination ability of the recognition model for semantically ambiguous concepts through two-stage comparative learning, balancing the precision and recall rate by constructing an initial dictionary based on the human phenotype ontology official database and the Chinese human phenotype ontology official data, merging and disambiguating the current input text after accurate matching and semantic similarity matching to obtain a fused candidate result list, and finally obtaining the structured recognition result corresponding to each current input text by post-processing each candidate fragment in the fused candidate result list, which can provide effective data support for disease diagnosis.

As shown in the reference image Fig. 2, search conditions for search items (12) are set in the search condition setting window (10), and past processing performed on wafers, etc. is searched for, and a search result summary (3) and search result list chart (4) are displayed in area (2) (for example, a time series graphic display or a time axis trend plot display). The desired processing is selected from the search result list chart (4), and the data for the selected processing is displayed in area (5). Area (5) shows, for example, a wafer map (9) that displays data for each part on the wafer in different colors.

The invention relates to a data processing method and a related device. The data processing method comprises the following steps: converting data into TreeMap; traversing the key value pair on the outermost layer of the TreeMap by using a breadth-first search algorithm, adding the current key value pair into the initialized two-way queue, circularly executing, taking out an element located at the queue head in the two-way queue, and if the value of the element belongs to a basic type, adding a character string obtained by splicing the key value pair into a temporary list; if the value of the element belongs to the object or the array, the element is processed and then added into a bidirectional queue in a queue head insertion mode; when the two-way queue is empty, the circulation is ended, the character strings in the temporary list are spliced and then added into a result list, and the temporary list is initialized; and after the key value pair on the outermost layer of the TreeMap is processed, splicing the character strings in the result list into a to-be-signed character string and returning the to-be-signed character string. The keys in the JSON character string are sequenced according to a dictionary ascending order, and the keys supporting the nested hierarchy are also sequenced according to the dictionary ascending order.

Disclosed are a method and device for communication using a fronthaul interface. An operation method of an O-RU includes the steps of: generating a plurality of first measurement results by performing measurement operations on a first measurement target in a single notification period according to a notification interval; generating a first measurement result list including the plurality of first measurement results; and transmitting, to an O-DU, a message including the first measurement result list.

The application discloses an automatic high-division meter selecting method and system based on a heterogeneous platform, relates to the technical field of electric energy meter detection, and has the following advantages: in the task configuration stage, a user can conveniently check and select verification items and set related parameters through a graphical interface; in the instruction conversion and issuing stage, the configured task can be parsed into standardized control instructions and converted into special communication messages recognizable by the heterogeneous platform; in the verification execution and data collection stage, key data can be collected in real time and verification states can be dynamically displayed; in the automatic fine adjustment and rechecking stage, fine adjustment operations are automatically performed according to error data and rechecking is circularly performed; in the intelligent scoring and ranking stage, a comprehensive score is calculated and ranked by adopting a differentiated scoring rule and combining a weight coefficient; and in the result output stage, a detailed verification report can be generated and a meter selecting result list can be output, so that the adaptation of the heterogeneous platform, data processing and process control are realized.

As shown in Reference Image Figure 2, search conditions for search items (7) are set in the search condition setting window (1), search for past processes performed on wafers, etc., select desired processes from the search result list (2), and display data for the selected processes on a graph (3) or map (4). Graph (3), for example, displays the temperature change of the wafer over processing time, and may also display data for a specific time. Map (4), for example, displays a color-coded temperature map on the wafer together with a color sample (5).

As shown in the reference image Fig. 2, search conditions for search items (7) are set in the search condition setting window (1), search for processing performed on wafers etc. in the past, select the desired processing from the search result list (2), and display the data of the selected processing in graphs (3) and (4). Graphs (3) and (4) may, for example, display the change in wafer temperature over processing time, or display data for a specific time.

The invention discloses a person and post matching method, device and system and a storage medium. The person and post matching method comprises a first stage and a second stage, wherein the first stage is pre-matching, quantitative evaluation pre-matching is carried out based on structured information extracted from resumes and position description, and candidates basically meeting position conditions are screened out; the second stage is a model training and application stage, firstly, a mixed deep learning architecture of a double-tower model and a cross encoder is adopted to construct a resume coding tower, a position coding tower and a cross encoder model, so that deep semantic matching analysis and vector similarity matching are performed on candidate resumes passing pre-matching, and then the candidate resumes are obtained. The candidate resumes with the highest similarity of the TOPK are obtained, then the candidate resumes of the TOPK are finely arranged, and finally a matched result list is output.

The application discloses a content recommendation scene-based experiment strategy evaluation method, device and medium, comprising: creating opposite first and second shunt paths for first and second virtual recommendation requests; generating a first recommendation combination strategy based on N recommendation strategies associated with N path nodes in the first shunt path, and generating a second recommendation combination strategy based on N strategies associated with N path nodes in the second shunt path; obtaining a first recommendation result list by processing the first virtual recommendation request based on the first recommendation combination strategy, and obtaining a second recommendation result list by processing the second virtual recommendation request based on the second recommendation combination strategy; obtaining a target recommendation result list by combining the first and second recommendation result lists; and evaluating the effect difference of different recommendation strategies based on user feedback data of the target recommendation result list. The application can improve the AB test accuracy in the content recommendation scene.

1. The name of the design product: the graphical user interface of the function query setting condition in the front-end software of the electronic device. 2. The use of the design product: an electronic device. 3. The design points of the design product: the graphical user interface in the electronic device. 4. The picture or photo that best indicates the design points: the front view. 5. The electronic device is a conventional design, and other views are omitted. 6. The use of the graphical user interface: the product interface is the interface for displaying and interacting with the function query setting condition in the front-end software; the front view interface is the interactive interface when entering Distribution All, part of the interface is the query condition, and the lower part is the query result list; in the front view interface, after clicking 【New】, the interface change state diagram 1 is displayed; in the interface change state diagram 1 interface, after entering the information and clicking 【Add】, the interface change state diagram 2 is displayed.

1. The name of the design product: computer with search application graphical user interface. 2. The use of the design product: for running programs and display. 3. The design points of the design product: the graphical user interface content in the screen. 4. The picture or photo that best indicates the design points: interface change state diagram 2. 5. The computer is a common design, and other views are omitted. 6. The use of the graphical user interface: relationship analysis of specific targets and associated targets in the system. 7. The human-computer interaction mode of the graphical user interface: the main view is the system main interface; enter the keyword in the search box of the main view and click "search" to appear interface change state diagram 1; click any account name in the result list of interface change state diagram 1 to appear interface change state diagram 2; click any account avatar in the relationship graph of interface change state diagram 2 to appear interface change state diagram 3; click the "view evidence" icon in the relationship list of interface change state diagram 3 to appear interface change state diagram 4; move the mouse to any account avatar in the relationship graph of interface change state diagram 4 and click the right mouse button to appear interface change state diagram 5; select any type in the "expand type" drop-down menu of interface change state diagram 5 to appear interface change state diagram 6. It should be noted that the gray areas corresponding to the main view and each interface change state diagram are deleted content screens, and this area is not related to the design points of the product.

The application discloses a search result reordering method and device, equipment and a computer storage medium, and relates to the technical field of media content search. In the method, a suitable timing is intelligently selected to trigger a reordering process according to real-time interaction between a target object and media content during a search session, the un-consumed media content in a search result list of the search session is reordered, and the reordered search result list is presented to the target object. The method adjusts the order of each media content in the search result according to instant interest, can improve the accuracy of the result presentation of the search session, improves the content searching efficiency, and solves the problem that the interaction behavior feedback of the target object is not timely in the media content search scene. The order of the search result is adjusted in time to meet the search intention of the user.

The application discloses an open source platform task matching method based on a developer behavior graph and double-layer filtering, and the method comprises the following steps: S1, dividing the behavior characteristics of the developer into three dimensions to obtain a developer behavior characteristic data set; S2, identifying and screening out developers who do not meet the basic requirements; S3, constructing a lightweight knowledge graph G containing the nodes of the developers, the nodes of the tasks and the associated edges of the nodes by combining a relational database with a graph storage dev_task ; S4, obtaining the task requirement text Text req of the historical open source task and the open source task to be matched, obtaining a plurality of subtasks by decomposing the requirement of the open source task to be matched, and generating corresponding vectorized representations; S5, adopting a double-layer filtering strategy, constructing a primary matching layer to obtain a preliminary candidate developer set, and constructing a fine screening scoring layer to obtain a recommended result list.