Component search method, component search device, electronic device, and storage medium

Abstract

The application provides a component retrieval method, a component retrieval device, an electronic device and a storage medium, and belongs to the field of financial technology. The method comprises the following steps: obtaining a component retrieval request, and obtaining an original component from a component library; performing deconstruction processing on the original component to obtain key component data of the original component; constructing a component multi-dimensional feature model based on the key component data, wherein the component multi-dimensional feature model comprises a plurality of candidate components and component features of each candidate component; constructing a multi-dimensional feature database based on the component multi-dimensional feature model and component asset data; performing weight scoring on the component features based on the multi-dimensional feature database and a preset algorithm to obtain a feature weight matrix; performing retrieval scoring on the candidate components based on the component retrieval request and the feature weight matrix to obtain component scoring data; and screening the candidate components based on the component scoring data to obtain a target component list. The application can improve the accuracy of component retrieval.

Description

Technical Field

[0001] This application relates to the field of financial technology, and in particular to a component retrieval method, component retrieval device, electronic device, and storage medium. Background Technology

[0002] With the development of computer technology and artificial intelligence, traditional offline businesses are gradually migrating online, and this trend has become irreversible. For example, online shopping, live streaming, and online transactions benefit institutions such as banks and online merchants.

[0003] To improve transaction efficiency and security, institutions such as banks and online merchants often use components from existing component libraries to design financial transaction software for trading and communicating with online users.

[0004] In the design of financial trading software, it is often necessary to extract the required components from the component library. Most related technologies use manual search to find and compare components in the component library. This method relies heavily on the search experience and familiarity of the relevant personnel with the component library. However, manual search often results in errors, leading to low accuracy in component retrieval. Therefore, how to improve the accuracy of component retrieval has become an urgent technical problem to be solved. Summary of the Invention

[0005] The main objective of this application is to provide a component retrieval method, component retrieval device, electronic device, and storage medium, aiming to improve the accuracy of component retrieval.

[0006] To achieve the above objectives, a first aspect of this application proposes a component retrieval method, the method comprising:

[0007] Obtain the component retrieval request and retrieve the original component from the preset component library;

[0008] The original component is deconstructed to obtain the key component data.

[0009] Based on the key component data, a multidimensional feature model of the components is constructed, wherein the multidimensional feature model of the components includes multiple candidate components and the component features of each candidate component;

[0010] Based on the multidimensional feature model of the components and the pre-acquired component asset data, a multidimensional feature database is constructed;

[0011] Based on the multidimensional feature database and the preset algorithm, the component features are weighted and scored to obtain a feature weight matrix;

[0012] The candidate components are retrieved and scored based on the component retrieval request and the feature weight matrix to obtain component score data, wherein the component score data is used to represent the degree of matching between the candidate components and the component retrieval request;

[0013] The candidate components are filtered based on the component scoring data to obtain a target component list. In some embodiments, the component features include first feature features and second feature features. The step of weighting and scoring the component features based on the multidimensional feature database and a preset algorithm to obtain a feature weight matrix includes:

[0014] The first element feature is scored for importance based on the preset algorithm to obtain the first importance matrix;

[0015] Based on the first importance matrix, feature scores are performed on the first element features to obtain the first weight matrix;

[0016] The second element feature is scored for importance based on the preset algorithm to obtain the second importance matrix;

[0017] The feature weight matrix is ​​obtained based on the first weight matrix and the second importance matrix.

[0018] In some embodiments, the step of scoring the first element features based on the first importance matrix to obtain a first weight matrix includes:

[0019] Based on the first importance matrix, feature scores are performed on the first element feature to obtain a preliminary score vector for the first element feature.

[0020] The preliminary scoring vector is standardized to obtain the target scoring vector;

[0021] The target score vector and the first importance matrix are validated to obtain validation data, which is used to indicate whether there are logical errors in the target score vector and the first importance matrix.

[0022] If the verification data indicates that there are no logical errors in the target score vector and the first importance matrix, then the first weight matrix is ​​obtained based on the target score vector.

[0023] In some embodiments, the step of retrieving and scoring the candidate components based on the component retrieval request and the feature weight matrix to obtain component score data includes:

[0024] Extract the component feature text of each candidate component from the multidimensional feature database, wherein the component feature text includes the text content information of the component features of the candidate component;

[0025] Content extraction is performed on the component feature text to obtain the component semantic vector corresponding to the component feature text;

[0026] Content extraction is performed on the component retrieval request to obtain the retrieval semantic vector corresponding to the component retrieval request;

[0027] A vector similarity matrix is ​​constructed based on the component semantic vector and the retrieval semantic vector, wherein the vector similarity matrix is ​​used to represent the degree of semantic similarity between the component features and the component retrieval request;

[0028] The component scoring data is obtained by scoring the components based on the feature weight matrix and the vector similarity matrix.

[0029] In some embodiments, the deconstruction of the original component to obtain key component data includes:

[0030] The original components are first deconstructed to obtain first-level feature data;

[0031] The primary element data is then subjected to a second deconstruction to obtain secondary element data;

[0032] The key component data is obtained based on the primary element data and the secondary element data.

[0033] In some embodiments, constructing a multidimensional feature database based on the component multidimensional feature model and pre-acquired component asset data includes:

[0034] The candidate components are numbered according to a preset numbering rule to obtain the component number of the candidate components, wherein any two component numbers are different;

[0035] Row data is obtained based on the component number, column data is obtained based on the component name and component features of the candidate components, and a feature data table is constructed based on the row data and the column data;

[0036] The feature data table is populated with the component asset data to obtain the multidimensional feature database.

[0037] In some embodiments, the step of filtering the candidate components based on the component scoring data to obtain a target component list includes:

[0038] Compare the component scoring data with a preset threshold;

[0039] Candidate components whose component score data is greater than or equal to the preset threshold are selected as intermediate components;

[0040] The intermediate components are sorted in descending order based on the component rating data to obtain the target component list.

[0041] To achieve the above objectives, a second aspect of this application provides a component retrieval apparatus, the apparatus comprising:

[0042] The acquisition module is used to acquire component retrieval requests and obtain the original components from a preset component library;

[0043] The deconstruction module is used to deconstruct the original component to obtain the key component data of the original component;

[0044] The model building module is used to build a multi-dimensional feature model of components based on the key component data, wherein the multi-dimensional feature model of components includes multiple candidate components and component features of each candidate component;

[0045] The database construction module is used to construct a multidimensional feature database based on the component multidimensional feature model and pre-acquired component asset data;

[0046] The weight scoring module is used to perform weight scoring on the component features based on the multidimensional feature database and a preset algorithm to obtain a feature weight matrix.

[0047] The retrieval and scoring module is used to score the candidate components based on the component retrieval request and the feature weight matrix to obtain component scoring data, wherein the component scoring data is used to represent the degree of matching between the candidate component and the component retrieval request;

[0048] The component filtering module is used to filter the candidate components based on the component rating data to obtain a list of target components.

[0049] To achieve the above objectives, a third aspect of the present application provides an electronic device, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method described in the first aspect.

[0050] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0051] The component retrieval method, device, electronic device, and storage medium proposed in this application acquire a component retrieval request and obtain original components from a pre-set component library; deconstruct the original components to obtain key component data; and construct a multi-dimensional feature model of the components based on the key component data. This multi-dimensional feature model includes multiple candidate components and the component features of each candidate component. This approach facilitates the deconstruction of each original component, hierarchically deconstructing it to obtain the constituent elements used to compose the original component, enabling analysis at the element level to determine whether each original component meets the retrieval requirements. Furthermore, based on the multi-dimensional feature model and pre-acquired component asset data, a multi-dimensional feature database is constructed, allowing direct utilization of the feature content information in the multi-dimensional feature database during subsequent retrieval processes, effectively saving component retrieval time and improving component retrieval efficiency. Furthermore, based on a multi-dimensional feature database and a pre-defined algorithm, component features are weighted and scored to obtain a feature weight matrix. Candidate components are then scored based on the component retrieval request and the feature weight matrix to obtain component score data. This component score data represents the degree of matching between candidate components and the component retrieval request. It quantifies the overall relevance of each candidate component to the retrieval request, clearly and directly reflecting which candidate components are more relevant, thus making the component scoring process more granular. Finally, candidate components are filtered based on the component score data to obtain a target component list. This effectively improves the accuracy of component retrieval, enabling rapid retrieval of components suitable for financial trading software design. This significantly improves the design efficiency and accuracy of financial trading software, resulting in better security and timeliness in financial transactions between the designed software and network objects. Attached Figure Description

[0052] Figure 1 This is a flowchart of the component retrieval method provided in the embodiments of this application;

[0053] Figure 2 yes Figure 1 The flowchart of step S102 in the document;

[0054] Figure 3 yes Figure 1 The flowchart of step S104 in the process;

[0055] Figure 4 yes Figure 1 The flowchart of step S105 in the process;

[0056] Figure 5 yes Figure 4 The flowchart of step S402 in the document;

[0057] Figure 6 yes Figure 1 The flowchart of step S106 in the process;

[0058] Figure 7 yes Figure 1 The flowchart of step S107 in the process;

[0059] Figure 8 This is a schematic diagram of the component retrieval device provided in the embodiments of this application;

[0060] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0063] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0064] First, let's analyze some of the terms used in this application:

[0065] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0066] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0067] Information Extraction (NER) is a text processing technique that extracts factual information such as entities, relationships, and events from natural language text and outputs it as structured data. Information extraction is a technique for extracting specific information from text data. Text data is composed of specific units, such as sentences, paragraphs, and chapters. Text information is composed of smaller, specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these units. Extracting noun phrases, names of people, and place names from text data is an example of text information extraction. Of course, text information extraction techniques can extract information of various types.

[0068] With the development of computer technology and artificial intelligence, traditional offline businesses are gradually migrating online, and this trend has become irreversible. For example, online shopping, live streaming, and online transactions benefit institutions such as banks and online merchants.

[0069] To improve transaction efficiency and security, institutions such as banks and online merchants often use components from existing component libraries to design financial transaction software for trading and communicating with online users.

[0070] In the design of financial trading software, it is often necessary to extract the required components from the component library. Most related technologies use manual search to find and compare components in the component library. This method relies heavily on the search experience and familiarity of the relevant personnel with the component library. However, manual search often results in errors, leading to low accuracy in component retrieval. Therefore, how to improve the accuracy of component retrieval has become an urgent technical problem to be solved.

[0071] Based on this, embodiments of this application provide a component retrieval method, a component retrieval device, an electronic device, and a storage medium, aiming to improve the accuracy of component retrieval.

[0072] The component retrieval method, apparatus, electronic device, and storage medium provided in the embodiments of this application are specifically described through the following embodiments. First, the component retrieval method in the embodiments of this application is described.

[0073] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0074] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0075] The component retrieval method provided in this application relates to the field of artificial intelligence technology. The component retrieval method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the component retrieval method, but is not limited to the above forms.

[0076] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0077] Figure 1 This is an optional flowchart of the component retrieval method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S107.

[0078] Step S101: Obtain a component retrieval request and retrieve the original component from the preset component library;

[0079] Step S102: Deconstruct the original component to obtain the key component data;

[0080] Step S103: Based on the key component data, construct a multi-dimensional feature model of the components, wherein the multi-dimensional feature model of the components includes multiple candidate components and the component features of each candidate component;

[0081] Step S104: Construct a multidimensional feature database based on the component multidimensional feature model and the pre-acquired component asset data;

[0082] Step S105: Based on the multidimensional feature database and the preset algorithm, the component features are weighted and scored to obtain the feature weight matrix;

[0083] Step S106: Based on the component retrieval request and the feature weight matrix, the candidate components are retrieved and scored to obtain component score data, wherein the component score data is used to represent the degree of matching between the candidate components and the component retrieval request;

[0084] Step S107: Filter candidate components based on component scoring data to obtain a list of target components.

[0085] Steps S101 to S107 of this embodiment involve obtaining a component retrieval request and acquiring original components from a pre-set component library; deconstructing the original components to obtain key component data; and constructing a multi-dimensional feature model of the components based on the key component data. This multi-dimensional feature model includes multiple candidate components and the component features of each candidate component. This method facilitates the deconstruction of each original component, hierarchically deconstructing it to obtain the constituent elements used to compose the original components, enabling analysis at the element level to determine whether each original component meets the retrieval requirements. Furthermore, based on the multi-dimensional feature model and pre-acquired component asset data, a multi-dimensional feature database is constructed, allowing direct utilization of the feature content information in the multi-dimensional feature database during subsequent retrieval processes, effectively saving component retrieval time and improving component retrieval efficiency. Furthermore, based on a multi-dimensional feature database and a pre-defined algorithm, component features are weighted and scored to obtain a feature weight matrix. Based on the component retrieval request and the feature weight matrix, candidate components are scored to obtain component score data. This component score data represents the degree of matching between candidate components and the component retrieval request. It quantifies the overall relevance of each candidate component to the retrieval request, clearly and directly reflecting which candidate components are more relevant, thus making the component scoring process more granular. Finally, candidate components are filtered based on the component score data to obtain a target component list, effectively improving the accuracy of component retrieval.

[0086] In step S101 of some embodiments, when obtaining a component retrieval request, it can be a component retrieval request obtained from the target object's input (typed or voice input). That is, the component retrieval request can be a piece of voice data, a piece of descriptive text data, etc., without limitation. The component retrieval request includes descriptive content information of the required component or descriptive information of the current development requirements. The target object can be development staff, etc.

[0087] When the server receives a component retrieval request for the target object, it retrieves the original component from a pre-defined component library. This pre-defined component library can be built based on existing components, that is, storing components generated in the historical development and design in the same database to obtain the component library. This component library includes original components of different types, different development levels, and different scenarios. For example, original components include text box components, graphic components, password input components for financial transactions, transaction prompt box components, etc.

[0088] Please see Figure 2 In some embodiments, step S102 may include, but is not limited to, steps S201 to S203:

[0089] Step S201: Perform a first deconstruction on the original components to obtain first-level feature data;

[0090] Step S202: Perform a second deconstruction on the primary element data to obtain secondary element data;

[0091] Step S203: Based on the primary element data and the secondary element data, obtain the key component data.

[0092] In step S201 of some embodiments, the process of deconstructing the original component can be understood as breaking down the original component into basic elements that can make up the original component. First, the original component is deconstructed in a first stage, which is a first-level deconstruction, deconstructing the original component into multiple first-level elements, and using the first-level elements obtained after the first-level deconstruction as first-level element data.

[0093] It should be noted that after the initial deconstruction of the original components, the resulting primary element data includes view elements, logical elements, model elements, interaction elements, and communication elements. Among them, view elements refer to the visual interface that users can see and interact with; logical elements refer to the logical data related to the control of business processes and the execution of business rules; model elements refer to the data model, including the definition of data objects and the manipulation of data, etc.; interaction elements refer to the way users operate and the system displays information; and communication elements refer to the process of transmitting information between the original components through a medium.

[0094] In step S202 of some embodiments, since some primary elements can be further decomposed into smaller individual elements, the process of second decomposing the primary element data can be understood as splitting the primary element data into specific elements that can constitute the primary elements. That is, performing secondary decomposition on the original components, decomposing each primary element into multiple secondary elements, and using the secondary elements obtained after secondary decomposition as secondary element data. Among them, view elements, model elements, interaction elements, and communication elements can all be decomposed into smaller individual elements.

[0095] It should be noted that the process of deconstructing the primary element data to obtain the secondary element data includes: deconstructing the view elements to obtain the corresponding secondary elements, which include structural elements, element elements, and style elements. Structural elements include page layout structure and hierarchical structure; element elements include radio buttons, forms, label bars, etc.; and style elements include font, font size, font color, and background color, etc. Deconstructing the interaction elements to obtain the corresponding secondary elements includes input elements, output elements, and perception elements; deconstructing the communication elements to obtain the corresponding secondary elements includes information elements, media elements, and transmission elements; and deconstructing the model elements to obtain the corresponding secondary elements includes entity elements and method elements.

[0096] In step S203 of some embodiments, the primary element data and secondary element data obtained through the first and second deconstruction are summarized as a data whole to obtain the key component data of the original component. The key component data of each original component can be used to fully express the original component.

[0097] Through the above steps S201 to S203, each original component can be deconstructed multiple times in a hierarchical manner to obtain the first-level and second-level elements used to compose the original component. This allows for analysis of whether each original component meets the retrieval requirements at the element level. Compared with component retrieval at the component level, this makes the component retrieval process more granular, improves the precision of component retrieval, and helps to improve the accuracy of component retrieval.

[0098] In step S103 of some embodiments, when constructing a multi-dimensional feature model of a component based on key component data, each primary element and secondary element is first structured according to its own attribute features, and the structured primary and secondary elements are used as component features of the original component. For example, the component feature corresponding to a view element is the view feature, and the component features corresponding to the secondary elements under the view element are the view structure feature, the view element feature, and the view style feature, respectively.

[0099] Furthermore, based on the hierarchical relationship between the original components and their component features, a multidimensional feature model of the components is constructed. The multidimensional feature model of the components includes multiple candidate components and the component features of each candidate component. The multidimensional feature model of the components can be represented in the form of multiple tree diagrams, with each candidate component as an independent tree structure.

[0100] Taking a candidate component as an example, the root node of the tree structure is the candidate component itself. The component features of the first-level leaf nodes of the root node are the first element features, which correspond to the first-level element data. That is, the first element features include view features, model features, logical features, interaction features, and communication features. Furthermore, the first-level leaf nodes also include the next-level leaf nodes, that is, the second-level leaf nodes of the root node. The component features of the second-level leaf nodes are the second element features, which correspond to the second-level element data. That is, the second-level leaf nodes under the view features have view structure features, view element features, and view style features; the second-level leaf nodes under the model features have model entity features and model method features; the second-level leaf nodes under the interaction features have interaction input features, interaction output features, and interaction perception features; and the second-level leaf nodes under the communication features have communication information features, communication medium features, and communication transmission features.

[0101] This approach can conveniently characterize the key constituent data of each original component to obtain the component features of the original component. Based on the hierarchical relationship between the original component and its component features, a multidimensional component feature model can be constructed. This multidimensional component feature model can aggregate component features from multiple different dimensions and can also clearly reflect the hierarchical relationship between each candidate component and its component features.

[0102] Please see Figure 3 In some embodiments, step S104 may include, but is not limited to, steps S301 to S303:

[0103] Step S301: Number the candidate components according to the preset numbering rules to obtain the component number of the candidate components, wherein any two component numbers are different.

[0104] Step S302: Obtain row data based on component number, obtain column data based on component name and component features of candidate components, and construct feature data table based on row data and column data;

[0105] Step S303: Populate the feature data table with content based on the component asset data to obtain a multidimensional feature database.

[0106] In step S301 of some embodiments, the preset numbering rule can be set according to actual conditions and is not limited. For example, the numbering rule can be set as "component number: CP + an incrementing four-digit number, where CP represents the abbreviation of the English name of the component, for example, component numbers are CP0001, CP0002, CP0201, etc." Therefore, each candidate component is numbered according to this numbering rule to obtain the component number of each candidate component. In order to distinguish different candidate components, any two component numbers are different, that is, the component number of each candidate component is unique.

[0107] In addition, when setting encoding rules, different encoding formats can be set for different categories of candidate components. For example, different types of candidate components can start with different letters, and candidate components of the same type can be named with an incrementing four-digit number. Alternatively, multiple strings can be randomly generated, with each string corresponding to a candidate component, and the string corresponding to the candidate component can be used as its component number.

[0108] In step S302 of some embodiments, after numbering each candidate component, the component number can be used as row data, and the component name and component features of each candidate component can be used as column data to form a feature data table. That is, each row of the feature data table represents all the feature information of a candidate component, and each column includes a certain type of component content information of all candidate components. The feature data table can include a component number column, a component name column, a view structure feature column, a view element feature column, a view style feature column, a logic feature column, a model entity feature column, a model method feature column, an interaction input feature column, an interaction output feature column, an interaction perception feature column, a communication medium feature column, a communication transfer feature column, and a communication information feature column.

[0109] In step S303 of some embodiments, the component asset data can be compiled and summarized by developers during component development. Therefore, the already written component asset data can be directly called from a preset data source. The component asset data includes the detailed feature data of the component features of each candidate component. The component asset data can be composed of a series of Excel tables, and each Excel table is used to summarize the detailed feature data of the same component feature of candidate components. For example, the detailed feature data of model features includes the number, English name, Chinese name, and description of the model features of each candidate component, etc. The detailed feature data of model method features includes the number, Chinese name, English name, and description of each model method feature, etc. For example, the detailed feature data of a certain model feature includes "Number: MD0001; Chinese name: Insured Information Model; English name: InputInsuredInfoModel; Description: Empty". The detailed feature data of a certain model method feature includes "Number: MM0001; Chinese name: Get Customer List; English name: getCustomerList; Description: Convert model customer information into JSON required by the interface". The detailed feature data of a certain view style feature includes "Number: ST0001; Chinese name: Title Style; English name: TitleStyle; Attribute: Size is 10, color number is gray, bold, Description: Empty". And so on. Traverse all the tables and table contents of the component asset data, and supplement the extracted table contents to the corresponding content areas of the feature data table to obtain a feature data table with complete content. Use the feature data table with complete content as a multi-dimensional feature database.

[0110] Through the above steps S301 to S303, the feature details of each component feature can be extracted from the component asset data relatively easily, and all component features are summarized into a data table to obtain a multidimensional feature database containing feature information of all component features. This allows the feature content information of the multidimensional feature database to be directly used in the subsequent retrieval process, which can effectively save component retrieval time and improve component retrieval efficiency.

[0111] Please see Figure 4 In some embodiments, the component features include a first feature and a second feature, and step S105 may include, but is not limited to, steps S401 to S404:

[0112] Step S401: Based on a preset algorithm, the importance of the first element feature is scored to obtain the first importance matrix;

[0113] Step S402: Based on the first importance matrix, feature scores are performed on the first element features to obtain the first weight matrix;

[0114] Step S403: Based on a preset algorithm, the importance of the second element features is scored to obtain the second importance matrix;

[0115] Step S404: Based on the first weight matrix and the second importance matrix, the feature weight matrix is ​​obtained.

[0116] In step S401 of some embodiments, the preset algorithm may be the analytic hierarchy process (AHP), etc. When using the AHP to score the importance of the first element feature, a Santy-based 1-9 scale method can be introduced to calculate the importance relationship between each pair of elements.

[0117] Specifically, a scale of 1 indicates that the two primary features are equally important; a scale of 3 indicates that the former is slightly more important than the latter; a scale of 5 indicates that the former is significantly more important than the latter; a scale of 7 indicates that the former is extremely more important than the latter; a scale of 9 indicates that the former is strongly more important than the latter; scales of 2, 4, 6, and 8 represent the median values ​​of the aforementioned ratios of importance of the primary features, and the reciprocals of 1 to 9 represent the positive and negative numbers of the aforementioned ratios of importance of the two primary features.

[0118] For example, the scale between component names is 1, the scale between component names and views is 5, meaning the component name is more important, and the scale between component names and logic is 1 / 3, meaning the logic is more important.

[0119] In a specific scenario, the first importance matrix of a candidate component is presented in tabular form as shown in Table 1:

[0120]

[0121]

[0122] Table 1

[0123] Based on this, an importance comparison matrix is ​​constructed for every two first element features to obtain the first importance matrix. In the first importance matrix, the element value is the scale of the two first element features. For example, the P-th row and Q-th column of the first importance matrix represents the scale value (i.e., the importance score result) between the Q-th first element feature in the horizontal direction and the P-th first element feature in the vertical direction.

[0124] In step S402 of some embodiments, when scoring the first feature based on the first importance matrix, a preliminary score vector for each horizontal first feature is first calculated according to the first importance matrix. Then, the preliminary score vector is standardized, and the standardized preliminary score vector is used as the target score vector for each horizontal first feature. Further, the target score vector and the first importance matrix are verified according to preset verification rules to obtain verification data. If the verification data indicates that there are no logical errors in the target score vector and the first importance matrix, the target score vector is used as the final weight for each first feature, thereby obtaining the first weight matrix for the first feature.

[0125] In step S403 of some embodiments, the importance of the second element features is scored based on a preset algorithm to obtain a second importance matrix. This second importance matrix is ​​mainly used to represent the importance ratio of each second element feature to the first element feature at its higher level. The specific importance scoring process is basically the same as the scoring process in step S401 above, and will not be repeated here. For example, the second importance matrix shows that the importance ratios of view structure features, view style features, and view element features to view features are [0.1667, 0.1667, 0.6667].

[0126] In step S404 of some embodiments, when obtaining the feature weight matrix based on the first weight matrix and the second importance matrix, the importance ratios in the second importance matrix are weighted using the weight values ​​of the first weight matrix to obtain the feature weight matrix. For example, the weight of the view feature obtained from the first weight matrix is ​​0.187, and the importance ratios of the view structure feature, view style feature, and view element feature to the view feature obtained from the second importance matrix are [0.1667, 0.1667, 0.6667]. Therefore, the weight value of the view structure feature is 0.187 * 0.1667 = 0.031. The weight value of the view structure feature can represent the degree of influence of the view structure feature on the search results in component retrieval. The feature weight matrix can be used to represent the importance weight of each component feature of the candidate component.

[0127] Through the above steps S401 to S404, the weight quantization of each component feature can be performed relatively easily, so that a candidate component can be identified more accurately and comprehensively based on multiple component features, which can effectively improve the accuracy of component retrieval.

[0128] Please see Figure 5 In some embodiments, step S402 may include, but is not limited to, steps S501 to S504:

[0129] Step S501: Based on the first importance matrix, the first element feature is scored to obtain the preliminary score vector of the first element feature;

[0130] Step S502: Standardize the preliminary scoring vector to obtain the target scoring vector;

[0131] Step S503: Perform verification processing on the target score vector and the first importance matrix to obtain verification data. The verification data is used to indicate whether there are logical errors in the target score vector and the first importance matrix.

[0132] Step S504: If the verification data indicates that there are no logical errors in the target score vector and the first importance matrix, then the first weight matrix is ​​obtained based on the target score vector.

[0133] In step S501 of some embodiments, when scoring the first feature based on the first importance matrix, the m-th power of the first feature in each row of the first importance matrix is ​​first calculated to obtain an m-dimensional vector, which is the preliminary score vector of the first feature. The calculation process can be represented as shown in formula (1):

[0134]

[0135] in, Let be the initial score vector of the first feature in the i-th row, m be the feature dimension (m is not greater than the total number of columns in the first importance matrix), j be the j-th column of the first importance matrix, and a ij The value of the element in the i-th row and j-th column of the first importance matrix (i.e., the scale value).

[0136] In a specific implementation scenario, the first importance matrix of a candidate component is shown in Table 1 above. The initial score vector calculation process for this candidate component is as follows:

[0137] Component name:

[0138] View features:

[0139] Model features:

[0140] Logical characteristics:

[0141] Interaction features:

[0142] Communication characteristics:

[0143] In step S502 of some embodiments, when the preliminary scoring vector is standardized to obtain the target scoring vector, the standardization process can be represented as shown in formula (2):

[0144]

[0145] Among them, w i Let i be the target score vector of the first feature element in the i-th row. Let m be the initial score vector for the first feature element in the i-th row, where m is the feature dimension and m is no greater than the total number of columns in the first importance matrix. Sum the initial score vectors for the first element features of all rows.

[0146] Furthermore, the target score vectors of all first element features belonging to the same candidate component are integrated to obtain the target score vector W of the candidate component.

[0147] In a specific implementation scenario, the first importance matrix of a candidate component is shown in Table 1 above. The calculation process of the target score vector of the candidate component is as follows:

[0148] Component name:

[0149] View features:

[0150] Model features:

[0151] Logical characteristics:

[0152] Interactive features:

[0153] Communication characteristics:

[0154] Therefore, the target score vector of this candidate component can be represented as W = [0.244, 0.187, 0.385, 0.23, 0.244, 0.08].

[0155] In step S503 of some embodiments, when verifying the target scoring vector and the first importance matrix to obtain verification data, a consistency verification of the first importance matrix can be performed. First, the maximum eigenvalue λ of the first importance matrix A of each candidate component is calculated. max The calculation process can be represented as shown in formula (3):

[0156] λmax=AW1 / W1+AW2 / W2+AW3 / W3+…+AWn / Wn formula (3)

[0157] Where n is the dimension of the first importance matrix, AW represents the matrix multiplication of the first importance matrix and the target score vector, and W is the target score vector of the candidate component.

[0158] Then, based on the largest eigenvalue λ max The consistency verification index CI is calculated, and the calculation process can be represented as shown in formula (4).

[0159]

[0160] Where n is the dimension of the first importance matrix.

[0161] The closer the consistency check metric (CI) is to 0, the better the consistency. When CI is less than a preset threshold, it indicates that the first importance matrix has passed the consistency check, the target score vector meets the requirements, and the check data (the first importance matrix and the target score vector) contains no logical errors. If CI is greater than or equal to the preset threshold, it indicates that the first importance matrix has failed the consistency check, the current target score vector does not meet the requirements, and the check data (the first importance matrix and the target score vector) contains logical errors, requiring the first importance matrix to be reconstructed.

[0162] In step S504 of some embodiments, when CI is less than a preset threshold, it indicates that the first importance matrix has passed the consistency check, the target score vector meets the requirements, and the check data shows that there are no logical errors in the first importance matrix and the target score vector. Therefore, the current target score vector can be directly used as the final weight of the first feature, thereby obtaining the first weight matrix of the first feature.

[0163] Steps S501 to S504 above can conveniently quantify the weights of each component feature, and a consistency check is introduced to determine whether the current first importance matrix and target score vector can be used to represent the feature weights of the component features, which can improve the accuracy of component feature weight allocation, thereby improving the accuracy of component retrieval.

[0164] Please see Figure 6 In some embodiments, step S106 includes, but is not limited to, steps S601 to S605:

[0165] Step S601: Extract the component feature text of each candidate component from the multidimensional feature database, wherein the component feature text includes the text content information of the component features of the candidate component.

[0166] Step S602: Extract the content from the component feature text to obtain the component semantic vector corresponding to the component feature text;

[0167] Step S603: Extract content from the component retrieval request to obtain the retrieval semantic vector corresponding to the component retrieval request.

[0168] Step S604: Construct a vector similarity matrix based on the component semantic vector and the retrieval semantic vector, wherein the vector similarity matrix is ​​used to represent the degree of semantic similarity between component features and component retrieval requests;

[0169] Step S605: Component scoring is performed based on the feature weight matrix and vector similarity matrix to obtain component scoring data.

[0170] In step S601 of some embodiments, since the multidimensional feature database includes feature details of all candidate components and their component features, the component feature text of each candidate component can be extracted from the multidimensional feature database by means of web crawling or data calling through a data interface. The component feature text includes the text content information of the component features. For example, the component feature text is extracted from the text box of the view style feature of a candidate component CP0005. The text content information of the component feature text is "Set the font size of the title style to 10 and the font to KaiTi".

[0171] In step S602 of some embodiments, when extracting content from component feature text, a preset word segmentation tool (such as Jieba word segmentation, etc.) can be used to segment the component feature text to obtain the segmentation results. Then, operations such as part-of-speech filtering and phrase generation are performed on the segmentation results to obtain a candidate vocabulary. Finally, word embedding processing is performed on the feature content of the candidate vocabulary using a word vector algorithm (such as Bert encoding algorithm, etc.) to obtain the component semantic vector corresponding to the component feature text.

[0172] In step S603 of some embodiments, when extracting content from a component retrieval request, a preset word segmentation tool (such as Jieba word segmentation, etc.) can be used to segment the component retrieval request to obtain the segmentation results. Then, operations such as part-of-speech filtering and phrase generation are performed on the segmentation results to obtain a retrieval vocabulary. Finally, word embedding processing is performed on the feature content of the retrieval vocabulary using a word vector algorithm (such as Bert encoding algorithm, etc.) to obtain the retrieval semantic vector corresponding to the component retrieval request.

[0173] In step S604 of some embodiments, when constructing a vector similarity matrix based on the component semantic vector and the retrieval semantic vector, the similarity S[i,j] between the component semantic vector and the retrieval semantic vector is first calculated using methods such as cosine similarity algorithm or Euclidean distance method, where [i,j] is the j-th component feature of the i-th candidate component. A vector similarity matrix Q[M,N] is then constructed based on the similarity S[i,j] between each component semantic vector and the retrieval semantic vector. This vector similarity matrix represents the semantic similarity between each component feature of each candidate component and the component retrieval request. For example, the vector similarity matrix Q[M,N] has M rows and N columns, where M is the number of candidate components and N is the number of component features of the candidate components. The elements in the vector similarity matrix Q[M,N] are component features, and the element value of each element is the similarity between that component feature and the component retrieval request; that is, the element value of the element in the i-th row and j-th column is S[i,j].

[0174] As can be seen from step S103 above, the component features can include features corresponding to the secondary elements, namely, component name, view structure features, view element features, view style features, model entity features, logical features, model method features, communication medium features, and communication information features. The number of component features is 9.

[0175] In step S605 of some embodiments, each component feature is first scored and weighted based on the feature weight matrix and the vector similarity matrix to obtain a weighted vector similarity matrix. Specifically, the feature weight of each component feature is extracted from the feature weight matrix, and the feature weight is multiplied by the corresponding element value of that component feature in the vector similarity matrix (i.e., the similarity s between the text content information of each component feature and the component request) to obtain the weighted score value of that component feature. Further, after obtaining the weighted score value of each component feature, all weighted score values ​​belonging to the same candidate component are summed to obtain the total component score of the candidate component. This total component score is used as the component score data for the candidate component.

[0176] Through the above steps S601 to S605, the semantic similarity between each component feature and the component retrieval request can be calculated relatively easily. Thus, the correlation between the component feature and the component retrieval request can be determined based on the semantic similarity. Based on the component features, the overall relevance of each candidate component to the component retrieval request can be obtained. Component scoring data is used to quantify the overall relevance of each candidate component to the component retrieval request. This can clearly and directly reflect which candidate components are more relevant to the component retrieval request, making the component scoring process more granular. By comprehensively evaluating the relevance between candidate components and the component retrieval request based on multiple feature dimensions, it is beneficial to improve the accuracy of component retrieval.

[0177] Please see Figure 7 In some embodiments, step S107 may include, but is not limited to, steps S701 to S703:

[0178] Step S701: Compare the component scoring data with the preset threshold;

[0179] Step S702: Select candidate components whose component score data is greater than or equal to a preset threshold as intermediate components;

[0180] Step S703: Sort the intermediate components in descending order based on the component rating data to obtain the target component list.

[0181] In step S701 of some embodiments, the specific value of the preset threshold can be set according to actual business needs and is not limited. For example, the preset threshold can be set to 0.6. The component score data is compared with the preset threshold, and based on the relationship between the component score data and the preset threshold, it is determined whether the candidate component meets the requirements of the current development design.

[0182] In step S702 of some embodiments, since the larger the component score data of the candidate component, the more suitable the candidate component is as a component of the current development design, when the component score data of the candidate component is greater than or equal to a preset threshold, it indicates that the candidate component meets the requirements of the current development design; when the component score data of the candidate component is less than the preset threshold, it indicates that the candidate component does not meet the requirements of the current development design. Therefore, the candidate component with component score data greater than or equal to the preset threshold is selected as the intermediate component.

[0183] In step S703 of some embodiments, in order to retrieve as many candidate components as possible that are helpful to the current development and design, the intermediate components can be sorted in descending order according to the size of the component score data to obtain a target component list. The intermediate components with higher component score data are placed at the beginning of the target component list, which makes it convenient for the target object to extract the middle components at the beginning for development and design, thus improving the convenience of development and design.

[0184] Through the above steps S701 to S703, the component scoring data can be used to quantify the degree of fit between candidate components and current development and design requirements. Based on the idea of ​​threshold comparison, at least one candidate component that can meet the current development and design requirements can be selected as an intermediate component. The selection order of intermediate components can be determined by using the size of the component scoring data. This allows intermediate components with a higher degree of matching to be selected for development and design, which is beneficial to improving the accuracy of component retrieval and development and design.

[0185] The component retrieval method of this application embodiment obtains a component retrieval request and retrieves original components from a preset component library; it then deconstructs the original components to obtain key component data; based on the key component data, it constructs a multi-dimensional feature model of the components, wherein the multi-dimensional feature model includes multiple candidate components and component features of each candidate component. This method can conveniently deconstruct each original component, hierarchically deconstructing it to obtain the constituent elements used to compose the original components, enabling analysis at the element level to determine whether each original component meets the retrieval requirements. Furthermore, based on the multi-dimensional feature model of the components and the pre-acquired component asset data, a multi-dimensional feature database is constructed, allowing the feature content information of the multi-dimensional feature database to be directly utilized in subsequent retrieval processes, effectively saving component retrieval time and improving component retrieval efficiency. Furthermore, based on a multi-dimensional feature database and a preset algorithm, component features are weighted and scored to obtain a feature weight matrix. Candidate components are then scored based on the component retrieval request and the feature weight matrix to obtain component score data. This component score data represents the degree of matching between candidate components and the component retrieval request. It quantifies the overall relevance of each candidate component to the retrieval request, clearly and directly reflecting which candidate components are more relevant, thus making the component scoring process more granular. Finally, candidate components are filtered based on the component score data to obtain a target component list, effectively improving the accuracy of component retrieval. In addition, the component retrieval method of this embodiment can quickly retrieve components that meet current needs from an existing front-end component library, effectively reducing repetitive code writing, improving component reusability, and lowering development costs. Furthermore, it can quickly retrieve components suitable for financial trading software design based on the target component list, effectively improving the design efficiency and accuracy of financial trading software, and enabling better security and timeliness in financial transactions between the designed financial trading software and network objects.

[0186] Please see Figure 8 This application also provides a component retrieval device that can implement the above-described component retrieval method. The device includes:

[0187] The acquisition module 801 is used to acquire component retrieval requests and obtain original components from a preset component library;

[0188] The deconstruction module 802 is used to deconstruct the original component to obtain the key component data;

[0189] The model building module 803 is used to build a multi-dimensional feature model of components based on key component data. The multi-dimensional feature model of components includes multiple candidate components and the component features of each candidate component.

[0190] Database construction module 804 is used to build a multidimensional feature database based on the component multidimensional feature model and pre-acquired component asset data;

[0191] The weight scoring module 805 is used to score the component features based on a multi-dimensional feature database and a preset algorithm to obtain a feature weight matrix.

[0192] The retrieval scoring module 806 is used to score candidate components based on component retrieval requests and feature weight matrices to obtain component scoring data, wherein the component scoring data is used to represent the degree of matching between candidate components and component retrieval requests;

[0193] The component filtering module 807 is used to filter candidate components based on component rating data to obtain a list of target components.

[0194] The specific implementation of this component retrieval device is basically the same as the specific implementation of the component retrieval method described above, and will not be repeated here.

[0195] This application also provides an electronic device, which includes: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for communication between the processor and the memory. When the program is executed by the processor, it implements the aforementioned component retrieval method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0196] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0197] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0198] The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the component retrieval method of the embodiments of this application.

[0199] The input / output interface 903 is used to implement information input and output;

[0200] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0201] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0202] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0203] This application also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the above-described component retrieval method.

[0204] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0205] The component retrieval method, component retrieval device, electronic device, and computer-readable storage medium provided in this application obtain a component retrieval request and acquire original components from a preset component library; deconstruct the original components to obtain key component data; and construct a multi-dimensional feature model of the components based on the key component data. This multi-dimensional feature model includes multiple candidate components and component features of each candidate component. This method can conveniently deconstruct each original component hierarchically to obtain the constituent elements used to compose the original component, enabling analysis at the element level to determine whether each original component meets the retrieval requirements. Furthermore, based on the multi-dimensional feature model and pre-acquired component asset data, a multi-dimensional feature database is constructed, allowing direct use of the feature content information in the multi-dimensional feature database during subsequent retrieval processes, effectively saving component retrieval time and improving component retrieval efficiency. Furthermore, based on a multi-dimensional feature database and a pre-defined algorithm, component features are weighted and scored to obtain a feature weight matrix. Candidate components are then scored based on the component retrieval request and the feature weight matrix to obtain component score data. This component score data represents the degree of matching between candidate components and the component retrieval request. It quantifies the overall relevance of each candidate component to the retrieval request, clearly and directly reflecting which candidate components are more relevant, thus making the component scoring process more granular. Finally, candidate components are filtered based on the component score data to obtain a target component list. This effectively improves the accuracy of component retrieval, enabling rapid retrieval of components suitable for financial trading software design. This significantly improves the design efficiency and accuracy of financial trading software, resulting in better security and timeliness in financial transactions between the designed software and network objects.

[0206] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0207] It will be understood by those skilled in the art that Figure 1-7 The technical solutions shown do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0208] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0209] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0210] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0211] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0212] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0213] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0214] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0215] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0216] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A component retrieval method, characterized in that, The method includes: Obtain the component retrieval request and retrieve the original component from the preset component library; The original component is deconstructed to obtain the key component data. Based on the key component data, a multidimensional feature model of the components is constructed, wherein the multidimensional feature model of the components includes multiple candidate components and the component features of each candidate component; Based on the multidimensional feature model of the components and the pre-acquired component asset data, a multidimensional feature database is constructed; Based on the multidimensional feature database and the preset algorithm, the component features are weighted and scored to obtain a feature weight matrix; The candidate components are retrieved and scored based on the component retrieval request and the feature weight matrix to obtain component score data, wherein the component score data is used to represent the degree of matching between the candidate components and the component retrieval request; The candidate components are filtered based on the component scoring data to obtain a target component list; The component features include first feature features and second feature features. The component features are weighted and scored based on the multidimensional feature database and a preset algorithm to obtain a feature weight matrix, including: The first element feature is scored for importance based on the preset algorithm to obtain a first importance matrix; the first element feature is scored for importance based on the first importance matrix to obtain a first weight matrix; the second element feature is scored for importance based on the preset algorithm to obtain a second importance matrix; and the feature weight matrix is ​​obtained based on the first weight matrix and the second importance matrix. The process of deconstructing the original component to obtain its key constituent data includes: The original components are first deconstructed to obtain primary element data; the primary element data is then second deconstructed to obtain secondary element data; and the key component data is obtained based on the primary element data and the secondary element data.

2. The component retrieval method according to claim 1, characterized in that, The step of scoring the first element feature based on the first importance matrix to obtain the first weight matrix includes: Based on the first importance matrix, feature scores are performed on the first element feature to obtain a preliminary score vector for the first element feature. The preliminary scoring vector is standardized to obtain the target scoring vector; The target score vector and the first importance matrix are validated to obtain validation data, which is used to indicate whether there are logical errors in the target score vector and the first importance matrix. If the verification data indicates that there are no logical errors in the target score vector and the first importance matrix, then the first weight matrix is ​​obtained based on the target score vector.

3. The component retrieval method according to claim 1, characterized in that, The step of retrieving and scoring the candidate components based on the component retrieval request and the feature weight matrix to obtain component scoring data includes: Extract the component feature text of each candidate component from the multidimensional feature database, wherein the component feature text includes the text content information of the component features of the candidate component; Content extraction is performed on the component feature text to obtain the component semantic vector corresponding to the component feature text; Content extraction is performed on the component retrieval request to obtain the retrieval semantic vector corresponding to the component retrieval request; A vector similarity matrix is ​​constructed based on the component semantic vector and the retrieval semantic vector, wherein the vector similarity matrix is ​​used to represent the degree of semantic similarity between the component features and the component retrieval request; The component scoring data is obtained by scoring the components based on the feature weight matrix and the vector similarity matrix.

4. The component retrieval method according to claim 1, characterized in that, The construction of a multidimensional feature database based on the component multidimensional feature model and pre-acquired component asset data includes: The candidate components are numbered according to a preset numbering rule to obtain the component number of the candidate components, wherein any two component numbers are different; Row data is obtained based on the component number, column data is obtained based on the component name and component features of the candidate components, and a feature data table is constructed based on the row data and the column data; The feature data table is populated with the component asset data to obtain the multidimensional feature database.

5. The component retrieval method according to any one of claims 1 to 4, characterized in that, The process of filtering candidate components based on the component scoring data to obtain a target component list includes: Compare the component scoring data with a preset threshold; Candidate components whose component score data is greater than or equal to the preset threshold are selected as intermediate components; The intermediate components are sorted in descending order based on the component rating data to obtain the target component list.

6. A component retrieval device, characterized in that, The device includes: The acquisition module is used to acquire component retrieval requests and obtain the original components from a preset component library; The deconstruction module is used to deconstruct the original component to obtain the key component data of the original component; The model building module is used to build a multi-dimensional feature model of components based on the key component data, wherein the multi-dimensional feature model of components includes multiple candidate components and component features of each candidate component; The database construction module is used to construct a multidimensional feature database based on the component multidimensional feature model and pre-acquired component asset data; The weight scoring module is used to perform weight scoring on the component features based on the multidimensional feature database and a preset algorithm to obtain a feature weight matrix. The retrieval and scoring module is used to score the candidate components based on the component retrieval request and the feature weight matrix to obtain component scoring data, wherein the component scoring data is used to represent the degree of matching between the candidate component and the component retrieval request; The component filtering module is used to filter the candidate components based on the component rating data to obtain a list of target components; The component features include first feature features and second feature features. The component features are weighted and scored based on the multidimensional feature database and a preset algorithm to obtain a feature weight matrix, including: The first element feature is scored for importance based on the preset algorithm to obtain a first importance matrix; the first element feature is scored for importance based on the first importance matrix to obtain a first weight matrix; the second element feature is scored for importance based on the preset algorithm to obtain a second importance matrix; and the feature weight matrix is ​​obtained based on the first weight matrix and the second importance matrix. The process of deconstructing the original component to obtain its key constituent data includes: The original components are first deconstructed to obtain primary element data; the primary element data is then second deconstructed to obtain secondary element data; and the key component data is obtained based on the primary element data and the secondary element data.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the component retrieval method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the component retrieval method according to any one of claims 1 to 5.

Images