Industrial supply and demand recommendation method, system and device and storage medium

Abstract

The invention relates to the technical field of data processing, and particularly provides an industrial supply and demand recommendation method, system and device and a storage medium, and the method comprises the steps: constructing a case library of two-dimensional indexes based on predefined industry and link classification, and associating each index item with a service provider case set; collecting diagnosis data of a demand side enterprise, wherein the diagnosis data at least comprises industries, links and multi-dimensional diagnosis indexes; recalling an initial service provider set from an index database based on industry and link information in the diagnosis data; constructing an enterprise portrait based on the diagnosis data, calculating the matching degree between each service provider and the enterprise by using a multi-level dynamic weight model at least comprising a basic matching layer, a context awareness adjustment layer and a collaborative filtering layer, and generating a preliminary recommendation list; and filtering the preliminary list according to a preset access rule, and outputting a final recommendation list. According to the invention, efficient, accurate and reliable supply and demand matching is realized.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, specifically relating to an industrial supply and demand recommendation method, system, device, and storage medium. Background Technology

[0002] In the industrial services sector, accurately matching supply and demand is key to improving industry efficiency. Traditional service provider recommendation methods mainly rely on simple industry classification tags or matching based on fixed-weight rules. These methods have significant drawbacks: first, the matching dimensions are limited, failing to reflect the in-depth needs of enterprises in specific business processes; second, the recommendation rules are rigid, employing static "if-else" logic, unable to dynamically adjust according to the enterprise's real-time status (such as digitalization level, budget pressure, and implementation urgency); and third, the lack of effective quality control may result in recommendations from service providers with insufficient capabilities. Therefore, existing technologies struggle to achieve personalized, accurate, and high-quality industrial service recommendations, urgently requiring a solution that deeply integrates dynamic enterprise diagnostic profiles with intelligent matching algorithms. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present invention provides an industrial supply and demand recommendation method, system, device and storage medium to solve the above-mentioned technical problems.

[0004] In a first aspect, the present invention provides an industrial supply and demand recommendation method, comprising: Based on predefined industry categories and business processes, a case library is created indexed by industry identifiers and process identifiers, with each index item associated with a service provider's case set; Collect diagnostic data from demand-side enterprises, including at least industry, process, and multi-dimensional diagnostic indicators; Based on the industry and process information in the diagnostic data, the initial set of service providers is recalled from the index database; Based on the diagnostic data, a corporate profile is constructed, and a multi-level dynamic weight model is used to calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise, generating a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer, and a collaborative filtering layer; The preliminary recommendation list is filtered according to preset admission rules, and the final recommendation list is output.

[0005] In an optional implementation, a case library indexed by industry identifiers and process identifiers is created based on predefined industry categories and business processes, including: An index table is created in the relational database, with industry identifier and process identifier as a composite primary key. Each row in the index table corresponds to a combination of industry and process, and the record stores a set of case identifiers for all service providers serving that combination.

[0006] In one optional implementation, the predefined industry categories include 16 mutually exclusive industrial industry classifications, and the predefined business processes include 6 mutually exclusive core business processes.

[0007] In one optional implementation, diagnostic data from the demand-side enterprise is collected, including: A standardized diagnostic questionnaire is provided to the enterprise through a front-end interactive interface; the questionnaire answers submitted by the enterprise are received and the answers are encapsulated into a structured data message; wherein the diagnostic information is obtained by parsing the structured data message.

[0008] In one optional implementation, a company profile is constructed based on the diagnostic data, and a multi-level dynamic weight model is used to calculate the matching degree between each service provider in the initial service provider set and the demand-side company, generating a preliminary recommendation list, including: Based on the diagnostic data, a multidimensional diagnostic vector representing the multidimensional state of the enterprise is calculated and generated, and the enterprise's demand preference vector is derived based on the multidimensional diagnostic vector. The multi-level dynamic weight model calculates the matching degree in the following ways: First, a basic matching calculation is performed based on the enterprise demand preference vector and the corresponding capabilities of the service providers; second, based on the specific dimension values ​​in the multi-dimensional diagnostic vector, the weight coefficients of each demand preference are dynamically adjusted to perform context-aware calculation; simultaneously, the selection preferences of historical enterprises similar to the current enterprise profile are introduced to perform collaborative filtering calculation; finally, the results of the basic matching calculation, context-aware calculation, and collaborative filtering calculation are integrated to generate the comprehensive matching degree of each service provider.

[0009] In one optional implementation, the multidimensional diagnostic vector includes scores for five dimensions: technical foundation score, process maturity, budget pressure, urgency of improvement, and team readiness. The enterprise's demand preference vector is calculated by multiplying a pre-trained preference weight matrix with the multidimensional diagnostic vector, and it includes at least the intensity of demand for industry experience, service quality, price, implementation speed, technical support, and success stories.

[0010] In an optional implementation, the preliminary recommendation list is filtered according to preset admission rules to output a final recommendation list, including: Each service provider in the preliminary recommendation list is compared with a preset hard admission standard. The hard admission standard requires that the service provider has at least a preset number of successful cases in the industry to which the client company belongs, and that the average score of its historical cases in the business process that the company needs to improve is not lower than a preset score threshold. Only service providers that fully meet the hard admission standard are retained to form the final recommendation list.

[0011] Secondly, the present invention provides an industrial supply and demand recommendation system, comprising: The index building module is used to create a case library indexed by industry identifiers and process identifiers based on predefined industry categories and business processes. Each index item is associated with a set of service provider cases. The demand collection module is used to collect diagnostic data from demand-side enterprises. The diagnostic data includes at least industry, process, and multi-dimensional diagnostic indicators. The collection recall module is used to recall the initial service provider set from the index database based on the industry and process information in the diagnostic data; The case matching module is used to construct enterprise profiles based on the diagnostic data and calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise using a multi-level dynamic weight model to generate a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer, and a collaborative filtering layer. The case filtering module is used to filter the preliminary recommendation list according to preset admission rules and output the final recommendation list.

[0012] Thirdly, a device is provided, comprising: Memory, used to store industrial supply and demand recommendation programs; A processor is configured to implement the steps of the industrial supply and demand recommendation method as provided in the first aspect when executing the industrial supply and demand recommendation program.

[0013] Fourthly, a computer-readable storage medium is provided, on which an industrial supply and demand recommendation program is stored, which, when executed by a processor, implements the steps of the industrial supply and demand recommendation method provided in the first aspect.

[0014] The beneficial effects of this invention are as follows: the industrial supply and demand recommendation method, system, device, and storage medium provided by this invention achieve millisecond-level rapid recall of candidate service providers by constructing a two-dimensional index library, and innovatively introduce a multi-layer dynamic weight model that includes basic matching, context awareness, and collaborative filtering. This model can dynamically generate personalized matching weights based on real-time enterprise diagnostic data, significantly improving the accuracy and personalization of recommendations. Simultaneously, by setting hard admission rules to filter the recommendation results, the quality and reliability of the service providers in the final recommendation list are effectively guaranteed. The overall solution overcomes the shortcomings of traditional rule-based matching, which is rigid and has a single dimension, and greatly improves the efficiency, quality, and user satisfaction of supply and demand matching. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic block diagram of a system according to an embodiment of the present invention.

[0018] Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0020] 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 invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0021] The industrial supply and demand recommendation method provided in this embodiment of the invention is executed by a computer device, and correspondingly, the industrial supply and demand recommendation system runs on the computer device.

[0022] Figure 1This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The implementing entity can be an industrial supply and demand recommendation system. Depending on different needs, the order of steps in this flowchart can be changed, and some steps can be omitted.

[0023] like Figure 1 As shown, the method includes: S1. Based on predefined industry categories and business processes, create a case library indexed by industry identifiers and process identifiers, with each index item associated with a service provider case set; S2. Collect diagnostic data from demand-side enterprises, including at least industry, process, and multi-dimensional diagnostic indicators; S3. Based on the industry and process information in the diagnostic data, recall the initial service provider set from the index database; S4. Construct an enterprise profile based on the diagnostic data, and use a multi-level dynamic weight model to calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise, and generate a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer and a collaborative filtering layer; S5. Filter the preliminary recommendation list according to the preset admission rules and output the final recommendation list.

[0024] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0025] The two-dimensional case index is implemented using a relational database (such as MySQL). Specifically, a table named `case_index` is created, containing two core fields: `industry_id` (industry identifier, integer type) and `link_id` (stage identifier, integer type), which are set as a composite primary key. Each row in the table uniquely corresponds to an "industry-stage" combination. The `case_id_list` field (text or collection type) of this row is used to store a collection of service provider case identifiers, such as a JSON array ["case_001", "case_003"], thereby efficiently associating all successful cases serving that specific scenario.

[0026] Regarding the index dimensions, this embodiment predefines 16 mutually exclusive industrial sector categories, including but not limited to machinery manufacturing, electronic information, chemical materials, and auto parts; it also defines 6 mutually exclusive core business processes, covering R&D design, production and manufacturing, supply chain management, marketing and sales, after-sales service, and digital upgrades. This refined and mutually exclusive two-dimensional classification system ensures the accuracy and recall efficiency of the index, laying a structured data foundation for subsequent precise matching.

[0027] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0028] The collection of diagnostic data is achieved through a standardized front-end interaction process. The system presents the requesting enterprise with a standardized diagnostic questionnaire containing eight core questions, which is presented through a web or mobile interface. The questions cover multiple dimensions, including the enterprise's industry, business processes that urgently need improvement, equipment digitization rate, existing system coverage, annual IT budget range, and expected improvement timeline.

[0029] After a company submits a questionnaire, the front-end serializes the answers and sends them to the back-end service via an HTTP request. The back-end receiving module encapsulates the answers into a structured JSON data message. This message has a predefined key-value pair structure, for example: {"industry_id":3, "link_id":2, "digitization_rate": 65.5, "budget_level":"B", "urgency_score":4}.

[0030] The system's built-in parser extracts and verifies the values ​​of each field from the JSON message, converting it into a standard diagnostic information object that the system can process internally, for use by subsequent profile building and matching calculation modules. This process ensures the structured, standardized, and efficient access of the data source.

[0031] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0032] The process of retrieving the initial set of service providers from the index based on industry and process information in the diagnostic data is achieved through an efficient query service. The specific process is as follows: After receiving diagnostic data containing industry_id and link_id, the recall module combines these two identifiers into a joint query condition. Subsequently, the module initiates a parameterized query to the case_index table in the relational database (e.g., the SQL statement: SELECT case_id_list FROM case_index WHERE industry_id=? AND link_id=?).

[0033] The database uses a composite primary key for efficient retrieval, returning a unique matching row record within milliseconds. The system extracts a set of service provider case identifiers (e.g., a JSON array) from the `case_id_list` field of this record. This set represents a list of all service provider IDs with successful cases in the target industry and target stage.

[0034] This list is parsed and encapsulated into an initial service provider collection object, which is then directly passed as output to the downstream dynamic weight matching module. This recall mechanism based on precise key-value pair queries ensures high relevance of the candidate set and extremely fast recall, laying a data foundation of controllable quantity and quality for subsequent fine-grained ranking.

[0035] In one embodiment of the present invention, based on step S4, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0036] S401. In-depth Construction and Quantification of Enterprise Diagnostic Profiles After receiving structured diagnostic data, the system first performs data cleaning and standardization preprocessing to ensure the uniformity and validity of the input indicators. Then, it activates the enterprise diagnostic profiling engine, whose core task is to generate a comprehensive and quantitative digital profile that reflects the enterprise's current status and internal needs.

[0037] 1. Refined calculation of the multidimensional diagnostic vector E This vector consists of five rigorously defined dimensions, each synthesized from multiple diagnostic metrics, aiming to go beyond a single digitization rate and comprehensively assess a company's fundamentals: Technical Foundation Score: This score comprehensively assesses the company's "hard" technical asset level. The calculation formula is: Technical Foundation Score = 0.4 * Equipment Digitization Rate + 0.3 * Core Business System Coverage Rate + 0.2 * Automatic Data Collection Rate + 0.1 * IT Infrastructure Standardization Level. All indicator data are from questionnaires and have been standardized and scaled to the [0,100] range using a min-max method before being substituted into the calculation. The final result is normalized to a score of 0-5.

[0038] Process Maturity: This dimension measures the standardization and optimization level of a company's core business processes. By analyzing responses to questions in the questionnaire regarding "Do key business processes have standard operating procedures (SOPs)?", "Is data transfer between process stages automatic and seamless?", and "Are process audits and optimizations conducted regularly?", a rule engine is used to score the process (e.g., 5 points for all, 3 points for some, 1 point for none). Fine-tuning is then performed based on industry benchmarking results for potential process efficiency indicators (such as order delivery cycle and inventory turnover days) found in the diagnostic data.

[0039] Budget Stress Level: This indicator reflects a company's financial flexibility. The system maps the "Annual Budget Range Available for This Improvement" option in the questionnaire to numerical levels (e.g., High = 1, Medium = 2, Low = 3). Budget Stress Level = 1 - (Budget Level Value - 1) / (Total Number of Levels - 1). For example, in a 3-level classification, the stress level corresponding to the low budget level (value 3) is 1, indicating an extremely tight budget; the stress level corresponding to the high budget level (value 1) is 0, indicating a relaxed budget. This value ranges from 0 to 1 and will subsequently be linearly mapped to a 0-5 score range.

[0040] Urgency of Improvement: This directly reflects the intensity of a company's intrinsic motivation to seek external services. The score is a weighted composite of two parts: first, the company's subjectively reported "expected time window for initiating improvement" (e.g., within 1 month = 5 points, 1-3 months = 4 points, 3-6 months = 3 points, more than 6 months = 1 point); and second, indicators indirectly reflecting urgency through analysis of diagnostic data, such as the percentage of revenue caused by current issues (after standardization). The weighting of subjective and objective indicators is typically set at 7:3.

[0041] Team Readiness: Assessing the organization's and personnel's ability to adapt to change. Calculation criteria include: the presence of a dedicated IT / digital team (5 points for a mature team, 3 points for a part-time team, 1 point for none), the percentage of employees who have received relevant training in the past year, and senior management support (derived from specific questionnaire scores). The final score is obtained by weighted averaging of all sub-scores.

[0042] After all dimensions have been calculated, a five-dimensional vector E = [e1, e2, e3, e4, e5] is formed, where each e i ∈[0,5]. The system persists this vector as an authoritative snapshot of the enterprise's current state.

[0043] 2. Derivation and generation of the demand preference vector P This is a crucial step in mapping a firm's state to its potential preferences when selecting a service provider. The system maintains an evolutionary preference weight matrix W. preference This is a 6x5 matrix. Its physical meaning is: each row corresponds to one demand preference (6 in total), and each column corresponds to one dimension of the diagnostic vector (5 in total). Matrix elements. The weight representing the influence of the j-th diagnostic dimension on the intensity of the i-th demand preference.

[0044] The demand preference vector P is defined as: P = [p1, p2, p3, p4, p5, p6], representing: p1: Industry experience required; p2: Service quality requirements; p3: Price sensitivity (the higher the value, the more sensitive the price). p4: Implementation speed requirements; p5: Technical support requirements; p6: Success case requirements.

[0045] Calculation process: P=W through matrix multiplication preference • E is realized. That is, for each demand preference i, its intensity After the calculation is completed, P is usually normalized so that the sum of the preferences is 1 or within a fixed range, which facilitates subsequent comparisons.

[0046] Weight matrix W preference Training and updating: This matrix is ​​not statically configured but continuously optimized through machine learning. Training data comes from historical transaction cases: When a company (whose diagnostic vector E is known) ultimately chooses a service provider, the system analyzes the service provider's significant advantages (such as numerous industry cases, low prices, and extremely fast implementation) to infer the actual preference intensity distribution of the company during decision-making (as the true label of P). Through a large number of such (E, P_true) data pairs, algorithms such as ridge regression are used to periodically train and update W. preference This makes its predicted P increasingly closer to the actual preferences exhibited by firms in their actual decision-making.

[0047] S402. Detailed Execution Flow of the Multi-Level Dynamic Weight Calculation Model At this stage, the system targets each candidate service provider S in the initial service provider set. k Start the dynamic weight calculation pipeline.

[0048] 1. Basic matching layer calculation The goal of this layer is to align the company's general needs and preferences with the service provider's stated capabilities.

[0049] Input: Enterprise demand preference vector P; Service Provider S k Static capability profile vector S static =[s1,s2,s3,s4], where s1 is the industry focus (the industry's revenue share), s2 is the historical average satisfaction score, s3 is the typical case complexity score, and s4 is the price level index (logarithmically processed and smoothed).

[0050] The matching process is not a simple dot product, but rather involves constructing a basic matching weight vector W. base Its initial value is usually set as the average weight (e.g., [0.25, 0.25, 0.25, 0.25] corresponds to the four static capability dimensions, or the mapped weights aligned with the dimensions of P). Then the basic matching score is calculated: .in, It is a matching function. For example, for "price sensitivity p3" and "price level index s4", the matching value might be 1 / (1+|p3*10-s4|), indicating that the closer the preference is to the quote, the higher the score. This layer of calculation ensures that the service provider's fundamental capabilities match the core needs publicly stated by the company.

[0051] 2. Context-Aware Adjustment Layer Computation (Core Innovation) This layer is the intelligent core of the system. It dynamically reshapes the relative importance of each dimension of demand based on the enterprise's unique, real-time contextual state.

[0052] Context recognition engine: The system scans key dimension values ​​in the diagnostic vector E, triggering a predefined set of business rules. The rules are in "IF-THEN" form, but the output is a weight adjustment coefficient.

[0053] Example Rule 1 (Emergency Scenario): IFe4 (Improvement Urgency) > 4.0 THEN Implementation Speed ​​Need (p4) Weighting Coefficient * = 1.5; Price Sensitivity (p3) Weighting Coefficient * = 0.8 (Willingness to Pay for Speed).

[0054] Example Rule 2 (Budget Tight Scenario): IFe3 (Budget Pressure) > 0.7 THEN Price Sensitivity (p3) Weighting Coefficient * = 1.8; Meanwhile, in the service provider characteristics, the flag "Provide Flexible Installment Payment Plans" will receive additional points in subsequent calculations.

[0055] Example Rule 3 (Weak Capability Situation): If IFe5 (team readiness) < 0.3, the weight coefficient of THEN technical support requirement (p5) is increased to 1.6; the weight of the capability item "providing in-depth training and on-site support" in the service provider profile is increased.

[0056] Weight Adjustment and Calculation: After applying all triggered rules, a dynamically adjusted context weight vector W is obtained for this match. situation This vector may already be quite different from W_base. Then, using W... situation Recalculate the matching score: Here S dynamic Service Provider The dynamic profile of a company relative to its current characteristics includes: Industry similarity: The proportion of cases under the target industry-process combination to the total number of cases.

[0057] Scale matching degree: based on Cosine similarity between historical customer size and current enterprise size.

[0058] Challenge coping ability: targeting low-scoring dimensions in E (such as low technical foundation score). Success rate of improvements in similar historical cases.

[0059] ROI Prediction: The projected return on investment for this collaboration, based on similar historical collaboration data and using a regression model.

[0060] 3. Collaborative Filtering Enhancement Layer Calculation This layer introduces "collective intelligence" to avoid inaccurate recommendations due to biases or limitations of a single company profile.

[0061] Similar enterprise search: Using the current enterprise diagnostic vector E as the query point, the system quickly finds the K most similar historical enterprises (e.g., K=50) in the historical enterprise database using Euclidean distance or cosine similarity.

[0062] Group Preference Mining: Analyzing the decision-making data of these K similar companies when choosing a service provider. For example, statistics show that over 80% of the companies ultimately chose the service provider with the highest score in the "industry experience" dimension, while only 20% chose the service provider with the "lowest price." This indicates that in this specific context, "industry experience" is a strong implicit preference.

[0063] Collaborative weight generation: Based on the above analysis, a collaborative weight vector W_collaborative is generated, which strengthens the demand dimensions that are generally valued by similar groups (such as industry experience) and weakens the dimensions that are generally ignored (such as price).

[0064] Collaborative calculation: ,in Service Provider The comprehensive capability assessment value on the i-th demand dimension.

[0065] S403. Merging, sorting, and generating the initial recommendation list. 1. Fraction fusion: After obtaining the scores M1, M2, and M3 from the three layers of computation, the system does not simply average them, but rather performs a weighted fusion based on the system's current learning stage and confidence level: Final_Score_k = α * M1 k +β*M2 k +γ*M3 k .

[0066] α, β, γ are fusion coefficients, and α + β + γ = 1.

[0067] In the early stages of system deployment (when data is scarce), α is relatively high (e.g., 0.5), relying on basic rules.

[0068] As the context rule is proven effective, the β weight gradually increases (e.g., to 0.4).

[0069] As a sufficient number of successful historical cases are accumulated, the gamma weight (collaborative filtering) will continuously increase (e.g., reaching 0.3), reflecting the system's ability to learn collective wisdom from data. These coefficients can be configured by the administrator or automatically adjusted based on A / B testing results.

[0070] 2. Sorting and List Generation: For all service providers in the initial set Sort the services in descending order by Final_Score_k. The system usually sets a display threshold (e.g., only output the Top 20) or an absolute score threshold (e.g., Final_Score > 0.7), and extracts service providers that meet the criteria from the sorted list to form a preliminary recommendation list L.

[0071] 3. Add contextual information to the list: For each service provider entry in the list, the system generates a summary of its key recommendation reasons. These summaries are directly derived from outstanding performance in the three-layer calculation, such as: "This service provider performs best in 'implementation speed' (due to the high urgency of your project) and 'industry experience' (commonly valued by companies similar to yours) which are of utmost importance to you." This list and its metadata are encapsulated into a structured object, ready to be passed to the downstream quality filtering module.

[0072] In one embodiment of the present invention, based on step S5, a possible embodiment will be given below, and its specific implementation will be described in a non-limiting manner.

[0073] The step of filtering the initial recommendation list according to preset admission rules is performed by a separate quality filtering module. This module defines and maintains a set of hard admission criteria, which in this embodiment includes at least two quantifiable core conditions: The service provider must have at least 5 successful cases completed within the specific industry (determined by industry_id) of the client company.

[0074] The service provider must have an average customer rating of at least 4.8 out of 5.0 for all historical cases related to the specific business process that the company needs to improve (determined by link_id).

[0075] The filtering process is as follows: The system iterates through each service provider in the initial recommendation list and, based on its unique identifier, initiates a joint query to the case database to verify whether it simultaneously meets the above two criteria. For example, a query is executed to obtain the total number of cases and average rating for the service provider in a specified industry and process. Then, the query results are compared with the preset quantity threshold (5) and rating threshold (4.8).

[0076] A service provider is considered compliant only if the query results simultaneously meet the criteria of "number of cases ≥ 5" and "average rating ≥ 4.8". All verified service providers are retained and included in a new set to form the final recommendation list; any service provider that does not meet either criterion is filtered out. This process ensures that every service provider in the final output list possesses verified, high-quality industry-specific experience and reputation, thereby guaranteeing the reliability and usability of the recommendation results.

[0077] In some embodiments, the industrial supply and demand recommendation system may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the industrial supply and demand recommendation system may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) Function for recommending industrial supply and demand.

[0078] In this embodiment, the industrial supply and demand recommendation system can be divided into multiple functional modules based on its functions, such as... Figure 2 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0079] The index building module is used to create a case library indexed by industry identifiers and process identifiers based on predefined industry categories and business processes. Each index item is associated with a set of service provider cases. The demand collection module is used to collect diagnostic data from demand-side enterprises. The diagnostic data includes at least industry, process, and multi-dimensional diagnostic indicators. The collection recall module is used to recall the initial service provider set from the index database based on the industry and process information in the diagnostic data; The case matching module is used to construct enterprise profiles based on the diagnostic data and calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise using a multi-level dynamic weight model to generate a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer, and a collaborative filtering layer. The case filtering module is used to filter the preliminary recommendation list according to preset admission rules and output the final recommendation list.

[0080] Figure 3The industrial supply and demand recommendation method provided in the embodiments of this application can be applied to equipment. Those skilled in the art will understand that the equipment structures involved in the embodiments of this invention do not constitute a limitation on the equipment. Equipment may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, equipment includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Equipment may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0081] The device 300 may include a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0082] The memory 320 can be used to store execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 is able to perform some or all of the steps in the above method embodiments.

[0083] The processor 310 serves as the control center of the storage device, connecting various parts of the electronic device via various interfaces and lines. It executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 310 may consist only of a central processing unit (CPU). In this embodiment of the invention, the CPU may have a single processing core or include multiple processing cores.

[0084] The communication unit 330 is used to establish a communication channel, enabling the storage device to communicate with other devices. It can receive user data sent by other devices or send user data to other devices.

[0085] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, which, when executed, may include some or all of the steps provided in the embodiments of the present invention. The storage medium may be a magnetic disk, an optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0086] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other medium capable of storing program code. It includes several instructions to cause a computer device (which may be a personal computer, a server, or a second device, network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0087] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

[0088] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules 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 systems or modules may be electrical, mechanical, or other forms.

[0089] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0091] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. An industrial supply and demand recommendation method, characterized in that, include: Based on predefined industry categories and business processes, a case library is created indexed by industry identifiers and process identifiers, with each index item associated with a service provider's case set; Collect diagnostic data from demand-side enterprises, including at least industry, process, and multi-dimensional diagnostic indicators; Based on the industry and process information in the diagnostic data, the initial set of service providers is recalled from the index database; Based on the diagnostic data, a corporate profile is constructed, and a multi-level dynamic weight model is used to calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise, generating a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer, and a collaborative filtering layer; The preliminary recommendation list is filtered according to preset admission rules, and the final recommendation list is output.

2. The method according to claim 1, characterized in that, Based on predefined industry categories and business processes, a case library indexed by industry identifiers and process identifiers is created, including: An index table is created in the relational database, with industry identifier and process identifier as a composite primary key. Each row in the index table corresponds to a combination of industry and process, and the record stores a set of case identifiers for all service providers serving that combination.

3. The method according to claim 2, characterized in that, The predefined industry categories include 16 mutually exclusive industrial industry classifications, and the predefined business processes include 6 mutually exclusive core business processes.

4. The method according to claim 1, characterized in that, Collect diagnostic data from the demand-side companies, including: A standardized diagnostic questionnaire is provided to the enterprise through a front-end interactive interface; the questionnaire answers submitted by the enterprise are received and the answers are encapsulated into a structured data message; wherein the diagnostic information is obtained by parsing the structured data message.

5. The method according to claim 1, characterized in that, Based on the diagnostic data, a company profile is constructed, and a multi-level dynamic weight model is used to calculate the matching degree between each service provider in the initial service provider set and the demand-side company, generating a preliminary recommendation list, including: Based on the diagnostic data, a multidimensional diagnostic vector representing the multidimensional state of the enterprise is calculated and generated, and the enterprise's demand preference vector is derived based on the multidimensional diagnostic vector. The multi-level dynamic weight model calculates the matching degree in the following ways: First, a basic matching calculation is performed based on the enterprise demand preference vector and the corresponding capabilities of the service providers; second, based on the specific dimension values ​​in the multi-dimensional diagnostic vector, the weight coefficients of each demand preference are dynamically adjusted to perform context-aware calculation; simultaneously, the selection preferences of historical enterprises similar to the current enterprise profile are introduced to perform collaborative filtering calculation; finally, the results of the basic matching calculation, context-aware calculation, and collaborative filtering calculation are integrated to generate the comprehensive matching degree of each service provider.

6. The method according to claim 5, characterized in that, The multidimensional diagnostic vector includes scores for five dimensions: technical foundation, process maturity, budget pressure, urgency of improvement, and team readiness. The enterprise's demand preference vector is calculated by multiplying a pre-trained preference weight matrix with the multidimensional diagnostic vector, and it includes at least the intensity of demand for industry experience, service quality, price, implementation speed, technical support, and success stories.

7. The method according to claim 1, characterized in that, The preliminary recommendation list is filtered according to preset admission rules to output a final recommendation list, including: Each service provider in the preliminary recommendation list is compared with a preset hard admission standard. The hard admission standard requires that the service provider has at least a preset number of successful cases in the industry to which the client company belongs, and that the average score of its historical cases in the business process that the company needs to improve is not lower than a preset score threshold. Only service providers that fully meet the hard admission standard are retained to form the final recommendation list.

8. An industrial supply and demand recommendation system, characterized in that, include: The index building module is used to create a case library indexed by industry identifiers and process identifiers based on predefined industry categories and business processes. Each index item is associated with a set of service provider cases. The demand collection module is used to collect diagnostic data from demand-side enterprises. The diagnostic data includes at least industry, process, and multi-dimensional diagnostic indicators. The collection recall module is used to recall the initial service provider set from the index database based on the industry and process information in the diagnostic data; The case matching module is used to construct enterprise profiles based on the diagnostic data and calculate the matching degree between each service provider in the initial service provider set and the demand-side enterprise using a multi-level dynamic weight model to generate a preliminary recommendation list; the multi-level dynamic weight model includes at least a basic matching layer, a context-aware adjustment layer, and a collaborative filtering layer. The case filtering module is used to filter the preliminary recommendation list according to preset admission rules and output the final recommendation list.

9. An industrial supply and demand recommendation device, characterized in that, include: Memory, used to store industrial supply and demand recommendation programs; A processor, configured to implement the steps of the industrial supply and demand recommendation method as described in any one of claims 1-7 when executing the industrial supply and demand recommendation program.

10. A computer-readable storage medium storing a computer program, characterized in that, The readable storage medium stores an industrial supply and demand recommendation program, which, when executed by a processor, implements the steps of the industrial supply and demand recommendation method as described in any one of claims 1-7.

Images