Recommended methods, apparatus and computer-readable storage media for technical objects

Abstract

This application discloses a method, apparatus, and computer-readable storage medium for recommending technical objects. Relating to the field of big data technology, the method includes: receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, determining the similarity between the N historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technical objects corresponding to the target historical work order texts based on a preset collaboration graph network, determining the target objects, where the collaboration graph network represents the relationship between the N historical work order texts and the technical objects; and determining the recommended technical objects corresponding to the work order texts based on the target objects. This application solves the problem that recommendations based on past experience by technical personnel in related technologies are inefficient and inaccurate.

Description

Technical Field

[0001] This application relates to the field of big data technology, and more specifically, to a method, apparatus, and computer-readable storage medium for recommending technical objects. Background Technology

[0002] In bank IT departments, handling complex system failures involving core accounting, interbank payments, and regulatory reporting relies heavily on the experience of a few senior experts. However, experts have limited time and energy, and when recommending technical personnel for different work orders, relying solely on experts leads to low recommendation efficiency and inaccurate results.

[0003] There is currently no effective solution to the problem that recommendations from technical personnel based on past experience are inefficient and inaccurate in related technologies. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, and computer-readable storage medium for recommending technical objects, in order to solve the problems that recommendations made by those skilled in the art based on past experience are inefficient and inaccurate.

[0005] To achieve the above objectives, according to one aspect of this application, a method for recommending technical objects is provided. The method includes: receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, and determining the similarity between the N historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technical objects corresponding to the target historical work order texts based on a preset collaboration graph network, and determining target objects, wherein the collaboration graph network represents the relationship between the N historical work order texts and the technical objects; and determining recommended technical objects corresponding to the work order texts based on the target objects.

[0006] Optionally, the work order text is converted into a target semantic vector, including: performing word segmentation on the work order text to obtain M words, where M is a positive integer; inputting the M words into a preset text embedding model to obtain an initial vector of the work order text; determining the weights corresponding to each of the M words based on preset word frequency weights; and determining the target semantic vector based on the weights of each of the M words and the initial vector of the work order text.

[0007] Optionally, the method further includes: obtaining N historical work order texts and the corresponding work order information for each of the N historical work order texts, wherein the work order information includes technical objects; converting the N historical work order texts into semantic vectors corresponding to each of the N historical work order texts based on a preset text embedding model; determining nodes and edges based on the semantic vectors and work order information corresponding to each of the N historical work order texts, wherein the nodes include work order nodes and technical personnel nodes; and constructing a collaboration graph network based on the nodes and edges.

[0008] Optionally, based on the target object, the recommended technology object corresponding to the work order text is determined, including: when there are L target objects, obtaining the activity level and work order number corresponding to each of the L target objects, where L is a positive integer greater than 1; sorting the L target objects based on the activity level and work order number corresponding to each of the L target objects to determine the target object sequence; and determining the recommended technology object based on the target object sequence.

[0009] Optionally, it also includes: receiving the target solution provided by the recommended technology object based on a preset account; performing a detection operation on the target solution to obtain a detection result, wherein the detection operation includes detecting whether there are syntax errors in the target solution and detecting whether there are security injection vulnerabilities in the target solution; and sending the target solution to the target account if the detection result is that the detection passes.

[0010] Optionally, it also includes: receiving the processing results from the target account after receiving the target solution; and adjusting the collaboration graph network based on the processing results.

[0011] To achieve the above objectives, according to another aspect of this application, a technology object recommendation device is provided. The device includes: a receiving module for receiving a work order text sent by a target account; a conversion module for converting the work order text into a target semantic vector; a calculation module for calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, and determining the similarity between the N historical work order texts, where N is a positive integer; a selection module for selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; a first determination module for selecting a technology object corresponding to the target historical work order text based on a preset collaboration graph network, and determining the target object, wherein the collaboration graph network represents the relationship between the N historical work order texts and the technology object; and a second determination module for determining the recommended technology object corresponding to the work order text based on the target object.

[0012] To achieve the above objectives, according to another aspect of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to execute the method recommended in any of the above-mentioned technical objects.

[0013] To achieve the above objectives, according to another aspect of this application, an electronic device is provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the method recommended by any of the above-mentioned technical objects during runtime.

[0014] To achieve the above objectives, according to another aspect of this application, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the method recommended by any of the above-described technical objects.

[0015] In this embodiment, a technology object recommendation method is adopted. This involves receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technology objects corresponding to the target historical work order texts based on a preset collaboration graph network, where the collaboration graph network represents the relationship between the N historical work order texts and the technology objects; and determining the recommended technology objects corresponding to the work order texts based on the target objects. This achieves the goal of recommending technical personnel by combining collaboration graph networks and similar work order recommendations, thereby improving recommendation efficiency and accuracy. This solves the technical problem in related technologies where recommending technical personnel based on past experience is inefficient and inaccurate. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 A hardware structure block diagram of a computer terminal for implementing a technology object recommendation method is shown.

[0018] Figure 2 This is a flowchart of the method for recommending technical objects according to the embodiments of this application;

[0019] Figure 3 This is a structural block diagram of the recommended device based on the embodiments of this application;

[0020] Figure 4 This is a structural block diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, 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.

[0023] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0024] Example 1

[0025] According to an embodiment of this application, a method embodiment for recommending technical objects is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a technology object recommendation method is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0027] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0028] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the technical object recommendation method in the embodiments of this application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned technical object recommendation method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0029] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0030] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0031] Under the aforementioned operating environment, this application provides the following: Figure 2 Recommended methods for the technical objects shown. Figure 2 This is a flowchart of a method for recommending technical objects according to embodiments of this application.

[0032] Step S201: Receive the work order text sent by the target account.

[0033] In this step, a work order refers to a document or electronic record used to record, track, and manage specific issues, tasks, or service requests.

[0034] Typically, the system provides one or more ticket submission interfaces or channels. The target account (i.e., the user or technician submitting the ticket) can submit the ticket text through these channels. Before receiving the ticket text, user authentication is performed to ensure that the submitter is the target account with submission permissions. This is accomplished through security measures such as username / password, two-factor authentication, and access tokens to prevent unauthorized access and malicious submissions.

[0035] The received work order texts may come from different channels, with varying formats and content structures. Therefore, we can first standardize this text information, extracting key fields such as work order title, problem description, requester information, priority, and category, and then store this information in a structured manner for easier subsequent processing and retrieval.

[0036] Step S202: Convert the work order text into a target semantic vector.

[0037] In this step, the work order text can first be preprocessed, for example, by cleaning the text to remove irrelevant symbols, stop words, and noise, while retaining the core information. Then, the continuous text is segmented into independent lexical units. A pre-trained word embedding model can be used to map each word into a multi-dimensional vector, reflecting the meaning of the word in context. These word embedding vectors are combined to form sentence-level vectors, which can be achieved using averaging, weighted summation, or more complex models such as RNNs (Recurrent Neural Networks) and LSTMs (Long Short-Term Memory Networks). Finally, the entire work order text (including title, description, system environment, error log, etc.) is treated as a document, and sentence vectors are aggregated to generate work order-level semantic vectors.

[0038] Through the above steps, the work order text is transformed into a vector representation that reflects its core semantics and intent, which can be used for accurate matching and recommendation. It helps the system understand the real needs of each work order, thereby finding the most relevant solutions and expert resources.

[0039] Step S203: Calculate the similarity between the target semantic vector and the semantic vectors corresponding to each of the N preset historical work order texts, and determine the similarity between each of the N historical work order texts, where N is a positive integer.

[0040] In this step, a set of semantic vectors for N historical work orders can be read from a pre-prepared database or file. Then, the similarity between the target semantic vector and the semantic vectors corresponding to each of the N historical work order texts is calculated. Specifically, cosine similarity can be calculated; for each historical work order's semantic vector, a cosine similarity is calculated between it and the target semantic vector. During the calculation process, the ID of each historical work order and its similarity value can be stored as a pair, such as (work order ID, similarity value).

[0041] Step S204: Select historical work order texts with similarity exceeding a preset threshold as target historical work order texts.

[0042] In this step, a reasonable similarity threshold, such as 0.75, can be set based on past experience and the characteristics of the current work order. The similarity of each work order is checked one by one from the previously calculated historical work orders and their similarity list. If the similarity score of a historical work order is higher than the pre-set threshold, the work order is retained; otherwise, it is removed from the candidate list. This step will generate a new, streamlined list of historical work orders, in which all work orders have a high semantic similarity to the target work order.

[0043] Specifically, it can also sort the historical work orders from high to low similarity and recommend the Top-N most similar ones to the technical staff, including their solutions and technical staff.

[0044] By following the steps above, we can selectively recommend historical records that are truly similar in meaning to the current work order, thereby providing users with more valuable information, speeding up the resolution of problems, and reducing the interference of invalid information.

[0045] Step S205: Based on the preset collaboration graph network, select the technical object corresponding to the target historical work order text and determine the target object. The collaboration graph network represents the relationship between N historical work order texts and technical objects.

[0046] In this step, the collaboration graph network includes work order nodes, technical tag nodes, and technical personnel nodes, as well as the edges connecting them. The ID of the target historical work order can be determined to locate its position in the network. Each work order node represents a specific work order and stores all information about that work order, including its semantic vector and the technical tag and technical personnel nodes connected to it.

[0047] In the collaboration graph network, identify edges directly connected to the target historical work order node, especially those pointing to technical tag nodes. For each connection edge, extract the connected technical tag nodes to obtain the name of the technical field or specific technical point involved in the work order. If the edges in the collaboration graph network have weights, further analysis can be performed to determine which technical tags are more closely related to the work order; technical tags with higher weights are likely the core technical focus of the work order. Combining the work order description and the technical tag information extracted from the network, the target object can be identified, and recommended technical personnel can be identified.

[0048] By following the steps above, a graph network model based on historical work orders and expert collaboration can be used to quickly identify the technical personnel and potential solutions most relevant to the current problem, greatly improving the efficiency of troubleshooting and knowledge sharing.

[0049] Step S206: Based on the target object, determine the recommended technology object corresponding to the work order text.

[0050] In this step, the technical objects and corresponding experts can be sorted according to relevance weights to ensure that the recommended experts and solutions are the most relevant. An information list containing recommended technical objects and their related experts can be created, including the expert's contact information, problem-solving history, and technical expertise. A brief explanation should be attached to each recommendation, explaining why this expert or solution is relevant to the current ticket, for example: "Zhang San successfully solved 10 database performance optimization related issues in the past year." The recommendation list should be displayed to technical personnel or operations teams through a user-friendly interface for quick viewing and selection.

[0051] This method can accurately identify historical work orders with similar semantics, thus surpassing the limitations of simple keyword matching and achieving deep semantic retrieval, significantly improving the accuracy and efficiency of information retrieval. Historical work orders with similarity exceeding a preset threshold are selected as target historical work orders. Based on a preset collaboration graph network, the technical objects associated with these work orders are identified, and technical personnel are recommended. This process effectively mines and shares tacit knowledge within the organization, solves the expert location problem, and thus improves the speed and efficiency of troubleshooting.

[0052] As an optional embodiment, converting the work order text into a target semantic vector includes: performing word segmentation on the work order text to obtain M words, where M is a positive integer; inputting the M words into a preset text embedding model to obtain an initial vector of the work order text; determining the weights corresponding to each of the M words based on preset word frequency weights; and determining the target semantic vector based on the weights of each of the M words and the initial vector of the work order text.

[0053] Optionally, the process of converting the work order text into a target semantic vector first involves word segmentation of the work order text, resulting in M ​​words. These words are then input into a pre-defined text embedding model to generate an initial vector for the work order text. To improve the representativeness of the vector, weight values ​​are determined for each of the M words based on pre-defined word frequency weights. Finally, the target semantic vector is calculated using a weighted fusion method, combining the weights of the M words and the initial vector of the work order text.

[0054] Specifically, the work order text is segmented using an appropriate word segmentation algorithm (such as TF-IDF segmentation, jieba segmentation, or other word segmentation tools in natural language processing libraries) to obtain M words. M is a positive integer representing the number of words in the segmented text. The M words are then input into a pre-trained text embedding model, where each word is transformed into a fixed-length vector reflecting its semantic features. Based on the entire work order corpus or a corpus of a specific domain, the term frequency (i.e., the frequency of the word's occurrence in documents) and document frequency (i.e., the number of documents containing the word) of each word are calculated. Term frequency-inverse document frequency (TF-IDF) or other weighting algorithms can be used to determine the weights of the M words. The weights reflect the importance of the word in the document; generally, words with higher term frequencies but lower document frequencies have higher weights because these words are more likely to be key information in a specific document or domain. Based on the weights of each word obtained in the above steps, the embedding vectors of each word are weighted and summed to generate the initial vector of the work order text.

[0055] The above steps enable us to more accurately capture the semantic features of work order texts, providing more detailed analytical basis for subsequent similar work order matching and expert recommendations, thereby effectively improving the accuracy and efficiency of search and recommendation.

[0056] As an optional embodiment, N historical work order texts and their corresponding work order information are obtained, wherein the work order information includes technical objects; based on a preset text embedding model, the N historical work order texts are transformed into semantic vectors corresponding to each of the N historical work order texts; based on the semantic vectors and work order information corresponding to each of the N historical work order texts, nodes and edges are determined, wherein the nodes include work order nodes and technical personnel nodes; based on the nodes and edges, a collaborative graph network is constructed.

[0057] Optionally, N historical work order texts and their corresponding work order information are first obtained, whereby the work order information includes technical objects associated with the problem. Using a pre-defined text embedding model, these historical work order texts are transformed into semantic vectors. This process allows the textual description of the problem to be represented in mathematical vector form, facilitating subsequent similarity calculations. Based on the transformed semantic vectors and work order information, the nodes and edges of the collaboration graph network are determined. Nodes include work order nodes and technical personnel nodes, while edges represent the relationships between work orders and technical personnel / technical objects. By constructing the collaboration graph network, the system can intuitively display the connections between problems, solvers, and technical tags, providing a foundation for intelligent recommendation. When a new work order enters the system, the collaboration graph network can quickly locate historical work orders with similar semantics, as well as technical personnel who have handled similar problems.

[0058] Specifically, the text and information of N historical work orders can be loaded from a database or file system. Work order information should include, but is not limited to, work order ID, problem description, solution, handler, and related technical objects (tags). Each historical work order text is preprocessed, including word segmentation, stop word removal, stemming, or lemmatization, to improve the quality of the text vector. A pre-defined text embedding model (such as Word2Vec, BERT, FastText, etc.) is used to convert the preprocessed text into a semantic vector. Each historical work order will correspond to a vector reflecting its semantic features. A "work order node" is created for each historical work order, with attributes including work order ID, semantic vector, and related technical objects; a "technical tag node" is created for each mentioned technical object; and a "technical personnel node" is created for each technical personnel involved in solving the problem. Relationships between nodes are determined, such as the "inclusion" relationship between work orders and technical tags, the "solution" relationship between work orders and technical personnel, and the "expertise" relationship between technical personnel and technical tags. Edges are created between nodes based on the work order information. For example, if historical work order A is related to technical object B, then create an "include" edge between work order node A and technical tag node B; if technician C resolved work order A, then create a "resolved" edge between technician node C and work order node A. Using graph database technology or graph data structures, the nodes and edges defined in the above steps are integrated to form a collaborative graph network.

[0059] To improve the network's query efficiency and recommendation accuracy, weights can be assigned to each edge. For example, for the "Resolve" edge, the weight could be based on factors such as the speed and success rate at which technicians resolve the work order. The collaboration graph network should be updated periodically or based on the addition of new work orders. Ensure the network reflects the latest work order processing status and changes in personnel skills.

[0060] This graph network-based recommendation mechanism not only accelerates the problem-solving process and improves the accuracy and quality of solutions, but also promotes knowledge sharing and expert identification within the organization. At the same time, through automated simulation and inspection, it reduces operational risks, enables the automatic expansion of experiential knowledge and continuous system optimization, and effectively improves team collaboration efficiency and system stability.

[0061] As an optional embodiment, determining the recommended technology object corresponding to the work order text based on the target object includes: when there are L target objects, obtaining the activity level and work order number corresponding to each of the L target objects, where L is a positive integer greater than 1; sorting the L target objects based on the activity level and work order number corresponding to each of the L target objects to determine the target object sequence; and determining the recommended technology object based on the target object sequence.

[0062] Optionally, when determining the recommended technical object corresponding to the work order text based on the target object, the system first obtains the activity level and number of work orders for each of the L target objects, where L represents a positive integer greater than 1. This step quantifies the participation and expertise of technical personnel. Subsequently, the system sorts the L target objects based on the collected data, constructing a target object sequence. This process allows the system to prioritize technical personnel with extensive experience in specific fields and recent activity. Finally, based on the target object sequence, the system selects the most suitable potential experts for recommendation, ensuring that the recommendation results reflect both individual expertise and their contribution to the team and response speed, thereby improving problem-solving efficiency and satisfaction.

[0063] For example, in the collaboration graph, find the "Technical Personnel" nodes connected to the recommended similar historical work orders. These technical personnel are then sorted and filtered (e.g., based on the number of similar work orders they have resolved, recent activity levels, etc.). Finally, a list of potential experts is generated, with clear annotations indicating the reasons for the recommendations, such as: "Recommended Consultant: Li Si (Backend Team); Reason for Recommendation: Successfully resolved 3 issues with semantic similarity exceeding 85%."

[0064] This data-driven approach not only allows us to quickly identify the most suitable experts for solving problems, but also promotes the effective dissemination and utilization of knowledge within the team, creating a positive learning and working environment.

[0065] As an optional embodiment, the recommended technical object receives a target solution based on feedback from a preset account; a detection operation is performed on the target solution to obtain a detection result, wherein the detection operation includes detecting whether there are syntax errors in the target solution and detecting whether there are security injection vulnerabilities in the target solution; if the detection result is that the detection passes, the target solution is sent to the target account.

[0066] Optionally, the system can inspect the target solutions provided by the recommended technical objects. By automatically detecting syntax errors and security injection vulnerabilities in the target solutions, the accuracy and security of the solutions are ensured. When the inspection result indicates that the solution passes, the system sends the target solution to the target account, realizing the efficient reuse of faulty solutions and the effective transfer of expert experience.

[0067] Specifically, automated checks can be initiated once the technical staff has drafted the solution and is ready to respond:

[0068] (1) Script and code analysis: For SQL scripts, Shell commands, configuration code, etc. included in the solution, the system can perform static code analysis and dynamic security scanning in an isolated sandbox environment.

[0069] (2) Risk identification and warning: Check for risks such as syntax errors, potential slow queries, security injection vulnerabilities, and incompatible configuration changes.

[0070] (3) Generate a risk report: Before sending, pop up a prompt to the technical staff, such as: "Warning: The SQL script you submitted is missing a WHERE condition, which may result in a full table update. Please confirm." or "Tip: This configuration change is inconsistent with the official best practice. See the link for reference."

[0071] By testing and verifying the feedback solutions, the quality of the solutions was ensured, team trust and collaboration were strengthened, and the stability and operational efficiency of the bank's IT system were improved.

[0072] As an optional embodiment, the processing result fed back by the target account after receiving the target solution is received; based on the processing result, the collaboration graph network is adjusted.

[0073] Optionally, after receiving feedback from the target account regarding the processing results after resolving a work order, the system automatically adjusts the collaboration graph network based on these results. When the target account (usually a technical staff member encountering the problem) adopts a solution recommended by the system (or a solution developed after consulting with potential experts recommended by the system), they provide feedback on the processing results through the work order system, including but not limited to whether the problem was resolved, the actual effectiveness of the solution, and any additional comments or suggestions. The system then analyzes this feedback data and dynamically adjusts the nodes and edges in the collaboration graph network to reflect the latest practical results. For example, if a solution consistently receives high satisfaction feedback, the edge weight between that node and related work orders and technical tags will increase, giving it a higher ranking in future recommendations for similar problems. Conversely, if a solution receives negative feedback, its weight will be reduced accordingly, and it may even be removed from the recommendation list. Furthermore, the system can automatically identify and suggest new knowledge to be added to the knowledge base based on user feedback, helping to update and enrich the knowledge base in a timely manner and ensuring the timeliness and practicality of its content.

[0074] Specifically, once the target account marks the issue as "resolved," the knowledge evolution process can be initiated:

[0075] (1) Collect feedback: Initiate a simple feedback process to the problem submitter or relevant parties, such as rating the "satisfaction" of the solution from 1 to 5 stars.

[0076] (2) Dynamic weight adjustment: The system links feedback data to corresponding work orders and solutions. Based on the feedback score, the system dynamically adjusts the ranking weight of the solution in the recommendation of similar problems. Solutions with high satisfaction will be given priority in future recommendations.

[0077] (3) Intelligent knowledge base prompts: Continuously track the number of times each solution is recommended and the average satisfaction level. When a solution is verified to be effective multiple times (for example, recommended more than 10 times and with an average satisfaction level ≥ 4.5 stars), the system will automatically send a prompt to the administrator or the original author of the solution: "This solution has been successfully applied 10 times with a satisfaction level of 4.8 stars. Would you like to consider officially including it in the official knowledge base?"

[0078] This mechanism enables continuous self-optimization, improving recommendation accuracy and solution quality, forming a closed loop of continuous improvement and learning, thereby more effectively enhancing the team's problem-solving capabilities and speed.

[0079] The method for recommending technical objects provided in this application involves receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, thus determining the similarity between the N historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technical objects corresponding to the target historical work order texts based on a preset collaboration graph network, thus determining the target objects, where the collaboration graph network represents the relationship between the N historical work order texts and the technical objects; and determining the recommended technical objects corresponding to the work order texts based on the target objects. This achieves the goal of recommending technical personnel by combining collaboration graph networks and similar work order recommendations, thereby improving the technical effect of recommendation efficiency and accuracy, and solving the technical problem in related technologies where the recommendation of technical personnel based on past experience is inefficient and inaccurate.

[0080] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0081] Example 2

[0082] This application also provides a technology object recommendation device. It should be noted that the technology object recommendation device of this application can be used to execute the technology object recommendation method provided in this application. The technology object recommendation device provided in this application will be described below.

[0083] According to an embodiment of this application, an apparatus for implementing the above-described method for recommending technical objects is also provided. Figure 3 This is a structural block diagram of the recommended device based on the embodiments of this application, such as... Figure 3 As shown, the device includes:

[0084] The receiving module 301 is used to receive the work order text sent by the target account.

[0085] The conversion module 302, connected to the receiving module 301, is used to convert the work order text into a target semantic vector.

[0086] The calculation module 303, connected to the conversion module 302, is used to calculate the similarity between the target semantic vector and the semantic vectors corresponding to each of the N preset historical work order texts, and to determine the similarity between each of the N historical work order texts, where N is a positive integer.

[0087] The selection module 304, connected to the calculation module 303, is used to select historical work order texts with similarity exceeding a preset threshold as target historical work order texts.

[0088] The first determining module 305, connected to the selection module 304, is used to select the technical object corresponding to the target historical work order text based on a preset collaboration graph network, and to determine the target object. The collaboration graph network represents the relationship between N historical work order texts and technical objects.

[0089] The second determining module 306, connected to the first determining module 305, is used to determine the recommended technology object corresponding to the work order text based on the target object.

[0090] The technology object recommendation device provided in this application embodiment receives a work order text sent by a target account; converts the work order text into a target semantic vector; calculates the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, and determines the similarity between the N historical work order texts, where N is a positive integer; selects historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selects technology objects corresponding to the target historical work order texts based on a preset collaboration graph network, and determines the target objects, where the collaboration graph network represents the relationship between the N historical work order texts and the technology objects; and determines the recommended technology objects corresponding to the work order texts based on the target objects. This achieves the purpose of recommending technical personnel by combining collaboration graph networks and similar work order recommendations, thereby improving the technical effect of recommendation efficiency and accuracy, and solving the technical problem in related technologies where the recommendation of technical personnel based on past experience is inefficient and inaccurate.

[0091] Optionally, the conversion module is used to convert the work order text into a target semantic vector, including: a word segmentation unit for segmenting the work order text to obtain M words, where M is a positive integer; an input unit for inputting the M words into a preset text embedding model to obtain an initial vector of the work order text; a first determination unit for determining the weights corresponding to each of the M words based on preset word frequency weights; and a second determination unit for determining the target semantic vector based on the weights of each of the M words and the initial vector of the work order text.

[0092] Optionally, the above device further includes: an acquisition module for acquiring N historical work order texts and work order information corresponding to each of the N historical work order texts, wherein the work order information includes technical objects; a vector conversion module for converting the N historical work order texts into semantic vectors corresponding to each of the N historical work order texts based on a preset text embedding model; a third determination module for determining nodes and edges based on the semantic vectors corresponding to each of the N historical work order texts and the work order information corresponding to each of the N historical work order texts, wherein the nodes include work order nodes and technical personnel nodes; and a construction module for constructing a collaborative graph network based on the nodes and edges.

[0093] Optionally, the second determining module is used to determine the recommended technology object corresponding to the work order text based on the target object, including: when there are L target objects, the acquisition unit is used to acquire the activity level and work order number corresponding to each of the L target objects, where L is a positive integer greater than 1; the sorting unit is used to sort the L target objects based on the activity level and work order number corresponding to each of the L target objects to determine the target object sequence; the third determining unit is used to determine the recommended technology object based on the target object sequence.

[0094] Optionally, the above apparatus further includes: a scheme receiving module for receiving a target scheme fed back by the recommended technology object based on a preset account; a detection module for performing a detection operation on the target scheme to obtain a detection result, wherein the detection operation includes detecting whether there are syntax errors in the target scheme and detecting whether there are security injection vulnerabilities in the target scheme; and if the detection result is that the detection is passed, a sending module for sending the target scheme to the target account.

[0095] Optionally, the above apparatus further includes: a result receiving module for receiving the processing results fed back by the target account after receiving the target scheme; and an adjustment module for adjusting the collaboration graph network based on the processing results.

[0096] It should be noted that the receiving module 301, conversion module 302, calculation module 303, selection module 304, first determination module 305, and second determination module 306 mentioned above correspond to steps S201 to S206 in Embodiment 1. The two modules and their corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of a device and run in the computer terminal 10 provided in Embodiment 1.

[0097] Example 3

[0098] Embodiments of this application may provide an electronic device. Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) Processor 402, memory 404, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0099] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0100] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, and determining the similarity between the N historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technical objects corresponding to the target historical work order texts based on a preset collaboration graph network, and determining the target objects, where the collaboration graph network represents the relationship between the N historical work order texts and the technical objects; and determining recommended technical objects corresponding to the work order texts based on the target objects.

[0101] The processor can access the information and application programs stored in the memory via the transmission device to perform the following steps: converting the work order text into a target semantic vector, including: performing word segmentation on the work order text to obtain M words, where M is a positive integer; inputting the M words into a preset text embedding model to obtain the initial vector of the work order text; determining the weights corresponding to each of the M words based on preset word frequency weights; and determining the target semantic vector based on the weights of each of the M words and the initial vector of the work order text.

[0102] The processor can access information and applications stored in memory via a transmission device to perform the following steps: obtaining N historical work order texts and their corresponding work order information, wherein the work order information includes technical objects; converting the N historical work order texts into semantic vectors corresponding to each of the N historical work order texts based on a preset text embedding model; determining nodes and edges based on the semantic vectors and work order information corresponding to each of the N historical work order texts, wherein the nodes include work order nodes and technical personnel nodes; and constructing a collaborative graph network based on the nodes and edges.

[0103] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: Based on the target object, determine the recommended technology object corresponding to the work order text, including: if there are L target objects, obtain the activity level and work order number corresponding to each of the L target objects, where L is a positive integer greater than 1; based on the activity level and work order number corresponding to each of the L target objects, sort the L target objects to determine the target object sequence; based on the target object sequence, determine the recommended technology object.

[0104] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: receiving a target solution from a recommended technology object based on feedback from a preset account; performing a detection operation on the target solution to obtain a detection result, wherein the detection operation includes detecting whether there are syntax errors and security injection vulnerabilities in the target solution; and sending the target solution to the target account if the detection result is a pass.

[0105] The processor can access information and applications stored in memory via a transmission device to perform the following steps: receiving the processing results from the target account after receiving the target solution; and adjusting the collaboration graph network based on the processing results.

[0106] This application provides a method for recommending technical objects. The method involves receiving a work order text from a target account; converting the work order text into a target semantic vector; calculating the similarity between the target semantic vector and the semantic vectors corresponding to N preset historical work order texts, where N is a positive integer; selecting historical work order texts with similarity exceeding a preset threshold as target historical work order texts; selecting technical objects corresponding to the target historical work order texts based on a preset collaboration graph network, where the collaboration graph network represents the relationship between the N historical work order texts and the technical objects; and determining the recommended technical objects corresponding to the work order texts based on the target objects. This method combines collaboration graph networks and similar work order recommendations to recommend technical personnel, thereby improving recommendation efficiency and accuracy. It also solves the technical problem in related technologies where recommending technical personnel based on past experience is inefficient and inaccurate.

[0107] Those skilled in the art will understand that Figure 4 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0108] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0109] Example 4

[0110] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the recommended method of any of the technical objects provided in Embodiment 1.

[0111] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0112] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing the recommended method steps of any of the above-described technical objects.

[0113] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0114] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0115] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units 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 displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0116] The units described 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.

[0117] 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.

[0118] 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 several 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0119] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for recommending technical objects, characterized in that, include: Receive work order text from the target account; Convert the work order text into a target semantic vector; Calculate the similarity between the target semantic vector and the semantic vectors corresponding to each of the N preset historical work order texts, and determine the similarity between each of the N historical work order texts, where N is a positive integer; Select historical work order texts with a similarity exceeding a preset threshold as target historical work order texts; Based on a preset collaboration graph network, a technical object corresponding to the target historical work order text is selected to determine the target object. The collaboration graph network represents the relationship between the N historical work order texts and the technical object. Based on the target object, the recommended technology object corresponding to the work order text is determined.

2. The method according to claim 1, characterized in that, The step of converting the work order text into a target semantic vector includes: The work order text is segmented into words to obtain M words, where M is a positive integer; The M words are input into a preset text embedding model to obtain the initial vector of the work order text; Based on the preset word frequency weights, the weights corresponding to each of the M words are determined; The target semantic vector is determined based on the weights of the M words and the initial vector of the work order text.

3. The method according to claim 1, characterized in that, Also includes: Obtain the N historical work order texts and the work order information corresponding to each of the N historical work order texts, wherein the work order information includes a technical object; Based on a preset text embedding model, the N historical work order texts are transformed into semantic vectors corresponding to each of the N historical work order texts. Based on the semantic vectors corresponding to each of the N historical work order texts and the work order information corresponding to each of the N historical work order texts, nodes and edges are determined, wherein the nodes include work order nodes and technician nodes; The collaborative graph network is constructed based on the nodes and edges.

4. The method according to claim 1, characterized in that, The step of determining the recommended technology object corresponding to the work order text based on the target object includes: When there are L target objects, obtain the activity level and number of work orders corresponding to each of the L target objects, where L is a positive integer greater than 1; Based on the activity level and number of work orders corresponding to each of the L target objects, the L target objects are sorted to determine the target object sequence; Based on the target object sequence, the recommended technology object is determined.

5. The method according to any one of claims 1 to 4, characterized in that, Also includes: Receive the target solution based on feedback from a preset account from the recommended technology object; A detection operation is performed on the target solution to obtain a detection result, wherein the detection operation includes detecting whether there are syntax errors in the target solution and detecting whether there are security injection vulnerabilities in the target solution; If the detection result is a pass, the target solution will be sent to the target account.

6. The method according to claim 5, characterized in that, Also includes: Receive the processing result returned by the target account after receiving the target solution; Based on the processing results, the collaborative graph network is adjusted.

7. A device for recommending technical objects, characterized in that, include: The receiving module is used to receive work order texts sent by the target account; The conversion module is used to convert the work order text into a target semantic vector; The calculation module is used to calculate the similarity between the target semantic vector and the semantic vectors corresponding to each of the N preset historical work order texts, and to determine the similarity between each of the N historical work order texts, where N is a positive integer; The selection module is used to select historical work order texts with a similarity exceeding a preset threshold as target historical work order texts. The first determining module is used to select the technical object corresponding to the target historical work order text based on a preset collaboration graph network, and determine the target object, wherein the collaboration graph network represents the relationship between the N historical work order texts and the technical object; The second determining module is used to determine the recommended technology object corresponding to the work order text based on the target object.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the recommended method of any one of claims 1 to 6.

9. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method for recommending the technical object as described in any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the technical object recommendation method according to any one of claims 1 to 6.

Images