Server structure data pushing method, system, device and medium

Abstract

The application provides a server structure data pushing method, system, device and medium, comprising: in response to detecting user input search question data, analyzing the search question data to obtain search key features, the search question data comprising image data and / or text data; based on the obtained search key features, searching the pre-built server structure knowledge base for server structure data to be processed matching the search key features; obtaining a user portrait matched by the user, and performing personalized processing on the server structure data to be processed based on the user portrait to generate target server structure data and push the target server structure data. The application realizes pushing of server structure data with high matching degree to the user, improves user experience; and further based on the pre-built comprehensive knowledge base, the accuracy and comprehensiveness of the pushed data are improved.

Description

Technical Field

[0001] This application relates to the field of server technology, and in particular to a server structure data push method, system, device and medium. Background Technology

[0002] Server architecture is a core component of a computer system, and its design and configuration directly affect the server's performance, reliability, and scalability. To better understand server architecture, existing technologies typically provide users with queryable server architecture data by building server architecture knowledge bases.

[0003] However, existing data push methods based on server-structured knowledge bases still have problems such as incomplete push content and inconsistencies with user search results.

[0004] Therefore, a highly accurate server-structured data push method is urgently needed to solve the above-mentioned technical problems. Summary of the Invention

[0005] Therefore, it is necessary to provide a server-structured data push method and system to address the aforementioned technical problems.

[0006] Equipment and media to solve the above-mentioned technical problems.

[0007] Firstly, this application provides a method for pushing server structure data, the method comprising:

[0008] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0009] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0010] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0011] In some embodiments, the method further includes building the server structure knowledge base:

[0012] Obtain existing server-related data and preprocess the existing server-related data to generate server structure block data guided by server structure keywords;

[0013] Based on the server structure block data, a server component graph model containing the server component structure and the server component text vector are generated.

[0014] Based on the preset embedding model and server component text vectors, generate server structure text vectors;

[0015] Based on the region convolutional neural network, the graph convolutional network, and the server component graph model, a server structure graph vector is generated.

[0016] A server structure knowledge base is constructed based on the server structure text vector and the server structure graph vector.

[0017] In some embodiments, generating a server component graph model containing the server component structure and the server component text vector based on the server structure block data includes:

[0018] Parse the server structure relationships contained in the server structure block data;

[0019] Based on the parsed server structure relationships and graph database, a server component graph model is constructed;

[0020] Based on the server structure relationship and text embedding technology, obtain the text vectors of the server components;

[0021] A server structure vector database is constructed based on the server component graph model and the server component text vectors.

[0022] In some embodiments, before parsing the retrieval question data to obtain key retrieval features, the method further includes:

[0023] The search question data is parsed to classify the search question data into search levels, which include a first level and a second level.

[0024] In response to the detection that the search problem data is of the first level, a preset first database is invoked to obtain server structure data to be processed that matches the search problem data;

[0025] In response to the detection that the search question data is at the second level, the search question data is parsed to obtain key search features;

[0026] The first database contains multiple first search questions and associated server structure data stored in advance.

[0027] The step of parsing the search question data to classify the search question data into search levels includes:

[0028] Compare the search question data with multiple first search questions contained in the first database;

[0029] If a first search question exists in the first database that matches the search question data, then the search question data is determined to be of the first level;

[0030] If there is no first search question in the first database that matches the search question data, then the search question data is determined to be of the second level.

[0031] In some embodiments, the key retrieval features include text retrieval features and / or image retrieval features, and parsing the retrieval question data to obtain the key retrieval features includes:

[0032] Based on a pre-defined large language model, the text data in the retrieval question data is parsed to obtain text retrieval features;

[0033] Based on a preset image processing module, the image data in the retrieval question data is parsed to obtain image retrieval features.

[0034] In some embodiments, the step of searching for server structure data to be processed that matches the obtained key retrieval features in a pre-built server structure knowledge base includes:

[0035] In response to the fact that the key retrieval feature is an image retrieval feature, an image retrieval vector matching the image retrieval feature is generated;

[0036] Based on the image retrieval vector, search the server structure knowledge base for a target server structure graph vector that matches the image retrieval vector.

[0037] Calculate the graph-text similarity between the target server structure graph vector and multiple server structure text vectors contained in the server structure knowledge base, and determine the target server structure text vector based on the graph-text similarity.

[0038] Generate server structure data to be processed based on the target server structure text vector and the target server structure graph vector;

[0039] The step of searching for server structure data to be processed that matches the obtained key retrieval features in a pre-built server structure knowledge base further includes:

[0040] In response to the fact that the key retrieval feature is a text retrieval feature, a text retrieval vector matching the image retrieval feature is generated;

[0041] Based on the text retrieval vector, search the server structure knowledge base for a target server structure text vector that matches the text retrieval vector.

[0042] Calculate the text-image similarity between the target server structure text vector and multiple server structure graph vectors contained in the server structure knowledge base, and determine the target server structure graph vector based on the text-image similarity.

[0043] The server structure data to be processed is generated based on the target server structure text vector and the target server structure graph vector.

[0044] In some embodiments, personalizing the server structure data to be processed based on the user profile to generate target server structure data and pushing it includes:

[0045] Based on the user profile, server push restrictions are determined and target prompt word templates matching the user profile are retrieved;

[0046] The server structure data to be processed is filtered based on the server push restrictions to generate transition server structure data;

[0047] Based on the target prompt word template, adjust the transition server structure data to generate the target server structure data and push it.

[0048] Secondly, this application provides a server-structured data push system, the system comprising:

[0049] The processing module is configured to, in response to the detection of user input search question data, parse the search question data to obtain key search features, wherein the search question data includes image data and / or text data;

[0050] The retrieval module is used to search for server structure data to be processed that matches the obtained key retrieval features in a pre-built server structure knowledge base.

[0051] The push module is used to obtain the user profile matched by the user, and to perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0052] Thirdly, this application provides an electronic device, which includes:

[0053] One or more processors;

[0054] and memory associated with one or more processors, the memory being used to store program instructions, which, when read and executed by one or more processors, perform the following operations:

[0055] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0056] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0057] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0058] Fourthly, this application also provides a computer-readable storage medium storing a computer program that causes a computer to perform the following operations:

[0059] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0060] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0061] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0062] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the following method:

[0063] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0064] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0065] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0066] The beneficial effects achieved by this application are as follows:

[0067] This application provides a method for pushing server structure data, including: responding to the detection of user-inputted search question data; parsing the search question data to obtain key search features, wherein the search question data includes image data and / or text data; based on the obtained key search features, searching for server structure data to be processed that matches the key search features in a pre-built server structure knowledge base; obtaining a user profile matching the user; and performing personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it. This method achieves the push of server structure data with a high degree of matching with the user, improving the user experience; furthermore, based on a comprehensive pre-established knowledge base, it improves the accuracy and comprehensiveness of the pushed data. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0069] Figure 1 This is a schematic diagram of a server structure data push method provided in an embodiment of this application;

[0070] Figure 2 This is a schematic diagram of raw data acquisition provided in an embodiment of this application;

[0071] Figure 3 This is a text segmentation diagram provided in an embodiment of this application;

[0072] Figure 4 This is a schematic diagram of a vector database generation provided in an embodiment of this application;

[0073] Figure 5 This is a schematic diagram of a server structure knowledge base generation provided in an embodiment of this application;

[0074] Figure 6 This is a server-structured data push system architecture diagram provided in an embodiment of this application;

[0075] Figure 7 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0077] It should be understood that, in the description of this application, unless the context explicitly requires it, the words "comprising," "including," and similar terms throughout the specification and claims should be interpreted as encompassing rather than being exclusive or exhaustive; that is, meaning "including but not limited to."

[0078] It should also be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0079] It should be noted that the terms "S1," "S2," etc., are used only for descriptive purposes and do not specifically refer to the order or sequence, nor are they intended to limit this application. They are merely for the convenience of describing the method of this application and should not be construed as indicating the sequential order of the steps. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0080] Example 1

[0081] This application provides a method for pushing server architecture data, applied in an information push system. This information push system integrates a server architecture knowledge base. Specifically, it implements the function of pushing server architecture data based on the server architecture knowledge base. Figure 1 As shown, it includes the following:

[0082] S1. In response to detecting user input search question data, parse the search question data to obtain key search features.

[0083] In a specific implementation scenario, before parsing the search question data to obtain key search features, this application embodiment further proposes: parsing the search question data to classify it into search levels, including a first level and a second level; in response to detecting that the search question data is at the first level, calling a preset first database to obtain server structure data to be processed that matches the search question data; in response to detecting that the search question data is at the second level, parsing the search question data to obtain key search features; wherein, the first database pre-stores multiple first search questions and associated server structure data; wherein, parsing the search question data to classify it into search levels includes: comparing the search question data with multiple first search questions contained in the first database; if there is a first search question in the first database that matches the search question data, then the search question data is determined to be at the first level; if there is no first search question in the first database that matches the search question data, then the search question data is determined to be at the second level. After the search question data input by the user is classified into the first level, the first database is called for data feedback; after being classified into the second level, the key search features contained therein are extracted for subsequent retrieval. By pre-classifying the search query data, when the search query is simple and a corresponding search query exists in the historical data, the first database can be directly called to provide the server structure data to be processed corresponding to the search query data. That is, after returning the server structure data with a high degree of matching with the search query data from the first database, further personalized processing is performed to improve the accuracy and relevance of the server structure data finally fed back to the user, and also speeds up the response speed to the search query input by the user, while maximizing the solution of computing resources.

[0084] Specifically, the aforementioned first database pre-stores a first search question and server structure data associated with that search question. The first search question can be a simple question about server structure defined by someone skilled in the art, or it can be a historical question that has already been searched and has uniquely corresponding server structure data. The specific method for comparing the search question data with multiple first search questions contained in the first database to determine if there is a matching first search question includes converting the search question data and the first search question into vectors and then calculating the cosine similarity between the two vectors using cosine similarity. If the cosine similarity reaches a first preset threshold, the search question data is considered similar to the first search question. This first preset threshold can be set by someone skilled in the art based on actual scenarios and experience; this application does not impose any restrictions on it. The specific method for determining whether a first search question matches the search question data can also be achieved by converting the search question data and the first search question into sets containing words, calculating the Jaccard similarity between these two sets (i.e., the ratio of the size of the intersection of the two sets to the size of the ice machine), and determining that the search question data is similar to the first search question when the Jaccard similarity reaches a second preset threshold. The second preset threshold is set by those skilled in the art based on actual scenarios and experience, and this application does not impose any constraints on it. It is understood that the specific method for determining whether a first search question matches the search question data can also be achieved by converting the search question data and the first search question into sequences and calculating the Levenshtein Distance between the two sequences; it can also be achieved by calculating the similarity between the first search question and the search question data based on a word embedding method; or it can be achieved by comparing the hash values ​​between the first search question and the search question data based on a hash method. Specific details will not be elaborated further. It is understood that this application can choose an appropriate determination method based on the actual scenario, and this application does not impose any constraints on the specific determination method.

[0085] It is understood that this application supports image-text retrieval; therefore, the aforementioned retrieval question data includes image data and / or text data. Specifically, obtaining key retrieval features from the retrieval question data includes: parsing the text data in the retrieval question data based on a pre-set large language model to obtain text retrieval features; and parsing the image data in the retrieval question data based on a pre-set image processing module to obtain image retrieval features. It is understood that this application uses a pre-trained large language model to parse the user-input retrieval question data to obtain keywords with retrieval intent and keywords related to the server structure. Based on the keywords obtained from the above parsing, text retrieval features are generated. The training process of the large language model is a conventional technique in this field and will not be elaborated upon here. This application also uses a pre-set image processing module to extract features from the input image. The image processing module performs preprocessing on the input image data, such as image cropping and resizing, grayscale conversion, and normalization, and then automatically extracts image features based on techniques such as SIFT (Scale Invariant Feature Transform), SURF (Speed-Up Robust Feature Transform), HOG (Histogram of Oriented Gradients), and Convolutional Neural Networks (CNN) to generate image retrieval features.

[0086] Furthermore, this application proposes further processing of the aforementioned image retrieval features: obtaining the generated image retrieval features and performing dimensionality reduction on them; calculating the matching degree between the dimensionality-reduced image retrieval features and the server structure, and selecting image retrieval features whose matching degree is greater than or equal to a third preset threshold as target image retrieval features; using the selected target image retrieval features as features for subsequent retrieval in the knowledge base. Specifically, the dimensionality reduction of the image retrieval features includes using methods such as principal component analysis and singular value decomposition to reduce the dimensionality of the feature vectors and lower the subsequent computational complexity. Based on the correlation between the image retrieval features and the server structure, this application further selects characteristic retrieval features with high matching degree with the server structure to further improve the accuracy and efficiency of retrieval in the knowledge base using image data.

[0087] S2. Based on the obtained key search features, search for server structure data to be processed that match the key search features in the pre-built server structure knowledge base.

[0088] In specific implementation scenarios, the above-mentioned search for server structure data matching the acquired key retrieval features in a pre-built server structure knowledge base includes:

[0089] In response to the retrieval key feature being an image retrieval feature, an image retrieval vector matching the image retrieval feature is generated; based on the image retrieval vector, a target server structure graph vector matching the image retrieval vector is searched in the server structure knowledge base; the graph-text similarity between the target server structure graph vector and multiple server structure text vectors contained in the server structure knowledge base is calculated, and the target server structure text vector is determined based on the graph-text similarity; the server structure data to be processed is generated based on the target server structure text vector and the target server structure graph vector.

[0090] Based on the acquired key retrieval features, the process involves searching for server structure data matching the key retrieval features in a pre-built server structure knowledge base. This includes: generating a text retrieval vector matching the image retrieval features in response to the key retrieval features being text retrieval features; searching for a target server structure text vector matching the text retrieval vector in the server structure knowledge base based on the text retrieval vector; calculating the image-text similarity between the target server structure text vector and multiple server structure graph vectors contained in the server structure knowledge base, and determining the target server structure graph vector based on the image-text similarity; and generating server structure data to be processed based on the target server structure text vector and the target server structure graph vector.

[0091] The formula for calculating image-text similarity is as follows: Where T represents the server structure text vector, I is the image feature vector, and Sim(T,I) represents the cosine similarity between the text and the image. Utilizing image-text similarity calculation technology, matching and searching between text and image are achieved by comparing their vector representations. Users can perform image-related searches by inputting a text description of the server structure data, or by providing a server structure image to find related text descriptions; this enables the push of multimodal server structure data, improving the user experience.

[0092] The process of building the pre-built server structure knowledge base includes the following: acquiring existing server-related data and preprocessing it to generate server structure block data guided by server structure keywords; generating server component graph models and server component text vectors containing server component structures based on the server structure block data; generating server structure text vectors based on a preset embedding model and server component text vectors; generating server structure graph vectors based on region convolutional neural networks, graph convolutional networks, and server component graph models; and constructing the server structure knowledge base based on the server structure text vectors and server structure graph vectors. This application achieves automated extraction of server structure data contained in existing server-related data, including server structure graph vectors and server structure text vectors, and constructs a knowledge base based on the extracted server structure data, reducing manual input, expanding the coverage of the knowledge base, and further improving the accuracy and comprehensiveness of server structure queries for users in the future.

[0093] Specifically, the above-mentioned process involves acquiring existing server-related data and preprocessing it to generate server structure block data guided by server structure keywords, including two parts: data cleaning and text segmentation.

[0094] like Figure 2 As shown, the data cleaning process described above includes: collecting all raw data related to existing servers. This raw data comes from sources including, but is not limited to, existing data in the databases of major server-related companies, offline data collection, and data downloaded from publicly available websites. This raw data includes any documents and materials related to server architecture, such as technical documents, papers, blogs, patents, and other formats from server manufacturers, as well as multimodal data such as tables, images, audio, and video. These documents and materials are sufficient to cover all knowledge regarding server configurations, network topologies, storage architectures, operating systems, and performance metrics in the current scenario. Data cleaning operations are then performed on the collected raw data related to existing servers, including server data standardization, server text cleaning, and server domain analysis, to generate existing server-related data. These operations reduce noise and redundant information in the raw data and extract core content for subsequent processing, ensuring the consistency and usability of the generated existing server-related data.

[0095] like Figure 3As shown, the above text segmentation process includes: pre-determining a list of keywords related to the server structure, including names of basic server structures such as "chassis," "CPU," "memory," "hard drive," "power supply," "cable," "rail," "label," "external card," "motherboard," "backplate," "fan," and "heat sink." Based on the keyword list, paragraphs containing these keywords in the existing server-related data are searched. Specifically, this can be implemented using a programming language or a keyword extraction algorithm, such as the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm. This application does not impose any restrictions on the specific keyword extraction algorithm. Then, these paragraphs containing the above keywords are segmented to generate server structure block data and stored. For example, this application discloses code that uses the Python programming language to segment existing server structure data and store the segmented data:

[0096] import re

[0097] keywords = ["Case", "CPU", "Memory", "Hard Drive", "Power Supply", "Cables", "Guides", "Labels", "External Cards", "Motherboard", "Backplate", "Fan", "Heater"]

[0098] paragraph_pattern=r'(?:(?<!\n)\n{2,})(.*?)(?=\n{2,}|$)'

[0099] text_data="""

[0100] This is the first paragraph of text, which does not contain keywords.

[0101] This is the second text, which contains the keywords CPU and memory. The CPU is the core component of a computer, and memory is used to store data.

[0102] The third paragraph also lacks keywords.

[0103] The fourth paragraph contains the keywords "hard drive" and "computer case." A hard drive is used to store data, while the computer case is the skeleton of the computer.

[0104] This is the last paragraph and does not contain keywords.

[0105] """

[0106] paragraphs=re.findall(paragraph_pattern,text_data,re.DOTALL)

[0107] filtered_paragraphs = []

[0108] forparagraph inparagraphs:

[0109] cleaned_paragraph=paragraph.strip()

[0110] ifany(keyword in cleaned_paragraph forkeyword inkeywords):

[0111] filtered_paragraphs.append(cleaned_paragraph)

[0112] for index,paragraph in enumerate(filtered_paragraphs,start=1):

[0113] print(f"paragraph{index}:\n{paragraph}\n")

[0114] The code described above uses regular expressions or string matching functions to search for paragraphs containing these keywords in existing server-related data. If the existing server-related data is already organized by paragraphs, it can directly segment paragraphs by recognizing paragraph markers such as newlines, blank lines, or specific markers. If the existing server-related data is not organized by paragraphs, natural language processing tools are used to identify sentence boundaries and combine these sentences into logically related paragraphs. This server-structure keyword-oriented paragraph segmentation, compared to segmentation by character count and segmentation by semantics, avoids the shortcomings of character count segmentation, such as semantic breaks, information loss, and poor flexibility, while also avoiding the computational complexity, context dependence, and difficulty in precise control of semantic segmentation. This results in server-structured block data with higher semantic coherence, strong adaptability, and ease of processing. Furthermore, because the paragraph segmentation is based on a list of relevant keywords specific to the server structure, it greatly improves the directionality and accuracy of the text segmentation results; subsequently, it helps improve retrieval efficiency and, to some extent, compensates for insufficient retrieval accuracy.

[0115] In specific implementation scenarios, such as Figure 4As shown, the above-mentioned generation of server component graph models and server component text vectors based on server structure block data includes: parsing the server structure relationships contained in the server structure block data; constructing a server component graph model based on the parsed server structure relationships and graph database; obtaining server component text vectors based on server structure relationships and text embedding technology; and constructing a server structure vector database based on the server component graph model and server component text vectors.

[0116] Specifically, after dividing existing server-related data into data blocks to generate server structure block data, relation extraction techniques are used to analyze the sentences in the server structure block data and extract the server structure relationships contained therein, such as "CPU is installed in the chassis" and "memory is connected to the motherboard." Then, based on a graph database, a server component graph model is constructed using the obtained server structure relationships. In the graph, each server component can be a node, and the relationships between components are edges. The attributes of the edges can contain detailed information about the relationships, such as "installed" and "connected." Then, text embedding techniques are used to associate a text vector with each node or variable in the graph, thereby converting the server structure block data into text vector representations of server component text vectors. The aforementioned text embedding techniques include, but are not limited to, the bag-of-words model, Word2Vec (Word to Vector, a group of related models used to generate word vectors), GloVe (Global Vectors for Word Representation, a word representation tool based on global word frequency statistics), FastText (a model for learning word embeddings and text classification), and BERT (Bidirectional Encoder Representations from Transformers). This application reduces storage pressure by vectorizing server block data and provides a foundation for subsequent image and text fusion push during retrieval.

[0117] Specifically, based on a pre-defined embedding model and server component text vectors, server structure text vectors are generated. This involves using the BERT model with the formula Embedding(X) = BERT(X) to calculate the vector representation of the text data contained in the aforementioned server structure block data. Here, X represents the server structure text in the server block data, and BERT(X) represents the text embedding vector generated by the BERT model. In BERT, the Tokenembedding layer converts each word into a fixed-dimensional vector; Segmentembeddings handle the classification task of input sentence pairs; and PositionEmbeddings introduce positional information in the sequence, facilitating the knowledge base's understanding of the sequential relationships of words or characters within the sequence. It is understood that other text embeddings besides the BERT model can also be used to achieve the above technical effects, and this application does not limit this.

[0118] This process involves generating server structure graph vectors based on region convolutional neural networks, graph convolutional networks, and server component graph models. A server structure knowledge base is then constructed based on server structure text vectors and server structure graph vectors. Specifically, this includes: first, using FastR-CNN to further process the obtained server component graph models, generating bounding boxes and category labels for these server components. The output of this step is the component regions in the image and their preliminary feature representations. FastR-CNN generates candidate regions through a Region Proposal Network (RPN) and extracts features from these regions using a convolutional neural network. R = R-CNN(Image); where V R These are the feature vectors of each detected region in the image. FastR-CNN feature extraction formula: f i =RoIAlign(x,r i ), where f i Let x be the feature vector of the i-th region, and r be the input image. i This is the i-th detected region. These features are further input into a Graph Convolutional Neural Network (GCN), which constructs a graph structure from the components detected by Fast R-CNN and their relationships. GCN is used to learn the relationships between nodes (components) in the graph. Through GCN, the interrelationships between components in the image are modeled, representing the image as a graph structure and generating a global feature vector of the image. This enables the generation of a global graph vector corresponding to the entire server based on the previously generated server component graph model. GCN generates a graph structure from the feature map extracted by Fast R-CNN, where nodes represent server components and edges represent the relationships between components. Then, graph convolution operations are performed through GCN to obtain the overall vector representation V of the image. G V G =GCN(VR (E). Here, E represents the edges in the graph structure, indicating the relationships between components. GCN graph convolution formula: in, It is the adjacency matrix of the graph. It is the node degree matrix, H (l) W is the feature matrix of the nodes in the l-th layer. (l) σ is a trainable weight matrix, and σ is the activation function.

[0119] Furthermore, to improve the user experience of server structure data pushed by the server structure knowledge base, a prompt template can be pre-written to guide the server structure knowledge base in generating output that meets specific requirements. The prompt template is written by those skilled in the art based on actual needs, and the specific writing method is a conventional technique in the field, which will not be elaborated upon here. The following is a specific prompt template, provided for illustrative purposes only and not as a constraint; the actual prompt template does not need to include all the content in the following template:

[0120] Introduction: Briefly introduce the basic concepts of server architecture, including its main components and their roles in the server. Server Components: Describe the core components of the server, such as CPU, memory, motherboard, storage devices (e.g., hard drives, SSDs), power supply, cooling system, expansion cards (e.g., graphics cards, network cards), etc. For each component, explain its function, importance, and role in server operation. Component Relationships and Connections: Describe the connection methods and relationships between server components, such as how the motherboard connects to the CPU, memory, and expansion cards, and how the power supply provides power to each component. Emphasize the compatibility and cooperation between components, and their impact on overall server performance. Server Types and Architectures: Introduce different types of servers (e.g., tower servers, rack servers, blade servers), their applicable scenarios, advantages, and disadvantages. Describe server architectures (e.g., Symmetric Multiprocessing (SMP), Asymmetric Multiprocessing (AMP), distributed architecture, etc.), and explain the characteristics and applicable situations of each architecture. Server Configuration and Optimization: Explain how to select appropriate server configurations based on requirements, including CPU model, memory capacity, storage capacity, etc. This section discusses how to optimize servers to improve performance, such as adjusting BIOS settings, updating drivers, and optimizing system configuration. Troubleshooting and maintenance describes common server failure types (such as hardware failures, software failures, and network failures) and corresponding troubleshooting methods. It explains how to perform daily server maintenance and management to ensure stable operation and extend its lifespan. Security and backup discuss the importance of server security and how to protect servers from attacks and data breaches. It introduces the importance of data backup and common backup methods and strategies. The summary and outlook summarize the main content of the server architecture knowledge base and emphasize the importance of servers in modern information technology. It also looks ahead to future trends in server technology and possible new technologies. The user guide and interaction provide guidance on how to use this knowledge base, including how to query information and understand the relationships between components. Users are encouraged to provide feedback and suggestions to continuously improve and optimize the content and functionality of the knowledge base. It is understood that the final data pushed by the knowledge base will not only contain server architecture data but may also include server-related content from the above templates.

[0121] like Figure 5As shown, this application utilizes existing server-related data, a vector database, and a prompt template to build and train a server structure knowledge base. This knowledge base can be understood as a model for retrieving server structure data based on user-input retrieval question data. Because the content related to server structure iterates rapidly, this application also proposes setting a regular update schedule, such as monthly or quarterly updates, and establishing a feedback system so that users or internal teams can report any errors, omissions, or content requiring updates. After collecting updated content, the above process is repeated, using a new dataset to update the server structure knowledge base.

[0122] S3. Obtain the user profile matching the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate the target server structure data and push it.

[0123] Specifically, the process involves determining server push restrictions based on user profiles and retrieving target prompt word templates that match those profiles; filtering the server structure data to be processed based on these restrictions to generate transitional server structure data; and adjusting the transitional server structure data based on the target prompt word templates to generate the target server structure data, which is then pushed. The server push restrictions determined by the user profiles include user needs, user industry, and service structure application scenarios. The target prompt word templates are personalized output templates pre-written based on the user profiles. It can be understood that the server structure knowledge base constructed above targets different user groups, such as sales and R&D. By further filtering the server structure data to be processed based on the server push restrictions to obtain server structure data with high user compatibility, and then using the target prompt word templates that match the user profiles for personalized pushes, the quality of the knowledge base's pushed content can be significantly improved, further enhancing the user experience.

[0124] Example 2

[0125] Corresponding to the above embodiment one, as Figure 6 As shown in the illustration, this application also provides a server-structured data push system, including:

[0126] Processing module 610 is configured to, in response to detecting user input search question data, parse the search question data to obtain key search features, wherein the search question data includes image data and / or text data;

[0127] The retrieval module 620 is used to search for server structure data to be processed that matches the retrieval key features in a pre-built server structure knowledge base based on the obtained retrieval key features.

[0128] The push module 630 is used to obtain the user profile matched by the user, and to perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0129] In some embodiments, the system further includes a preparation module (not shown in the figure) for building the server structure knowledge base: acquiring existing server-related data and preprocessing the existing server-related data to generate server structure block data guided by server structure keywords; generating a server component graph model containing server component structures and server component text vectors based on the server structure block data; generating server structure text vectors based on a preset embedding model and server component text vectors; generating server structure graph vectors based on a region convolutional neural network, a graph convolutional network, and the server component graph model; and constructing a server structure knowledge base based on the server structure text vectors and the server structure graph vectors.

[0130] In some embodiments, the preparation module is further configured to parse the server structure relationships contained in the server structure block data; construct a server component graph model based on the parsed server structure relationships and the graph database; obtain server component text vectors based on the server structure relationships and text embedding technology; and construct a server structure vector database based on the server component graph model and the server component text vectors.

[0131] In some embodiments, before parsing the search question data to obtain key search features, the processing module 610 is further configured to parse the search question data to classify the search question data into search levels, the search levels including a first level and a second level; in response to detecting that the search question data is of the first level, a preset first database is invoked to obtain server structure data to be processed that matches the search question data; in response to detecting that the search question data is of the second level, the search question data is parsed to obtain key search features; wherein, the first database pre-stores multiple first search questions and associated server structure data; parsing the search question data to classify the search question data into search levels includes: comparing the search question data with multiple first search questions contained in the first database; if there is a first search question in the first database that matches the search question data, the search question data is determined to be of the first level; if there is no first search question in the first database that matches the search question data, the search question data is determined to be of the second level.

[0132] In some embodiments, the processing module 610 is further configured to parse the text data in the retrieval question data based on a preset large language model to obtain text retrieval features; and to parse the image data in the retrieval question data based on a preset image processing module to obtain image retrieval features.

[0133] In some embodiments, the retrieval module 620 is further configured to: generate an image retrieval vector matching the image retrieval feature in response to the retrieval key feature being an image retrieval feature; search for a target server structure graph vector matching the image retrieval vector in the server structure knowledge base based on the image retrieval vector; calculate the image-text similarity between the target server structure graph vector and multiple server structure text vectors contained in the server structure knowledge base, and determine the target server structure text vector based on the image-text similarity; generate server structure data to be processed based on the target server structure text vector and the target server structure graph vector; and generate server structure data to be processed based on the target server structure text vector and the target server structure graph vector.

[0134] In some embodiments, the push module 630 is further configured to determine server push restrictions based on the user profile and retrieve a target prompt word template that matches the user profile; filter the server structure data to be processed based on the server push restrictions to generate transition server structure data; and adjust the transition server structure data based on the target prompt word template to generate target server structure data and push it.

[0135] Example 3

[0136] Corresponding to all the above embodiments, this application provides an electronic device, including: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions, which, when read and executed by the one or more processors, perform the following operations:

[0137] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0138] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0139] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0140] in, Figure 7 An exemplary architecture of an electronic device is shown, which may include a processor 710, a video display adapter 711, a disk drive 712, an input / output interface 713, a network interface 714, and a memory 720. The processor 710, video display adapter 711, disk drive 712, input / output interface 713, network interface 714, and memory 720 can communicate with each other via a bus 730.

[0141] The processor 710 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to achieve the technical solution provided in this application.

[0142] The memory 720 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 720 can store the operating system 721 for controlling the execution of the electronic device 700, and the basic input / output system (BIOS) 722 for controlling the low-level operations of the electronic device 700. Additionally, it can store a web browser 723, a data storage management system 724, and an icon font processing system 725, etc. The aforementioned icon font processing system 725 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when the technical solution provided in this application is implemented through software or firmware, the relevant program code is stored in the memory 720 and is called and executed by the processor 710.

[0143] Input / output interface 713 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0144] Network interface 714 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0145] Bus 730 includes a pathway for transmitting information between various components of the device, such as processor 710, video display adapter 711, disk drive 712, input / output interface 713, network interface 714, and memory 720.

[0146] In addition, the electronic device 700 can also obtain information on specific claim conditions from the virtual resource object claim condition information database for use in condition judgment, etc.

[0147] It should be noted that although the above-described device only shows the processor 710, video display adapter 711, disk drive 712, input / output interface 713, network interface 714, memory 720, bus 730, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0148] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, a cloud server, or a network device, etc.) to execute the methods of various embodiments or some parts of the embodiments of this application.

[0149] Example 4

[0150] Corresponding to all the above embodiments, this application also provides a computer-readable storage medium, characterized in that it stores a computer program that causes a computer to perform the following operations:

[0151] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0152] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0153] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0154] Example 5

[0155] Corresponding to all the above embodiments, this application also provides a computer program product, which, when executed by a processor, implements the steps of the following method:

[0156] In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data;

[0157] Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base;

[0158] Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

[0159] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0160] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for pushing data across a server structure, characterized in that, The method includes: In response to detecting user-input search question data, the search question data is parsed to obtain key search features, the search question data including image data and / or text data; Based on the obtained key retrieval features, search for server structure data to be processed that matches the key retrieval features in the pre-built server structure knowledge base; Obtain the user profile matched by the user, and perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it. The step of searching for server structure data matching the acquired key retrieval features in a pre-built server structure knowledge base includes: In response to the fact that the key retrieval feature is an image retrieval feature, an image retrieval vector matching the image retrieval feature is generated; Based on the image retrieval vector, search the server structure knowledge base for a target server structure graph vector that matches the image retrieval vector. Calculate the graph-text similarity between the target server structure graph vector and multiple server structure text vectors contained in the server structure knowledge base, and determine the target server structure text vector based on the graph-text similarity. Generate server structure data to be processed based on the target server structure text vector and the target server structure graph vector; In response to the fact that the key retrieval feature is a text retrieval feature, a text retrieval vector matching the image retrieval feature is generated; Based on the text retrieval vector, search the server structure knowledge base for a target server structure text vector that matches the text retrieval vector. Calculate the text-image similarity between the target server structure text vector and multiple server structure graph vectors contained in the server structure knowledge base, and determine the target server structure graph vector based on the text-image similarity. Generate server structure data to be processed based on the target server structure text vector and the target server structure graph vector; The method further includes: acquiring the image retrieval features; performing dimensionality reduction processing on the image retrieval features; calculating the matching degree between the multiple image retrieval features after dimensionality reduction and the server structure; selecting image retrieval features whose matching degree is greater than or equal to the matching threshold as target image retrieval features; and using the selected target image retrieval features as features for subsequent retrieval in the knowledge base.

2. The method according to claim 1, characterized in that, The method also includes building the server structure knowledge base: Obtain existing server-related data and preprocess the existing server-related data to generate server structure block data guided by server structure keywords; Based on the server structure block data, a server component graph model containing the server component structure and server component text vectors are generated. Based on the preset embedding model and server component text vectors, generate server structure text vectors; Based on the region convolutional neural network, the graph convolutional network, and the server component graph model, a server structure graph vector is generated. A server structure knowledge base is constructed based on the server structure text vector and the server structure graph vector.

3. The method according to claim 2, characterized in that, The step of generating a server component graph model containing the server component structure and the server component text vectors based on the server structure block data includes: Parse the server structure relationships contained in the server structure block data; Based on the parsed server structure relationships and graph database, a server component graph model is constructed; Based on the server structure relationship and text embedding technology, obtain the text vectors of the server components; A server structure vector database is constructed based on the server component graph model and the server component text vectors.

4. The method according to claim 1, characterized in that, Before parsing the retrieval question data to obtain key retrieval features, the method further includes: The search question data is parsed to classify the search question data into search levels, which include a first level and a second level. In response to the detection that the search problem data is of the first level, a preset first database is invoked to obtain server structure data to be processed that matches the search problem data; In response to the detection that the search question data is at the second level, the search question data is parsed to obtain key search features; The first database contains multiple first search questions and associated server structure data stored in advance. The step of parsing the search question data to classify the search question data into search levels includes: Compare the search question data with multiple first search questions contained in the first database; If a first search question exists in the first database that matches the search question data, then the search question data is determined to be of the first level; If there is no first search question in the first database that matches the search question data, then the search question data is determined to be of the second level.

5. The method according to claim 2, characterized in that, The key retrieval features include text retrieval features and / or image retrieval features. Parsing the retrieval question data to obtain the key retrieval features includes: Based on a pre-defined large language model, the text data in the retrieval question data is parsed to obtain text retrieval features; Based on a preset image processing module, the image data in the retrieval question data is parsed to obtain image retrieval features.

6. The method according to claim 1, characterized in that, Personalize the server structure data to be processed based on the user profile to generate target server structure data and push it, including: Based on the user profile, server push restrictions are determined and target prompt word templates matching the user profile are retrieved; The server structure data to be processed is filtered based on the server push restrictions to generate transition server structure data; Based on the target prompt word template, adjust the transition server structure data to generate the target server structure data and push it.

7. A server structure data push system for implementing the server structure data push method according to any one of claims 1 to 6, characterized in that, The system includes: The processing module is configured to, in response to the detection of user input search question data, parse the search question data to obtain key search features, wherein the search question data includes image data and / or text data; The retrieval module is used to search for server structure data to be processed that matches the obtained key retrieval features in a pre-built server structure knowledge base. The push module is used to obtain the user profile matched by the user, and to perform personalized processing on the server structure data to be processed based on the user profile to generate target server structure data and push it.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; And a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, It stores a computer program that causes the computer to perform the method described in any one of claims 1-6.

Images