Text search method and system, electronic equipment, storage medium and product

Abstract

The invention discloses a text search method and system, electronic equipment, a storage medium and a product, and relates to the field of large model technology and artificial intelligence. The method is applied to a server and comprises the steps that a first text sent by a client side and context information associated with the first text are received, the first text is used for representing a text selected in operation pages displayed on the client side, and the context information at least comprises page information of the operation pages and historical operation information of the operation pages; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; searching based on the semantic analysis result and the context information to obtain a search result; and sending the search result to the client. According to the method and the device, the technical problem of relatively low accuracy of a search result during intelligent word segmentation search is solved.

Description

Technical Field

[0001] The present application relates to the fields of large model technology and artificial intelligence, and specifically to a text search method, system, electronic device, storage medium and product. Background Art

[0002] As the functions of the products corresponding to the console become more and more abundant, it becomes gradually difficult to use the console for smart word search. First, as the functions of the products corresponding to the console increase, users need to understand and manage more options and settings. This complexity may cause confusion for new users and non-professional users. Secondly, as the functions continue to increase, the menus and submenus of the console may become larger and deeper, which makes it more difficult for users to find specific settings or functions. Furthermore, more functions and options may lead to a sharp increase in the amount of information on the user interface, making it difficult for users to find the key information they need in the massive information, resulting in lower accuracy of search results when conducting smart word search.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a text search method, system, electronic device, storage medium and product to at least solve the technical problem of low accuracy of search results when performing intelligent word search.

[0005] According to one aspect of an embodiment of the present application, a text search method is provided, which is applied to a server and includes: receiving a first text sent by a client and context information associated with the first text, wherein the first text is used to represent the text selected in an operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; performing a search based on the semantic analysis result and the context information to obtain a search result; and sending the search result to the client.

[0006] According to one aspect of an embodiment of the present application, a text search method is also provided, which is applied to a client, including: upon receiving a search instruction for searching text in an operation page, obtaining a first text corresponding to the search instruction in the operation page, and context information associated with the first text, wherein the context information at least includes page information of the operation page, and historical operation information of the operation page; sending the first text and the context information to a server, and receiving search results returned by the server, wherein the search results are the results obtained by the server based on the semantic analysis results of the first text and the context information; and displaying the search results in the operation page.

[0007] According to one aspect of an embodiment of the present application, a text search method is also provided, which is applied to a client, including: when a search instruction for searching text in a cloud product control page is received, obtaining a first text corresponding to the search instruction in the cloud product control page, and context information associated with the first text, wherein the context information at least includes page information of the cloud product control page, and historical operation information of the cloud product control page; sending the first text and the context information to a cloud product control server, and receiving search results returned by the cloud product control server, wherein the search results are the results obtained by the cloud product control server through a search based on the semantic analysis results of the first text and the context information, and the semantic analysis results are the results obtained by the cloud product control server through semantic analysis of the first text; and displaying the search results in the cloud product control page.

[0008] According to one aspect of an embodiment of the present application, a text search method is also provided, which is applied to a server and includes: obtaining a first text and context information associated with the first text by calling a first interface, wherein the first interface includes a first parameter, and the parameter value of the first parameter includes the first text and the context information, the first text is used to represent the selected text in the operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; performing a search based on the semantic analysis result and the context information to obtain a search result; and outputting the search result by calling a second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the search result.

[0009] According to one aspect of an embodiment of the present application, a text search system is also provided, including: a client, which is used to obtain a first text corresponding to the search instruction in the operation page, and context information associated with the first text when receiving a search instruction for searching the text in the operation page, wherein the context information at least includes page information of the operation page and historical operation information of the operation page; a server, connected to the client, is used to perform semantic analysis on the first text, obtain a semantic analysis result of the first text, and search based on the semantic analysis result and the context information to obtain search results; the client is also used to display the search results in the operation page.

[0010] According to another aspect of an embodiment of the present application, a text search device is also provided, which is applied to a server and includes: a receiving module for receiving a first text sent by a client and context information associated with the first text, wherein the first text is used to represent the selected text in an operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page; an analysis module for performing semantic analysis on the first text to obtain a semantic analysis result of the first text; a search module for performing a search based on the semantic analysis result and the context information to obtain a search result; and a sending module for sending the search result to the client.

[0011] According to another aspect of an embodiment of the present application, a text search device is also provided, which is applied to a client and includes: an acquisition module for acquiring, upon receiving a search instruction for searching text in an operation page, a first text corresponding to the search instruction in the operation page, and context information associated with the first text, wherein the context information at least includes page information of the operation page and historical operation information of the operation page; a sending module for sending the first text and the context information to a server, and receiving search results returned by the server, wherein the search results are the results obtained by the server through a search based on the semantic analysis results of the first text and the context information, and the semantic analysis results are the results obtained by the server through a semantic analysis of the first text; and a display module for displaying the search results in the operation page.

[0012] According to one aspect of an embodiment of the present application, a text search device is also provided, which is applied to a client and includes: an acquisition module for acquiring, upon receiving a search instruction for searching text in a cloud product control page, a first text corresponding to the search instruction in the cloud product control page, and context information associated with the first text, wherein the context information includes at least page information of the cloud product control page and historical operation information of the cloud product control page; a sending module for sending the first text and the context information to a cloud product control server, and receiving search results returned by the cloud product control server, wherein the search results are the results obtained by the cloud product control server through a search based on the semantic analysis results of the first text and the context information, and the semantic analysis results are the results obtained by the cloud product control server through semantic analysis of the first text; a display module for displaying the search results in the cloud product control page.

[0013] According to one aspect of an embodiment of the present application, a text search device is also provided, which is applied to a server and includes: an acquisition module for acquiring a first text and context information associated with the first text by calling a first interface, wherein the first interface includes a first parameter, and the parameter value of the first parameter includes the first text and the context information, the first text is used to represent the selected text in the operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page; an analysis module for performing semantic analysis on the first text to obtain a semantic analysis result of the first text; a search module for performing a search based on the semantic analysis result and the context information to obtain a search result; an output module for outputting the search result by calling a second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the search result.

[0014] According to another aspect of the embodiments of the present application, an electronic device is provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in the various embodiments of the present application when running.

[0015] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present application.

[0016] According to another aspect of the embodiments of the present application, a computer program product is further provided, including a computer program, which implements the methods in various embodiments of the present application when executed by a processor.

[0017] In an embodiment of the present application, a first text sent by a client and context information associated with the first text are received, wherein the first text is used to represent the text selected in the operation page displayed on the client, and the context information at least includes the page information of the operation page and the historical operation information of the operation page; a semantic analysis is performed on the first text to obtain a semantic analysis result of the first text; a search is performed based on the semantic analysis result and the context information to obtain a search result; and the search result is sent to the client. It is easy to notice that the client can provide the first text and the context information associated with the first text, and the server can then perform a semantic analysis on the first text to obtain a semantic analysis result, and then search based on the semantic analysis result combined with the context information to obtain a search result, and send the search result to the client. Since AI intelligent technology is combined in the search process, the user's search intention can be understood more accurately, thereby improving the search accuracy, thereby solving the technical problem of low accuracy of search results when performing intelligent word search.

[0018] It is easy to notice that the above general description and the following detailed description are merely for the purpose of exemplifying and explaining the present application, and do not constitute a limitation of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present application;

[0021] Figure 2 is a flowchart of a text search method according to Example 1 of the present application;

[0022] Figure 3 is a schematic diagram of a text search method according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of a text search method according to Example 2 of the present application;

[0024] Figure 5 is a schematic diagram of a text search method according to Example 3 of the present application;

[0025] Figure 6 is a schematic diagram of a text search method according to Example 4 of the present application;

[0026] Figure 7 is a schematic diagram of a text search system according to Example 5 of the present application;

[0027] Figure 8 is a schematic diagram of a text search device according to Example 6 of the present application;

[0028] Figure 9 is a schematic diagram of a text search device according to Example 7 of the present application;

[0029] Figure 10 is a schematic diagram of a text search device according to Example 8 of the present application;

[0030] Figure 11 is a schematic diagram of a text search device according to Example 9 of the present application;

[0031] Figure 12 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] The technical solution provided in this application is mainly implemented using large-scale model technology. The large model here refers to a deep learning model with large-scale model parameters, which can usually contain hundreds of millions, tens of billions, hundreds of billions, trillions, or even more than ten trillion model parameters. The large model can also be called a foundation model / foundation model. It is pre-trained by using large-scale unlabeled corpus to produce a pre-trained model with more than 100 million parameters. This model can adapt to a wide range of downstream tasks and has good generalization capabilities, such as large-scale language models (LLMs) and multimodal pre-training models.

[0035] It should be noted that when the large model is actually applied, the pre-trained model can be fine-tuned through a small number of samples, so that the large model can be applied to different tasks. For example, the large model can be widely used in natural language processing (NLP), computer vision, semantic processing and other fields. Specifically, it can be applied to computer vision tasks such as visual question answering (VQA), image description (IC), image generation, etc. It can also be widely used in natural language processing tasks such as text-based sentiment classification, text summary generation, and machine translation. Therefore, the main application scenarios of the large model include but are not limited to digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc. In the embodiment of the present application, the data processing through the natural language processing model in the cloud product console scenario is taken as an example for explanation,

[0036] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretations:

[0037] BERT model: Bidirectional Encoder Representations from Transformers can fine-tune the pre-trained BERT model to adapt to specific tasks and data, thereby improving the performance of the model in specific fields and obtaining the final BERT model.

[0038] Transformer model: Generative pre-trained transformers refer to models based on the Transformer architecture in addition to BERT. These models can also be used for semantic understanding, such as the GPT (Generative Pre-trained Transformer) series of models. These models can process long text sequences and are suitable for understanding and generating responses to user consultation questions.

[0039] Seq2Seq model: refers to the sequence-to-sequence model, also known as the Sequence to Sequence model. If the user's consultation question requires a detailed answer, you can consider using the Seq2Seq model. For example, using an encoder-decoder structure, in which the encoder encodes the user's question into a fixed-length representation, and the decoder generates the corresponding answer.

[0040] FastText: Also known as fast text, FastText is a lightweight word embedding model for smaller data or limited computing resources. It can also be used for classification and parsing of consulting questions.

[0041] Example 1

[0042] According to an embodiment of the present application, a text search method is provided. It should be noted that the steps shown in the flowchart of 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 can be executed in an order different from that shown here.

[0043] Considering the huge number of model parameters of large models and the limited computing resources of mobile terminals, the above text search method provided in the embodiment of the present application can be applied to Figure 1 The application scenarios shown are not limited to these. Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present application. Figure 1 In the illustrated application scenario, the large model is deployed on a server 10. The server 10 can be connected to one or more client devices 20 via a local area network, a wide area network, the Internet, or other types of data networks. The client devices 20 herein may include, but are not limited to, smartphones, tablet computers, laptops, PDAs, personal computers, smart home devices, and in-vehicle devices. The client devices 20 can interact with users via a graphical user interface to access the large model and thereby implement the methods provided in the embodiments of the present application.

[0044] In an embodiment of the present application, a system composed of a client device and a server can perform the following steps: the client device executes, sends a first text, and context information associated with the first text; the server executes, receives the first text sent by the client, and context information associated with the first text, wherein the first text is used to represent the text selected in the operation page displayed on the client, and the context information at least includes page information of the operation page, and historical operation information of the operation page; performs semantic analysis on the first text to obtain a semantic analysis result of the first text; performs a search based on the semantic analysis result and the context information to obtain a search result; and sends the search result to the client. It should be noted that, if the operating resources of the client device can meet the deployment and operation conditions of the large model, the embodiment of the present application can be performed in the client device.

[0045] Under the above operating environment, this application provides Figure 2 The text search method shown. Figure 2 Flowchart of the text search method according to Example 1 of the present application. Figure 2 As shown, the method is applied to the server and may include the following steps:

[0046] Step S202: Receive a first text sent by the client, and context information associated with the first text, wherein the first text is used to represent the selected text in the operation page displayed on the client, and the context information at least includes page information of the operation page and historical operation information of the operation page.

[0047] The above-mentioned client can be a web client or a mobile software client. Optionally, this application does not impose any specific restrictions on the type of client.

[0048] The above-mentioned server can be a cloud server console, or other servers. Optionally, there is no specific restriction on the type of server in this application. In this application, the server is a cloud server console as an example for illustration, wherein the cloud server console can include a console portal server (also called Console Portal Server), and a global traffic server (also called Global Traffic Server, abbreviated as GTS Server) and other servers.

[0049] The above-mentioned first text can be the text data selected by the user on the operation interface of the client, wherein the first text may come from multiple channels, such as: user work orders, online consultations, user suggestions and feedback, help documents, community forums, duty records, network interface services and parameter definitions, error code information, resource diagnosis reports, etc. Optionally, the data from the above-mentioned channels can be displayed on the operation interface of the client, and classified and labeled by the user to obtain the first text. For example, the user highlights a certain paragraph of text in the above-mentioned data on the operation interface to determine the first text. Optionally, intelligent word search can be performed based on the first text.

[0050] The above-mentioned context information associated with the first text may include page information on the operation page and historical operation information of the operation page, wherein the page information on the operation page may include page content, that is, based on the page content currently browsed by the user, such as specific error messages, resource identifiers, etc., as well as page content summaries, etc., wherein the historical operation information can be used to analyze the user's operation history, such as frequently visited services, frequently performed operations, etc.

[0051] In an optional embodiment, a client and a server may be connected, and a user may input a first text and context information associated with the first text on the client in any manner, or a user may input a first text on the client in any manner, and the client may match the corresponding context information associated with the first text based on the content of the first text input by the user. For example, the user may highlight a certain text segment on the current page of the client to give the first text, and the user may input the context information associated with the first text in text form, or the user may input the first text in text form on the operation page of the client, and the client may automatically match the context information associated with the first text based on the user's role and authority. Furthermore, the server may obtain the first text and the context information associated with the first text on the client. Optionally, accurate user behavior feature information may be obtained through the first text and the context information associated with the first text, thereby providing the user with accurate search results and suggestions.

[0052] Step S204: performing semantic analysis on the first text to obtain a semantic analysis result of the first text.

[0053] In an optional embodiment, after obtaining the first text and the context information associated with the first text, the first text can be semantically analyzed in any form to determine the user's search intention, that is, to obtain the semantic analysis result of the first text. Optionally, in this application, only the use of a natural language processing model (also known as Natural Language Processing, referred to as NLP model) and a bidirectional encoder representation transformer (also known as Bidirect ional Encoder Representation from Transformers, referred to as BERT model) when performing semantic analysis is used as an example for illustration.

[0054] Optionally, the NLP model can be used to analyze the first text, understand the semantics and intentions of word marking, and complete feature extraction, where the word marking intentions may include querying corresponding resources, explaining technical terms, viewing corresponding operation documents, obtaining possible solutions to error pop-ups, etc. The BERT model can be used to capture contextual information and semantic relationships. It can understand the semantics of the entire sentence corresponding to the first text, and is suitable for parsing complex consulting questions when the consulting questions corresponding to the first text are relatively complex. Optionally, in addition to the BERT model, other models based on the Transformer architecture can also be considered, such as the GPT series of models, which can process long text sequences. Suitable for understanding and generating responses to user consultation questions. If the consultation question corresponding to the first text needs to generate a detailed answer, you can consider using a Seq2Seq model, for example, using an encoder-decoder structure, where the encoder encodes the user question into a fixed-length representation, and the decoder generates the corresponding answer. For smaller-scale data or limited computing resources, you can also use FastText to classify and parse consultation questions. FastText is a lightweight word embedding model. Considering the complexity of the functions controlled by the cloud product console, the characteristics of the data, and the performance of the model, the BERT model can be used to parse and analyze the user's word-marking consultation questions.

[0055] Step S206: Search based on the semantic analysis result and context information to obtain search results.

[0056] In an optional embodiment, after obtaining the semantic analysis results, an intelligent word search can be performed based on the semantic analysis results and contextual information to obtain search results. Optionally, when searching, user roles and permissions, that is, user identity information (user name, user address, etc.), role information (such as administrator, developer) and corresponding permissions (master account, access control microcomputer authorization information, etc.) can be used to ensure that recommended resources and operations are available to users. Optionally, access rights to data on the server, that is, the cloud server console, can be restricted, that is, different user roles have different access rights, and only authorized user roles can access the corresponding data content. Therefore, when searching based on semantic analysis results and contextual information, user roles need to be considered. After determining the user role, the user role can be matched with the permissions corresponding to the user role, and then the corresponding search results can be determined based on the data content area within the permissions corresponding to the user role, that is, to prevent operation suggestions outside the role permissions; optionally, based on Recommend related operations or documents based on the status of the cloud service resources that the user is viewing or managing (such as running, stopped, configuring, etc.); recommend related resources or operations based on the resources the user is viewing (such as a specific virtual machine, database instance, etc.), that is, resource relevance, such as related network settings, security group rules, etc.; based on user preferences, such as language preferences, user interface settings, time zones, configuration preferences, etc., through user behavior patterns and selection preferences, intelligent word search can provide recommendations that are more in line with user habits. Some operations may be time-related, such as the formulation of snapshot backup plans. The recommendation system can consider the current time and date and recommend appropriate operations; based on the system's real-time load and performance data, recommend expansion, optimization or other performance improvement operations. Optionally, if the user is viewing alarms or events related to security, performance, etc., the recommendation system can also provide better practices or quick entry points for solving the problem. Optionally, the next logical step in the event process can be analyzed based on the event process and logic. For example, after the user completes resource creation, it is recommended to configure or deploy related services.

[0057] In another optional embodiment, relevant troubleshooting instructions or quick fix solutions can be recommended based on the current page content, that is, based on the page content currently browsed by the user, such as specific error messages, resource identifiers, etc., and more personalized recommendations can be provided based on historical operation information, such as frequently visited services, frequently performed operations, etc., that is, more personalized search results can be obtained.

[0058] Step S208: Send the search results to the client.

[0059] In an optional embodiment, after the server obtains the search results, it can send the search results to the client, and the client can display the search results to the user in any form, for example, by outputting on the client's graphical user interface, or through a data interface on the graphical user interface, and the user receives it through a USB flash drive, etc. Optionally, in this application, there is no specific restriction on the display form of the search results.

[0060] In an embodiment of the present application, a first text sent by a client and context information associated with the first text are received, wherein the first text is used to represent the text selected in the operation page displayed on the client, and the context information at least includes the page information of the operation page and the historical operation information of the operation page; a semantic analysis is performed on the first text to obtain a semantic analysis result of the first text; a search is performed based on the semantic analysis result and the context information to obtain a search result; and the search result is sent to the client. It is easy to notice that the client can provide the first text and the context information associated with the first text, and the server can then perform a semantic analysis on the first text to obtain a semantic analysis result, and then search based on the semantic analysis result combined with the context information to obtain a search result, and send the search result to the client. Since AI intelligent technology is combined in the search process, the user's search intention can be understood more accurately, thereby improving the search accuracy, thereby solving the technical problem of low accuracy of search results when performing intelligent word search.

[0061] In the above embodiment of the present application, semantic analysis is performed on the first text to obtain the semantic analysis result of the first text, including: preprocessing the first text to obtain the second text; and performing semantic analysis on the second text using a semantic analysis model to obtain the semantic analysis result.

[0062] The above-mentioned semantic analysis model can be a natural language processing agent (NLP Agent for short) task in context awareness (also known as Context Aware), for example, a BERT model, or a GPT series model. Optionally, in this application, the semantic analysis model is taken as an example of a BERT model.

[0063] In an optional embodiment, after obtaining the first text, the first text can be preprocessed, specifically, it can include text cleaning, word segmentation, removal of stop words, etc., so that the data in the first text is divided into a training set and a test set, that is, a second text is obtained, so that the second text can be semantically analyzed using a semantic analysis model to obtain a semantic analysis result. Optionally, a pre-trained BERT model can be selected, for example, a BERT model pre-trained on a large-scale general corpus, and the pre-trained model can be loaded using tools such as the Transformers library, so that the text can be converted into numerical features that the model can understand, that is, the second text is converted into understandable numerical features, so that the second text can be semantically analyzed using a semantic analysis model.

[0064] In the above embodiment of the present application, the method also includes: determining the current scene corresponding to the operation page; determining the preset analysis model corresponding to the current scene from multiple preset analysis models to obtain a semantic analysis model, wherein different preset analysis models correspond to different scenes.

[0065] In an optional embodiment, in order to ensure the accuracy of the search results obtained, a corresponding preset analysis model can be selected for different current scenarios on the operation page, and the preset analysis model corresponding to the current scenario is determined as a semantic analysis model. Optionally, after receiving the basic context information of the client, that is, the first text, the server calls the NLP Agent task corresponding to the scenario for the user's current scenario, provides more context information for the big model (such as: querying user roles and permissions, current service and resource status, operation history, events and alarms, etc.), and outputs it to the big model to complete the artificial intelligence workflow task (also called Artificial Intelligence Workflow task, referred to as AI WorkFlow task) of the corresponding scenario.

[0066] In the above embodiment of the present application, the method also includes: obtaining sample data, wherein the sample data includes sample text, label data corresponding to the sample text, and sample context information associated with the sample text; adjusting the pre-trained model using the sample text and label data to obtain an adjusted model; performing semantic analysis on the sample text using the adjusted model to obtain a sample analysis result of the sample text; searching based on the sample analysis result and the sample context information to obtain a sample search result; evaluating the adjusted model based on the sample search result to obtain an evaluation result of the adjusted model; adjusting the adjusted model based on the evaluation result to obtain a semantic analysis model.

[0067] The above-mentioned sample data and labeled data can be used to adjust the pre-trained model. Among them, the cloud product console sample data may come from various channels, such as user work orders, online consultations, user suggestions and feedback, help documents, community forums, duty records, network interface services and parameter definitions, error code information, resource diagnosis reports, etc. Optionally, sample data can be obtained from the above channels and classified and labeled according to data type or source to obtain labeled data.

[0068] In an optional embodiment, the pre-trained model can be adjusted using sample data and label data to obtain an adjusted model, and the sample text can be semantically analyzed using a semantic analysis model to obtain sample analysis results. Furthermore, a search can be performed based on the sample analysis results and sample context information to obtain sample search results. Optionally, the adjusted model can be evaluated using the accuracy of the sample search results, that is, the semantic analysis performance of the adjusted model can be evaluated. If the semantic analysis performance of the adjusted model has not yet met expectations, the sample search results can be used to adjust the model parameters until the semantic analysis performance of the adjusted model meets expectations, and the adjusted model whose semantic analysis performance still meets expectations is determined as the semantic analysis model.

[0069] Optionally, when fine-tuning the BERT model based on user consultation question data in the cloud product console, in order to adapt to specific tasks, an appropriate classification head can be added, and labeled data can be used for supervised learning to minimize the loss of the pre-trained model. Furthermore, the sample analysis results and sample context information (such as the current page, historical operations, etc.) can be integrated, and relevant resources can be queried through a matching algorithm, and personalized suggestions can be provided to users in combination with the recommendation system. The recommendation algorithm is based on collaborative filtering, content recommendation, deep learning and other technologies. Furthermore, the adjusted model can be evaluated, that is, the performance of the adjusted model can be evaluated using a test set. Specifically, it can include indicators such as accuracy, precision, and recall rate, and the adjusted model can be adjusted and optimized again based on the evaluation results to obtain a semantic analysis model.

[0070] In the above embodiment of the present application, the method also includes: sending the semantic analysis result to the client; receiving the feedback result sent by the client, wherein the feedback result is the result obtained by performing a feedback operation on the semantic analysis result; and adjusting the semantic analysis model based on the feedback result.

[0071] In an optional embodiment, after obtaining the semantic analysis results, the semantic analysis results can be sent to the client, and the client can provide feedback on the semantic analysis results, that is, feedback on the accuracy of the semantic analysis results. Optionally, when the feedback result shows that the accuracy of the semantic analysis results is low, the semantic analysis model can be adjusted again to improve the accuracy of the semantic analysis model. Optionally, a feedback mechanism can be set up in the user interface to regularly monitor the performance of the semantic analysis model and collect user feedback. If the semantic analysis model performs poorly in certain areas, user feedback can be used to further improve the semantic analysis model or increase training data.

[0072] In the above embodiment of the present application, the method further includes: storing the feedback results and / or the model performance of the semantic analysis model, and generating a model log.

[0073] In an optional embodiment, feedback results, that is, user satisfaction and the performance of the semantic analysis model, can be collected in real time, and a record log, that is, a model log, can be generated. Furthermore, the model log can be stored.

[0074] In the above embodiment of the present application, searching based on semantic analysis results and context information to obtain search results includes: inputting the semantic analysis results and context information into a text search model, and obtaining the search results output by the text search model.

[0075] The above-mentioned text search model can be used to perform text search based on semantic analysis results and context information to obtain search results. Optionally, this application does not impose specific restrictions on the type of text search model. In this application, the text search model is taken as an NLP model as an example for illustration.

[0076] In the above embodiment of the present application, sending search results to the client includes: storing the search results in a message queue; reading the search results from the message queue; and sending the search results to a communication endpoint that supports one-way communication, wherein the search results are pushed by the communication endpoint to the client connected to the communication endpoint.

[0077] In an optional embodiment, after obtaining the search results, the search results can be stored in a message queue, so that the search results can be read from the message queue. Further, the search results can be sent to a communication endpoint that supports unidirectional communication, so that the communication endpoint that supports unidirectional communication can push the search results to the client connected to the communication endpoint.

[0078] Figure 3 is a schematic diagram of a text search method according to an embodiment of the present application, such as Figure 3As shown, the cloud server console may include a console portal server and a global traffic server. The user may provide a first text and context information associated with the first text on the console and conduct a search. Optionally, assuming that the console is a console portal server, it is necessary to first perform initialization configuration and then perform a model service call, that is, adjust the pre-trained model to obtain a semantic analysis model, and use the speech analysis model to analyze the first text to obtain a semantic analysis result. Further, the global traffic server may be used for processing, that is, call the knowledge base and the large model to process the semantic analysis result and the context information to obtain the search result. Optionally, during the semantic analysis and search process, a message agent may be set to encapsulate and combine the search results generated by the large model and send them to the message queue of the message agent, wherein the message agent can ensure efficient message delivery and load balancing of consumer services. Furthermore, the search results are sent to the console client in the form of a message agent via the transmission control protocol.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0080] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0082] Example 2

[0083] According to an embodiment of the present application, a text search method is also provided, which is applied to a client. Figure 4 is a schematic diagram of a text search method according to Example 2 of the present application, such as Figure 4 As shown, the method includes the following steps:

[0084] Step S402: When a search instruction is received to search for text in an operation page, a first text corresponding to the search instruction and context information associated with the first text in the operation page are obtained, wherein the context information includes at least page information of the operation page and historical operation information of the operation page.

[0085] Step S404: sending the first text and context information to the server, and receiving search results returned by the server, wherein the search results are search results obtained by the server based on the semantic analysis results of the first text and the context information.

[0086] Step S406: Display the search results on the operation page.

[0087] In an optional embodiment, the user can give a search instruction on the operation page in any way to instruct to search for text in the operation page. After receiving the search instruction, the client can obtain the first text corresponding to the search instruction in the operation page, and the context information associated with the first text. Furthermore, the client can send the obtained first text and the context information associated with the first text to the server, and the server can search based on the first text and the context information associated with the first text to obtain search results. Optionally, after obtaining the search results, the server will send the search results to the client, so that the search results can be displayed on the operation page of the client.

[0088] In the above embodiment of the present application, when a search instruction for searching text in an operation page is received, obtaining the first text corresponding to the search instruction in the operation page includes one of the following: in response to a selection instruction for selecting text on the operation page, determining that the search instruction is received, and determining that the text corresponding to the selection instruction is the first text; in response to a selection instruction for selecting text on the operation page, displaying a preset icon on the operation page, and in response to a first preset operation acting on the preset icon, determining that the search instruction is received, and determining that the text corresponding to the selection instruction is the first text; in response to a first preset shortcut key being triggered, determining the text corresponding to the first preset shortcut key, and in response to a second preset shortcut key being triggered, determining that the search instruction is received, and determining that The text corresponding to the first preset shortcut key is determined to be the first text; in response to a selection instruction for selecting text on an operation page, the text corresponding to the selection instruction is determined; in response to a second preset operation acting on the text corresponding to the selection instruction, a menu list is displayed on the operation page; in response to a third preset operation acting on a preset item in the menu list, it is determined that a search instruction is received, and the text corresponding to the first preset shortcut key is determined to be the first text; in response to an operation instruction for copying or selecting text on the operation page, a prompt message is displayed on the operation page; in response to a confirmation instruction corresponding to the prompt message, it is determined that a search instruction is received, and the text corresponding to the operation instruction is determined to be the first text, wherein the prompt message is used to prompt whether to search for the text corresponding to the operation instruction.

[0089] The above-mentioned first preset shortcut key can be any one or more keys in the keyboard. Optionally, in this application, the first preset shortcut key is taken as shift+direction key as an example for explanation.

[0090] The above-mentioned second preset shortcut key can be any one or more keys in the keyboard. Optionally, in this application, the first preset shortcut key is command (ctrl) + space bar as an example for explanation.

[0091] In a first optional embodiment, when obtaining the first text on the operation page, it can be obtained by directly moving the floating icon (also called SmartBubble), and the search results can be displayed by the information display panel (also called SnapPanel).

[0092] In a second optional embodiment, if the received search instruction is in the form of a selection instruction for selecting text on the operation page in an immediate triggering manner, the selection content corresponding to the selection instruction can be directly determined as the first text.

[0093] In a third optional embodiment, if the received search instruction is in the form of a selection instruction for selecting text on the operation page through a floating icon, a floating icon, that is, a preset icon, can be displayed on the operation page. When the user clicks on the floating icon, that is, the first preset operation, it can be determined that the search instruction has been received, and the text corresponding to the selection instruction can be determined as the first text.

[0094] In a fourth optional embodiment, after the first preset shortcut key is triggered, the text corresponding to the first preset shortcut key can be determined first. After the second preset shortcut key is triggered, it can be determined that the search instruction has been received. Therefore, the content corresponding to the first preset shortcut key can be determined as the first text.

[0095] In a fifth optional embodiment, after receiving a selection instruction for selecting text on the operation page, the content corresponding to the selection instruction can be determined first. Furthermore, after receiving a second preset operation on the text corresponding to the selection instruction, a menu list is displayed on the operation page, that is, a context menu, wherein the second preset operation can be set by a technician in this field according to needs. In this application, right-clicking is used as an example for explanation. Further, when the user performs a third preset operation on the preset item in the menu list, it can be considered that a search instruction has been received, and the content corresponding to the first preset shortcut key is determined as the first text. In this application, there is no specific restriction on the operation method of the third preset operation. In this application, the third preset operation is taken as a click operation for explanation. The above-mentioned preset item can be an AI search item.

[0096] In a sixth optional embodiment, when an operation instruction to copy or select text on an operation page is received, a prompt message can be displayed on the operation page. When a confirmation instruction corresponding to the prompt message is received, it can be determined that a search instruction has been received, and the text corresponding to the operation instruction can be determined as the first text, wherein the prompt message can be information prompting whether to search for the text corresponding to the operation instruction.

[0097] In the above embodiment of the present application, in response to the first preset operation acting on the preset icon, it is determined that the search instruction is received, and the text corresponding to the selection instruction is determined to be the first text, including one of the following: in a case where the preset icon is a button control, in response to the first preset operation acting on the button control, it is determined that the search instruction is received, and the text corresponding to the selection instruction is determined to be the first text; in a case where the preset icon is a control set, in response to the first preset operation acting on the preset control in the control set, it is determined that the search instruction is received, and the text corresponding to the selection instruction is determined to be the first text.

[0098] The button control mentioned above may be an icon button.

[0099] The aforementioned control set may be a set including multiple button controls, such as a floating tool group.

[0100] The above-mentioned preset controls may be controls included in a control set, such as an AI search portal, etc.

[0101] In an optional embodiment, when the floating icon is an icon button, when a user clicks on the icon button, that is, a first preset operation, it can be determined that a search instruction has been received, and the text corresponding to the selection instruction is determined as the first text.

[0102] In another optional embodiment, when the floating icon is a floating tool group, when a user clicks on the floating icon in the floating tool group, that is, a first preset operation, it can be determined that a search instruction has been received, and the text corresponding to the selection instruction is determined as the first text.

[0103] In the above-mentioned embodiment of the present application, displaying search results on the operation page includes at least one of the following: displaying a floating card on the operation page, wherein the floating card displays summary information of the search results or link information of a preset page, and the page content of the preset page includes the search results; displaying the search results in a sidebar panel of the operation page; displaying a dialog window on the operation page, wherein the dialog window displays the search results.

[0104] In an optional embodiment, the search results may be displayed on the operation page in the following manner, for example, by displaying a floating card containing summary information of the search results or link information of a preset page on the operation page, or by displaying the search results directly in a sidebar panel of the operation page, or by displaying the search results on the operation page in the form of a dialog window.

[0105] In another optional embodiment, an event source (also called EventSource) connection can be established on the console front-end page, that is, in the client, the EventSource network interface is used to establish a connection with the server. The connection will be used to monitor events sent by the server through the application layer protocol. These events contain search result data, or use the intelligent search service (also called IntelliSearch Service) to receive search requests. That is, when the user initiates a search request, the client converts the user's word-marking information and context information into a unified request message format and sends it to the server.

[0106] Optionally, a message broker can be identified. When the IntelliSearch Service API receives a search request, it executes operations such as search engine calls, database queries, and NLP text processing. It then encapsulates and combines the search results generated by the large model and sends them to the message queue of the message broker. The message broker can ensure efficient message delivery and load balancing of the consumer service. Optionally, a consumer service can be used to process search requests. The IntelliSearchService includes a consumer listening service. The message broker configures the consumer service to listen to the corresponding Topic. When a new message arrives, the consumer service retrieves the message and processes it. Optionally, the search results can be sent via EventSource. After the IntelliSearch Service consumer service completes the processing, it sends the search results to an endpoint that supports Server-Sent Events (also known as Server-Sent Events, or SSE). This endpoint is responsible for "pushing" the data to all clients connected to it via EventSource. Optionally, you can also use the client's operation page to receive real-time search results and update the UI, and use the event source eventSource object on the browser side to continuously listen to events sent by the server. When the server sends new search results through the above endpoint, the client's event processing function will be triggered, and the client will then parse the received data and update the user interface, that is, display the search results on the user interface.

[0107] Optionally, if an error occurs during the entire process, such as a network problem, a server problem, or a message queue problem, the client's callback function will be triggered, and the client needs to provide an error prompt and handle it accordingly.

[0108] In the above embodiment of the present application, obtaining context information corresponding to the first text includes: obtaining initial context information associated with the first text; and unifying the format of the initial context information to obtain context information.

[0109] In an optional embodiment, considering that the cloud product console covers multiple product domains, in order to ensure the consistency of platform service calls in each product domain, a set of standardized context data formats can be defined to be compatible with various applications on various platforms, that is, the initial context information associated with the first text is obtained, and the format of the initial context information is unified to obtain context information.

[0110] In the above embodiment of the present application, the context information also includes: object information of the operation object corresponding to the operation page, operation preference information of the operation object, resource status of the resource to be operated corresponding to the operation page, system data corresponding to the operation page, event information or alarm information corresponding to the operation page, and the operation process corresponding to the operation page.

[0111] The above operation object may be the current user of the client.

[0112] In an optional embodiment, the context information may also include the object information of the current user, the operation preference information of the current user, the resource status of the resource to be operated corresponding to the operation page, the system data corresponding to the operation page, the event information or alarm information corresponding to the operation page, and the operation process corresponding to the operation page.

[0113] Example 3

[0114] According to an embodiment of the present application, a text search method is also provided, which is applied to a client. Figure 5 is a schematic diagram of a text search method according to Example 3 of the present application, such as Figure 5 As shown, the method includes the following steps:

[0115] Step S502: upon receiving a search instruction for searching text in a cloud product control page, obtaining first text corresponding to the search instruction in the cloud product control page, and context information associated with the first text, wherein the context information includes at least page information of the cloud product control page and historical operation information of the cloud product control page;

[0116] Step S504: Sending the first text and the context information to the cloud product control server, and receiving search results returned by the cloud product control server, wherein the search results are search results obtained by the cloud product control server based on the semantic analysis results of the first text and the context information, and the semantic analysis results are results obtained by the cloud product control server performing semantic analysis on the first text;

[0117] Step S506: Display the search results on the cloud product control page.

[0118] In an optional embodiment, when a search instruction is received to search for text in a cloud product control page, the first text corresponding to the search instruction and the context information associated with the first text in the cloud product control page can be obtained, and the first text and the context information can be sent to the cloud product control server. The cloud product control server generates corresponding search results and returns the search results to the cloud product control page, which displays the search results.

[0119] Example 4

[0120] According to an embodiment of the present application, a text search method is also provided, which is applied to a server. Figure 6 is a schematic diagram of a text search method according to Example 4 of the present application, such as Figure 6 As shown, the graphical user interface on the client 60 may display the first text and the context information associated with the first text, and send the first text and the context information to the server 61, so that the server 61 performs the following steps:

[0121] Step S602: Obtain the first text and context information associated with the first text by calling the first interface, wherein the first interface includes a first parameter, the parameter value of the first parameter includes the first text and the context information, the first text is used to represent the selected text in the operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page.

[0122] Step S604: performing semantic analysis on the first text to obtain a semantic analysis result of the first text.

[0123] Step S606: Search based on the semantic analysis result and context information to obtain search results.

[0124] Step S608: outputting the search results by calling the second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the search results.

[0125] The above-mentioned first interface can be an interface on the client, wherein the first interface can be an interface in the form of a virtual interface, a wired interface, etc. Optionally, this application does not impose any specific restrictions on the interface type of the first interface. In this application, the first interface can be used to obtain the first text and the context information associated with the first text.

[0126] The above-mentioned second interface can be an interface on the client, wherein the second interface can be an interface in the form of a virtual interface, a wired interface, etc. Optionally, this application does not impose specific restrictions on the interface type of the second interface. In this application, the second interface can be used to output search results.

[0127] In an optional embodiment, a user can input a first text and context information associated with the first text through a first interface on the client, and the server can obtain the first text and context information associated with the first text by calling the first interface and reading a parameter value at the first interface. Furthermore, the server can perform a semantic analysis on the first text to obtain a semantic analysis result of the first text, and thus perform a search based on the semantic analysis result and the context information to obtain a search result. Finally, the search result can be sent to the client and output through a second interface on the client. Optionally, the user can determine the search result by reading the parameter value of the second parameter on the second interface.

[0128] Example 5

[0129] According to an embodiment of the present application, a text search system is also provided. Figure 7 is a schematic diagram of a text search system according to Example 5 of the present application, such as Figure 7 As shown, the text search system 70 includes:

[0130] The client 701 is configured to, upon receiving a search instruction for searching text on an operation page, obtain first text on the operation page corresponding to the search instruction, and context information associated with the first text, wherein the context information includes at least page information of the operation page and historical operation information on the operation page;

[0131] The server 702 is connected to the client and is configured to perform semantic analysis on the first text to obtain a semantic analysis result of the first text, and perform a search based on the semantic analysis result and context information to obtain a search result;

[0132] The client 701 is also used to display the search results on the operation page.

[0133] In an optional embodiment, when the client 701 receives a search instruction to search for text in an operation page, it can obtain the first text corresponding to the search instruction in the operation page, as well as context information associated with the first text, and send the first text and the context information corresponding to the first text to the server 702, so that the first text can be semantically analyzed based on the server 702 to obtain the semantic analysis result of the first text, and a search can be performed based on the semantic analysis result and the context information to obtain search results. Further, the search results are sent to the client 701, and the client 701 displays the search results in the operation page.

[0134] Example 6

[0135] According to an embodiment of the present application, a device for implementing the above text search method is also provided, which is applied to a server. Figure 8 is a schematic diagram of a text search device according to Example 6 of the present application, such as Figure 8 As shown, the device includes: a receiving module 802 , an analyzing module 804 , a searching module 806 , and a sending module 808 .

[0136] Among them, the receiving module 802 is used to receive the first text sent by the client, and the context information associated with the first text, wherein the first text is used to represent the text selected in the operation page displayed on the client, and the context information at least includes the page information of the operation page, and the historical operation information of the operation page; the analysis module 804 is used to perform semantic analysis on the first text to obtain the semantic analysis result of the first text; the search module 806 is used to search based on the semantic analysis result and the context information to obtain the search result; the sending module 808 is used to send the search result to the client.

[0137] In the above embodiment of the present application, the analysis module 804 includes: a preprocessing unit, which is used to preprocess the first text to obtain a second text; and an analysis unit, which is used to perform semantic analysis on the second text using a semantic analysis model to obtain a semantic analysis result.

[0138] In the above embodiment of the present application, the device also includes: a first determination module, used to determine the current scene corresponding to the operation page; a second determination module, used to determine the preset analysis model corresponding to the current scene from multiple preset analysis models to obtain a semantic analysis model, wherein different preset analysis models correspond to different scenes.

[0139] In the above embodiment of the present application, the device also includes: an acquisition module for acquiring sample data, wherein the sample data includes sample text, label data corresponding to the sample text, and sample context information associated with the sample text; a first adjustment module for adjusting the pre-trained model using the sample text and label data to obtain an adjusted model; a second analysis module for performing semantic analysis on the sample text using the adjusted model to obtain a sample analysis result of the sample text; a second search module for searching based on the sample analysis result and the sample context information to obtain a sample search result; an evaluation module for evaluating the adjusted model based on the sample search result to obtain an evaluation result of the adjusted model; and a second adjustment module for adjusting the adjusted model based on the evaluation result to obtain a semantic analysis model.

[0140] In the above embodiment of the present application, the device also includes: a sending module for sending the semantic analysis result to the client; a receiving module for receiving the feedback result sent by the client, wherein the feedback result is the result obtained by performing a feedback operation on the semantic analysis result; and a third adjustment module for adjusting the semantic analysis model based on the feedback result.

[0141] In the above embodiment of the present application, the device also includes: a storage module, which is used to store the feedback results and / or the model performance of the semantic analysis model and generate a model log.

[0142] In the above embodiment of the present application, the search module 806 includes: a processing unit, which is used to input the semantic analysis results and context information into the text search model, and obtain the search results output by the text search model.

[0143] In the above embodiment of the present application, the sending module 808 includes: a storage unit for storing the search results in a message queue; a reading unit for reading the search results from the message queue; and a sending unit for sending the search results to a communication endpoint that supports one-way communication, wherein the search results are pushed by the communication endpoint to a client connected to the communication endpoint.

[0144] It should be noted that the receiving module 802, analyzing module 804, searching module 806, and sending module 808 correspond to steps S202 to S208 in Example 1. The examples and application scenarios implemented by the modules and corresponding steps are the same, but are not limited to the contents disclosed in Example 1. It should be noted that the modules or units can be hardware components or software components stored in a memory and processed by one or more processors. The modules can also be part of a device and can run in the server 10 provided in Example 1.

[0145] It should be noted that the preferred implementation scheme involved in the above embodiments of this application is the same as the scheme provided in Example 1, as well as the application scenario and implementation process, but is not limited to the scheme provided in Example 1.

[0146] Example 7

[0147] According to an embodiment of the present application, a device for implementing the above-mentioned text search method is also provided, which is applied to a client. Figure 9 is a schematic diagram of a text search device according to Example 7 of the present application, such as Figure 9 As shown, the device includes: an acquisition module 902 , a sending module 904 , and a display module 906 .

[0148] Among them, the acquisition module 902 is used to obtain the first text corresponding to the search instruction in the operation page and the context information associated with the first text when receiving a search instruction for searching the text in the operation page, wherein the context information at least includes the page information of the operation page and the historical operation information of the operation page; the sending module 904 is used to send the first text and the context information to the server, and receive the search results returned by the server, wherein the search results are the results obtained by the server based on the semantic analysis results and context information of the first text, and the semantic analysis results are the results obtained by the server performing semantic analysis on the first text; the display module 906 is used to display the search results in the operation page.

[0149] In the above embodiment of the present application, the acquisition module 902 includes one of the following: a first determination unit for determining that a search instruction is received in response to a selection instruction for selecting text on the operation page, and determining that the text corresponding to the selection instruction is the first text; a second determination unit for displaying a preset icon on the operation page in response to a selection instruction for selecting text on the operation page, determining that a search instruction is received in response to a first preset operation acting on the preset icon, and determining that the text corresponding to the selection instruction is the first text; a third determination unit for determining the text corresponding to the first preset shortcut key in response to the first preset shortcut key being triggered, determining that the search instruction is received in response to the second preset shortcut key being triggered, and determining that the text corresponding to the first preset shortcut key is the first text. a text; a fourth determining unit, for responding to a selection instruction for selecting text on an operation page, determining the text corresponding to the selection instruction, responding to a second preset operation acting on the text corresponding to the selection instruction, displaying a menu list on the operation page, responding to a third preset operation acting on a preset item in the menu list, determining that a search instruction is received, and determining that the text corresponding to the first preset shortcut key is the first text; a fifth determining unit, for responding to an operation instruction for copying or selecting text on the operation page, displaying a prompt message on the operation page, responding to a confirmation instruction corresponding to the prompt message, determining that a search instruction is received, and determining that the text corresponding to the operation instruction is the first text, wherein the prompt message is used to prompt whether to search for the text corresponding to the operation instruction.

[0150] In the above-mentioned embodiment of the present application, the second determination unit includes one of the following: a first determination subunit, which is used to, when the preset icon is a button control, respond to the first preset operation applied to the button control, determine that a search instruction has been received, and determine that the text corresponding to the selection instruction is the first text; a second determination subunit, which is used to, when the preset icon is a control set, respond to the first preset operation applied to the preset control in the control set, determine that a search instruction has been received, and determine that the text corresponding to the selection instruction is the first text.

[0151] In the above embodiment of the present application, the display module 906 includes at least one of the following: a first display unit, used to display a floating card on the operation page, wherein the floating card displays summary information of the search results or link information of a preset page, and the page content of the preset page includes the search results; a second display unit, used to display the search results in the sidebar panel of the operation page; a third display unit, used to display a dialog window on the operation page, wherein the dialog window displays the search results.

[0152] In the above embodiment of the present application, the acquisition module 902 includes: an acquisition unit, configured to acquire initial context information associated with the first text; and a unification unit, configured to unify the format of the initial context information to obtain context information.

[0153] It should be noted that the acquisition module 902, the sending module 904, and the display module 906 correspond to steps S402 to S406 in Example 2. The examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units can be hardware components or software components stored in a memory and processed by one or more processors. The above-mentioned modules can also be run as part of the device in the server 10 provided in Example 1.

[0154] It should be noted that the preferred implementation scheme involved in the above embodiments of this application is the same as the scheme provided in Example 2, as well as the application scenario and implementation process, but is not limited to the scheme provided in Example 2.

[0155] Example 8

[0156] According to an embodiment of the present application, a device for implementing the above-mentioned text search method is also provided, which is applied to a client. Figure 10 is a schematic diagram of a text search device according to Example 8 of the present application, such as Figure 10 As shown, the device includes: an acquisition module 1002 , a sending module 1004 , and a display module 1006 .

[0157] Among them, the acquisition module 1002 is used to obtain the first text corresponding to the search instruction in the cloud product control page and the context information associated with the first text when receiving a search instruction for searching the text in the cloud product control page, wherein the context information at least includes the page information of the cloud product control page and the historical operation information of the cloud product control page; the sending module 1004 is used to send the first text and the context information to the cloud product control server, and receive the search results returned by the cloud product control server, wherein the search results are the results obtained by the cloud product control server based on the semantic analysis results and context information of the first text, and the semantic analysis results are the results obtained by the cloud product control server performing semantic analysis on the first text; the display module 1006 is used to display the search results in the cloud product control page.

[0158] It should be noted that the acquisition module 1002, the sending module 1004, and the display module 1006 correspond to steps S502 to S506 in Example 3. The examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units can be hardware components or software components stored in a memory and processed by one or more processors. The above-mentioned modules can also be run as part of the device in the server 10 provided in Example 1.

[0159] It should be noted that the preferred implementation scheme involved in the above embodiments of this application is the same as the scheme provided in Example 3, as well as the application scenario and implementation process, but is not limited to the scheme provided in Example 3.

[0160] Example 9

[0161] According to an embodiment of the present application, a device for implementing the above text search method is also provided, which is applied to a server. Figure 11 is a schematic diagram of a text search device according to Example 9 of the present application, such as Figure 11 As shown, the device includes: an acquisition module 1102 , an analysis module 1104 , a search module 1106 , and an output module 1108 .

[0162] Among them, the acquisition module 1102 is used to obtain the first text and context information associated with the first text by calling the first interface, wherein the first interface includes a first parameter, the parameter value of the first parameter includes the first text and the context information, the first text is used to represent the selected text in the operation page displayed on the client, and the context information at least includes the page information of the operation page and the historical operation information of the operation page; the analysis module 1104 is used to perform semantic analysis on the first text to obtain the semantic analysis result of the first text; the search module 1106 is used to search based on the semantic analysis result and the context information to obtain the search result; the output module 1108 is used to output the search result by calling the second interface, wherein the second interface includes a second parameter, and the parameter value of the second parameter includes the search result.

[0163] It should be noted that the acquisition module 1102, analysis module 1104, search module 1106, and output module 1108 correspond to steps S602 to S608 in Example 3. The examples and application scenarios implemented by the modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units can be hardware components or software components stored in a memory and processed by one or more processors. The above-mentioned modules can also be run as part of the device in the server 10 provided in Example 1.

[0164] It should be noted that the preferred implementation scheme involved in the above embodiments of this application is the same as the scheme provided in Example 4, as well as the application scenario and implementation process, but is not limited to the scheme provided in Example 4.

[0165] Example 3

[0166] The embodiment of the present application may provide an electronic device, which may be any electronic device in a group of electronic devices. Optionally, in this embodiment, the electronic device may also be replaced by a terminal device such as a mobile terminal.

[0167] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0168] In this embodiment, the computer terminal can execute the program code in the method.

[0169] Optionally, Figure 12 is a structural block diagram of an electronic device according to an embodiment of the present application, such as Figure 12 As shown, the electronic device A may include: one or more (only one is shown in the figure) processors 1202, a memory 1204, a storage controller, and a peripheral interface, wherein the peripheral interface is connected to a radio frequency module, an audio module and a display.

[0170] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the methods in the above embodiments. 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 a memory remotely located relative to the processor, and these remote memories may be connected to terminal A via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0171] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: receiving a first text sent by the client and context information associated with the first text, wherein the first text is used to represent the selected text in the operation page displayed on the client, and the context information at least includes page information of the operation page and historical operation information of the operation page; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; performing a search based on the semantic analysis result and the context information to obtain a search result; and sending the search result to the client.

[0172] Optionally, the processor may further execute program code of the following steps: preprocessing the first text to obtain a second text; and performing semantic analysis on the second text using a semantic analysis model to obtain a semantic analysis result.

[0173] Optionally, the processor may also execute the program code of the following steps: determining the current scene corresponding to the operation page; determining the preset analysis model corresponding to the current scene from multiple preset analysis models to obtain a semantic analysis model, wherein different preset analysis models correspond to different scenes.

[0174] Optionally, the processor may also execute the program code of the following steps: obtaining sample data, wherein the sample data includes sample text, label data corresponding to the sample text, and sample context information associated with the sample text; adjusting the pre-trained model using the sample text and label data to obtain an adjusted model; performing semantic analysis on the sample text using the adjusted model to obtain a sample analysis result of the sample text; performing a search based on the sample analysis result and the sample context information to obtain a sample search result; evaluating the adjusted model based on the sample search result to obtain an evaluation result of the adjusted model; and adjusting the adjusted model based on the evaluation result to obtain a semantic analysis model.

[0175] Optionally, the processor may also execute the program code of the following steps: sending the semantic analysis result to the client; receiving the feedback result sent by the client, wherein the feedback result is the result obtained by performing a feedback operation on the semantic analysis result; and adjusting the semantic analysis model based on the feedback result.

[0176] Optionally, the processor may further execute program code of the following steps: storing the feedback results and / or the model performance of the semantic analysis model, and generating a model log.

[0177] Optionally, the processor may further execute program code of the following steps: inputting semantic analysis results and context information into a text search model, and obtaining search results output by the text search model.

[0178] Optionally, the processor may also execute the program code of the following steps: storing the search results in a message queue; reading the search results from the message queue; and sending the search results to a communication endpoint that supports one-way communication, wherein the search results are pushed by the communication endpoint to a client connected to the communication endpoint.

[0179] According to an embodiment of the present application, a method for text search is provided, which receives a first text sent by a client and context information associated with the first text, wherein the first text is used to represent the text selected in the operation page displayed on the client, and the context information at least includes page information of the operation page and historical operation information of the operation page; performs semantic analysis on the first text to obtain a semantic analysis result of the first text; performs a search based on the semantic analysis result and the context information to obtain a search result; and sends the search result to the client. It is easy to notice that the client can provide the first text and the context information associated with the first text, and then the server performs semantic analysis on the first text to obtain a semantic analysis result, and then performs a search based on the semantic analysis result combined with the context information to obtain a search result, and sends the search result to the client. Since AI intelligent technology is combined in the search process, the user's search intention can be understood more accurately, thereby improving the search accuracy, thereby solving the technical problem of low accuracy of search results when performing intelligent word search.

[0180] Those skilled in the art will appreciate that the structure shown in the figure is for illustrative purposes only, and the electronic device may also be a smartphone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (MID), a PAD, or other terminal device. This figure does not limit the structure of the above-mentioned electronic devices. For example, electronic device A may include more or fewer components (such as a network interface, a display device, etc.) than shown in the figure, or have a configuration different from that shown in the figure.

[0181] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0182] Example 4

[0183] The embodiment of the present application further provides a computer-readable storage medium. Optionally, in this embodiment, the computer-readable storage medium can be used to store the program code executed by the method provided in the above embodiment.

[0184] Optionally, in this embodiment, the storage medium may be located in any electronic device in a group of electronic devices in a computer network, or in any mobile terminal in a group of mobile terminals.

[0185] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for executing the following steps: receiving a first text sent by a client, and context information associated with the first text, wherein the first text is used to represent the selected text in the operation page displayed on the client, and the context information includes at least page information of the operation page, and historical operation information of the operation page; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; performing a search based on the semantic analysis result and the context information to obtain a search result; and sending the search result to the client.

[0186] Optionally, the computer-readable storage medium is also configured to store program code for executing the following steps: receiving a first text sent by a client, and context information associated with the first text, wherein the first text is used to represent the selected text in the operation page displayed on the client, and the context information includes at least page information of the operation page, and historical operation information of the operation page; performing semantic analysis on the first text to obtain a semantic analysis result of the first text; performing a search based on the semantic analysis result and the context information to obtain a search result; and sending the search result to the client.

[0187] Example 5

[0188] The embodiment of the present application further provides a computer program product. Optionally, in this embodiment, the computer program product may include a computer program, and when the computer program is executed by a processor, the method provided in the embodiment is implemented.

[0189] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0190] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0191] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0192] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0193] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0194] If the integrated unit is implemented in the form of 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 the present application is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0195] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A text search method, characterized in that: Applied to a server, the method includes: receiving a first text sent by a client, and context information associated with the first text, wherein the first text is used to represent text selected in an operation page displayed on the client, and the context information includes at least page information of the operation page and historical operation information of the operation page; Performing semantic analysis on the first text to obtain a semantic analysis result of the first text; Performing a search based on the semantic analysis result and the context information to obtain search results; The search results are sent to the client.

2. The method according to claim 1, characterized in that The performing semantic analysis on the first text to obtain a semantic analysis result of the first text includes: Preprocessing the first text to obtain a second text; Perform semantic analysis on the second text using a semantic analysis model to obtain the semantic analysis result.

3. The method according to claim 2, characterized in that The method further comprises: Determine the current scene corresponding to the operation page; A preset analysis model corresponding to the current scene is determined from a plurality of preset analysis models to obtain the semantic analysis model, wherein different preset analysis models correspond to different scenes.

4. The method according to claim 2 or 3, characterized in that The method further comprises: Acquire sample data, wherein the sample data includes sample text, label data corresponding to the sample text, and sample context information associated with the sample text; Adjusting the pre-trained model using the sample text and the label data to obtain an adjusted model; Performing semantic analysis on the sample text using the adjusted model to obtain a sample analysis result of the sample text; Performing a search based on the sample analysis result and the sample context information to obtain a sample search result; Evaluate the adjusted model based on the sample search results to obtain an evaluation result of the adjusted model; The adjusted model is adjusted based on the evaluation result to obtain the semantic analysis model.

5. The method according to claim 2 or 3, characterized in that The method further comprises: Sending the semantic analysis result to the client; receiving a feedback result sent by the client, wherein the feedback result is a result obtained by performing a feedback operation on the semantic analysis result; The semantic analysis model is adjusted based on the feedback result.

6. The method according to claim 5, characterized in that The method further comprises: The feedback result and / or the model performance of the semantic analysis model are stored to generate a model log.

7. The method according to claim 1, characterized in that The searching based on the semantic analysis result and the context information to obtain search results includes: The semantic analysis result and the context information are input into a text search model, and the search result output by the text search model is obtained.

8. The method according to claim 1, characterized in that The sending the search results to the client includes: Storing the search results in a message queue; Reading the search result from the message queue; The search result is sent to a communication endpoint that supports one-way communication, wherein the search result is pushed by the communication endpoint to the client connected to the communication endpoint.

9. A text search method, characterized in that: Applied to a client, the method includes: Upon receiving a search instruction for searching text in an operation page, obtaining first text corresponding to the search instruction in the operation page, and context information associated with the first text, wherein the context information includes at least page information of the operation page and historical operation information of the operation page; Sending the first text and the context information to a server, and receiving search results returned by the server, wherein the search results are search results based on semantic analysis results of the first text and the context information; The search results are displayed on the operation page.

10. The method according to claim 9, characterized in that The step of, upon receiving a search instruction for searching text in an operation page, obtaining a first text in the operation page corresponding to the search instruction, includes one of the following steps: In response to a selection instruction for selecting text on the operation page, determining that the search instruction is received, and determining that the text corresponding to the selection instruction is the first text; In response to a selection instruction for selecting text on the operation page, displaying a preset icon on the operation page, and in response to a first preset operation performed on the preset icon, determining that the search instruction has been received, and determining that the text corresponding to the selection instruction is the first text; In response to a first preset shortcut key being triggered, determining a text corresponding to the first preset shortcut key; in response to a second preset shortcut key being triggered, determining that the search instruction is received, and determining that the text corresponding to the first preset shortcut key is the first text; In response to a selection instruction for selecting text on the operation page, determining the text corresponding to the selection instruction, in response to a second preset operation acting on the text corresponding to the selection instruction, displaying a menu list on the operation page, in response to a third preset operation acting on a preset item in the menu list, determining that the search instruction is received, and determining that the text corresponding to the first preset shortcut key is the first text; In response to an operation instruction to copy or select text on the operation page, a prompt message is displayed on the operation page. In response to a confirmation instruction corresponding to the prompt message, it is determined that the search instruction has been received, and it is determined that the text corresponding to the operation instruction is the first text, wherein the prompt message is used to prompt whether to search for the text corresponding to the operation instruction.

11. The method according to claim 10, characterized in that The step of determining that the search instruction is received in response to the first preset operation on the preset icon and determining that the text corresponding to the selection instruction is the first text includes one of the following: In a case where the preset icon is a button control, in response to the first preset operation acting on the button control, determining that the search instruction is received, and determining that the text corresponding to the selection instruction is the first text; In a case where the preset icon is a control set, in response to the first preset operation acting on a preset control in the control set, it is determined that the search instruction is received, and it is determined that the text corresponding to the selection instruction is the first text.

12. The method according to claim 9, characterized in that Displaying the search results on the operation page includes at least one of the following: Displaying a floating card on the operation page, wherein the floating card displays summary information of the search results or link information of a preset page, and the page content of the preset page includes the search results; Displaying the search results in a sidebar panel of the operation page; A dialog window is displayed on the operation page, wherein the search result is displayed in the dialog window.

13. The method according to claim 9, characterized in that The acquiring of context information corresponding to the first text includes: Obtaining initial context information associated with the first text; The initial context information is formatted uniformly to obtain the context information.

14. The method according to claim 9 or 13, characterized in that The context information also includes: object information of the operation object corresponding to the operation page, operation preference information of the operation object, resource status of the resource to be operated corresponding to the operation page, system data corresponding to the operation page, event information or alarm information corresponding to the operation page, and operation process corresponding to the operation page.

15. A text search method, characterized in that: Applied to a client, the method includes: Upon receiving a search instruction for searching text in a cloud product control page, obtaining first text corresponding to the search instruction in the cloud product control page, and context information associated with the first text, wherein the context information includes at least page information of the cloud product control page and historical operation information of the cloud product control page; Sending the first text and the context information to a cloud product control server, and receiving a search result returned by the cloud product control server, wherein the search result is a result obtained by the cloud product control server through a search based on a semantic analysis result of the first text and the context information, and the semantic analysis result is a result obtained by the cloud product control server through a semantic analysis of the first text; The search results are displayed on the cloud product control page.

16. A text search method, characterized in that: Applied to a server, the method includes: Obtaining first text and context information associated with the first text by calling a first interface, wherein the first interface includes a first parameter, a parameter value of the first parameter includes the first text and the context information, the first text is used to represent text selected on an operation page displayed on a client, and the context information includes at least page information of the operation page and historical operation information of the operation page; Performing semantic analysis on the first text to obtain a semantic analysis result of the first text; Performing a search based on the semantic analysis result and the context information to obtain search results; The search result is output by calling a second interface, wherein the second interface includes a second parameter, and a parameter value of the second parameter includes the search result.

17. An electronic device, characterized in that: include: a memory storing an executable program; A processor, configured to run the program, wherein the program, when running, executes the method according to any one of claims 1 to 16.

18. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 16.

19. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 16.

Images