A data content query method and device, and a storage medium

Abstract

The application discloses a data content query method and device and a storage medium. The method comprises the following steps: acquiring query information input by a target object; searching based on the query information to display a plurality of search results matched with the target object; and in response to selection of the search results by the target object, displaying data content corresponding to the search results in a target interface and highlighting key knowledge points in the data content corresponding to the search results. Thus, a targeted data query process is realized. Since the key knowledge points are used for data content query and the key knowledge points are displayed, the query basis of the search results is embodied, the relevance of the obtained content is improved, the user can quickly obtain the query content, and thus the accuracy of data content query is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, and storage medium for querying data content. Background Technology

[0002] With the rapid development of internet technology, people have increasingly higher demands for data content. Furthermore, the querying and searching of data content is becoming increasingly important in people's lives.

[0003] Generally, the data content query process involves matching the user's query terms with the results, matching relevant results based on dimensions such as content relevance, authority, and content quality, and then displaying a partial summary on the results page for the user to gain an initial understanding.

[0004] However, the query terms entered by users may be generic, resulting in generalized search results. Users cannot quickly identify relevant search results, causing the data content of the query to not match their needs and affecting the accuracy of the data content query. Summary of the Invention

[0005] In view of this, this application provides a method for querying data content, which can effectively improve the accuracy of data content query.

[0006] The first aspect of this application provides a method for querying data content, which can be applied to a system or program in a terminal device that includes a data content query function, specifically including:

[0007] Retrieve the query information input by the target object;

[0008] Based on the query information, a content search is performed to display multiple search results that match the target object on the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information. The key knowledge points are obtained by extracting information from the data content corresponding to the search results.

[0009] In response to the target object's selection of the search result, the data content corresponding to the search result is displayed on the target interface, and the key knowledge points are highlighted in the data content corresponding to the search result.

[0010] Optionally, in some possible implementations of this application, the search result is the data content among the candidate content, and the method further includes:

[0011] Determine the start and end positions of the text in the candidate content to obtain the input information;

[0012] The input information is fed into a deep model to obtain the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0013] The key knowledge points corresponding to the candidate content are determined based on the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0014] The process of searching for content corresponding to the query information is based on the key knowledge points corresponding to the candidate content.

[0015] Optionally, in some possible implementations of this application, the method further includes:

[0016] Obtain preset samples;

[0017] Determine the content title corresponding to the preset sample;

[0018] Based on the content title, the knowledge points in the preset sample are sequentially labeled to obtain training samples;

[0019] The deep model is trained based on the training samples.

[0020] Optionally, in some possible implementations of this application, the step of performing a content search based on the query information to display multiple search results matching the target object on the target interface includes:

[0021] Based on the query information, knowledge points are matched to determine candidate knowledge points;

[0022] Obtain the object tag corresponding to the target object;

[0023] Input the object labels into the recommendation model to obtain recommended knowledge points;

[0024] The key knowledge points are determined by matching the recommended knowledge points with the candidate knowledge points.

[0025] Based on the key knowledge points, multiple search results matching the target object are determined and displayed on the target interface.

[0026] Optionally, in some possible implementations of this application, the method further includes:

[0027] Retrieve the query records corresponding to the target object;

[0028] Statistical analysis is performed on the click operations of knowledge points in the query records to obtain the click data corresponding to the target object;

[0029] The recommendation model is trained based on the click data.

[0030] Optionally, in some possible implementations of this application, the method further includes:

[0031] In the target interface that highlights the key knowledge points, obtain the operation information corresponding to the target object;

[0032] If the operation information indicates to stop the operation, an associated window containing related content will be displayed on the target interface. The associated content is determined based on the key knowledge points.

[0033] Optionally, in some possible implementations of this application, the step of displaying an associated window containing related content in the target interface if the operation information indicates to stop the operation includes:

[0034] If the operation information indicates to stop the operation, then candidate related content is determined based on the key knowledge points;

[0035] Obtain the search volume of the candidate related content within a preset period;

[0036] The relevant content is determined by sorting based on the search volume.

[0037] The associated window containing the associated content is displayed in the target interface.

[0038] A second aspect of this application provides a data content query device, comprising:

[0039] The acquisition unit is used to acquire the query information input by the target object;

[0040] The query unit is used to perform content search based on the query information to display multiple search results that match the target object in the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information. The key knowledge points are obtained by extracting information from the data content corresponding to the search results.

[0041] The query unit is further configured to respond to the target object's selection of the search result, display the data content corresponding to the search result in the target interface, and highlight the key knowledge points in the data content corresponding to the search result.

[0042] Optionally, in some possible implementations of this application, the search result is the data content in the candidate content, and the query unit is specifically used to determine the text start position and text end position in the candidate content to obtain input information;

[0043] The query unit is specifically used to input the input information into the deep model to obtain the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0044] The query unit is specifically used to determine the key knowledge points corresponding to the candidate content based on the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0045] The query unit is specifically used to perform a content search process for the query information based on the key knowledge points corresponding to the candidate content.

[0046] Optionally, in some possible implementations of this application, the query unit is specifically used to obtain a preset sample;

[0047] The query unit is specifically used to determine the content title corresponding to the preset sample;

[0048] The query unit is specifically used to perform sequence labeling on the knowledge points in the preset sample based on the content title in order to obtain training samples;

[0049] The query unit is specifically used to train the deep model based on the training samples.

[0050] Optionally, in some possible implementations of this application, the query unit is specifically used to perform knowledge point matching based on the query information to determine candidate knowledge points;

[0051] The query unit is specifically used to obtain the object tag corresponding to the target object;

[0052] The query unit is specifically used to input the object tags into the recommendation model to obtain recommended knowledge points;

[0053] The query unit is specifically used to match the recommended knowledge points with the candidate knowledge points to determine the key knowledge points;

[0054] The query unit is specifically used to determine multiple search results that match the target object based on the key knowledge points, and to display them on the target interface.

[0055] Optionally, in some possible implementations of this application, the query unit is specifically used to obtain the query record corresponding to the target object;

[0056] The query unit is specifically used to perform statistics based on the click operations of knowledge points in the query records in order to obtain the click data corresponding to the target object;

[0057] The query unit is specifically used to train the recommendation model based on the click data.

[0058] Optionally, in some possible implementations of this application, the method further includes:

[0059] The query unit is specifically used to obtain the operation information corresponding to the target object in the target interface of the highlighted key knowledge points;

[0060] The query unit is specifically used to display a related window containing related content on the target interface if the operation information indicates to stop the operation. The related content is determined based on the key knowledge points.

[0061] Optionally, in some possible implementations of this application, the query unit is specifically used to determine candidate related content based on the key knowledge points if the operation information indicates to stop the operation;

[0062] The query unit is specifically used to obtain the search volume of the candidate related content within a preset period;

[0063] The query unit is specifically used to sort based on the search volume in order to determine the associated content;

[0064] The query unit is specifically used to display the associated window containing the associated content in the target interface.

[0065] A third aspect of this application provides a computer device, comprising: a memory, a processor, and a bus system; the memory is used to store program code; the processor is used to execute the data content querying method described in the first aspect or any one of the first aspects according to instructions in the program code.

[0066] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the data content query method described in the first aspect or any one of the first aspects.

[0067] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the data content querying method provided in the first aspect or various optional implementations thereof.

[0068] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0069] By acquiring the query information input by the target user, and then performing a content search based on that query information, multiple search results matching the target user are displayed on the target interface. Each search result is configured with corresponding key knowledge points for display. These key knowledge points are obtained by matching the query information with the key knowledge points, which are extracted from the data content corresponding to the search results. Furthermore, in response to the target user's selection of search results, the corresponding data content is displayed on the target interface, with the key knowledge points highlighted within the data content. This achieves a targeted data query process. Because the data content is queried using key knowledge points, and the display of these key knowledge points reflects the basis of the search results, the relevance of the retrieved content is improved, allowing users to quickly obtain the query content and thus improving the accuracy of data content retrieval. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0071] Figure 1 Network architecture diagram for the data content query system;

[0072] Figure 2 A flowchart illustrating the process of querying data content provided in this application embodiment;

[0073] Figure 3 A flowchart illustrating a data content query method provided in an embodiment of this application;

[0074] Figure 4 A schematic diagram illustrating a data content query method provided in an embodiment of this application;

[0075] Figure 5 A schematic diagram illustrating a scenario for another data content query method provided in an embodiment of this application;

[0076] Figure 6 A flowchart illustrating another data content query method provided in this application embodiment;

[0077] Figure 7 A flowchart illustrating another data content query method provided in this application embodiment;

[0078] Figure 8A schematic diagram illustrating a scenario for another data content query method provided in an embodiment of this application;

[0079] Figure 9 A flowchart illustrating another data content query method provided in this application embodiment;

[0080] Figure 10 A flowchart illustrating another data content query method provided in this application embodiment;

[0081] Figure 11 A schematic diagram of the structure of a data content query device provided in an embodiment of this application;

[0082] Figure 12 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application;

[0083] Figure 13 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0084] This application provides a data content query method and related apparatus, which can be applied to systems or programs in terminal devices that include data content query functions. The method involves acquiring query information input by a target object; then performing a content search based on the query information to display multiple search results matching the target object on a target interface. Each search result is configured with corresponding key knowledge points for display. These search results are obtained by matching the key knowledge points with the query information, and the key knowledge points are derived from information extraction of the data content corresponding to the search results. Furthermore, in response to the target object's selection of search results, the data content corresponding to the search results is displayed on the target interface, and the key knowledge points are highlighted within the data content corresponding to the search results. This achieves a targeted data query process. Because key knowledge points are used for data content querying, and the display of these key knowledge points reflects the query basis of the search results, the relevance of the retrieved content is improved, allowing users to quickly obtain the query content, thereby improving the accuracy of data content querying.

[0085] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding to,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0086] First, some terms that may appear in the embodiments of this application will be explained.

[0087] Data annotation: Data annotation is the process of labeling metadata such as text, video, and images. The labeled data will be used by users to train machine learning models.

[0088] Key knowledge points: Key knowledge points refer to a summary of the author's core viewpoints in an article. For example: the ketogenic diet. Multiple key knowledge points can be extracted from an article.

[0089] A text segment (Span): A continuous piece of text in a longer text such as an article or paragraph, usually consisting of one or more consecutive words. A "sentence" can be considered a special type of "text segment".

[0090] Sequence labeling: marking one or more "segments of text" from a given long text, either manually or by a computer program.

[0091] User profiling: By analyzing a large amount of user behavior data and using machine learning and data mining techniques, we can create a multi-dimensional profile for each user currently being served. Common dimensions include age group, gender, education level, and interest markers.

[0092] Information extraction: The process of extracting target fragments from target text.

[0093] Inverted index: Web pages (articles) indexed by search engines are stored in an inverted chain data structure. An index is a special computer file format that can accelerate how computer programs (retrieval programs) retrieve articles related to a user's query.

[0094] Semantic vector indexing: This involves using machine learning to represent the semantics of natural language text as vectors, and storing these vectors in a specific data structure within the file system or memory. This data structure facilitates subsequent queries. A common application is "face recognition," where the main logic involves creating an index by representing images as vectors. Real-time sampling of the user's face is converted into vectors, and the final task is to find the vector in the index that is closest to the user's facial input vector.

[0095] Landing page: When a user clicks on one of the results returned by a search engine, they are taken to a new webpage, which is called the "landing page".

[0096] Training set: Machine "learning" usually requires a certain amount of data to eventually output a model with a certain ability. This process is called the "training process", and the data used for learning in this process is called the "training set".

[0097] It should be understood that the data content query method provided in this application can be applied to systems or programs in terminal devices that include data content query functions, such as search applications. Specifically, the data content query system can run on systems such as... Figure 1 In the network architecture shown, such as Figure 1 The diagram shows the network architecture of the data content query system. As can be seen, the system can provide query services for data from multiple information sources. Specifically, the terminal sends query information to the server, which then performs the query and displays the relevant data on the terminal interface. Figure 1 The document shows various terminal devices, which can be computer devices. In real-world scenarios, more or fewer types of terminal devices may participate in the data query process. The specific number and types depend on the actual scenario and are not limited here. Additionally, Figure 1 The image shows one server, but in real-world scenarios, multiple servers can be involved, with the specific number depending on the actual situation.

[0098] In this embodiment, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, smart voice interaction device, smart home appliance, in-vehicle terminal, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, and the terminal and server can be connected to form a blockchain network; this application does not impose any restrictions.

[0099] It is understood that the aforementioned data content query system can run on personal mobile terminals, such as as a search application, or it can run on a server, or it can run on third-party devices to provide data content queries and obtain the query processing results of the information source's data content. Specifically, the data content query system can run as a program on the aforementioned devices, or it can run as a system component of the aforementioned devices, or it can run as a cloud service program. This embodiment can be applied to scenarios such as cloud technology and autonomous driving. The specific operating mode depends on the actual scenario and is not limited here.

[0100] With the rapid development of internet technology, people have increasingly higher demands for data content. Furthermore, the querying and searching of data content is becoming increasingly important in people's lives.

[0101] Generally, the data content query process involves matching the user's query terms with the results, matching relevant results based on dimensions such as content relevance, authority, and content quality, and then displaying a partial summary on the results page for the user to gain an initial understanding.

[0102] However, the query terms entered by users may be generic, resulting in generalized search results. Users cannot quickly identify relevant search results, causing the data content of the query to not match their needs and affecting the accuracy of the data content query.

[0103] To address the aforementioned problems, this application proposes a data content query method, which is applied to... Figure 2 In the data content query process framework shown, such as Figure 2 The diagram shown is a flowchart of a data content query process provided in an embodiment of this application. The user performs a query operation on the terminal, the server performs knowledge point matching, and further performs targeted filtering based on the user to determine the search results containing the knowledge points.

[0104] Understandably, because users' queries are often broad, the provided answers tend to be lengthy and complex, leading to high reading costs and the possibility of not finding the desired results, resulting in low information retrieval efficiency. Therefore, this embodiment extracts key knowledge points from the content of articles to answer user questions. Based on these extracted knowledge points, it matches the user's personalized information to provide the most suitable knowledge, and can also sort the knowledge points according to current search volume, prioritizing the display of popular knowledge.

[0105] It is understood that the method provided in this application can be a program written as processing logic in a hardware system, or a data content query device, implemented through integration or external connection. As one implementation, the data content query device acquires query information input by the target object; then performs a content search based on the query information to display multiple search results matching the target object on the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information, and these key knowledge points are derived from information extraction of the data content corresponding to the search results. Furthermore, in response to the target object's selection of search results, the device displays the data content corresponding to the search results on the target interface, highlighting the key knowledge points within the data content. This achieves a targeted data query process. Because key knowledge points are used for data content querying, and the display of these key knowledge points reflects the query basis of the search results, the relevance of the retrieved content is improved, allowing users to quickly obtain the query content and thus improving the accuracy of data content querying.

[0106] Based on the above process architecture, the following section will introduce the methods for querying data content in this application. Please refer to [link / reference]. Figure 3 , Figure 3 The flowchart illustrates a data content query method provided in this application embodiment. This management method can be executed by a terminal or a server, and this application embodiment includes at least the following steps:

[0107] 301. Obtain the query information input by the target object.

[0108] In this embodiment, the target object can be a user, a terminal, or other functional entity that may perform query or search operations, and is not limited here.

[0109] In some possible scenarios, the target can input query information through keyboard input, voice input, or other haptic interaction methods. The specific input method depends on the actual scenario and is not limited here.

[0110] Specifically, the query information can be knowledge points extracted from the input information of the target object, or it can be a combination of multiple knowledge points. By parsing the query information, the efficiency of subsequent knowledge point matching can be improved.

[0111] 302. Perform content search based on query information to display multiple search results that match the target object on the target interface.

[0112] In this embodiment, each search result is configured with corresponding key knowledge points for display. This is because the search results are obtained by matching the key knowledge points with the query information, and the key knowledge points are obtained by extracting information from the data content corresponding to the search results. That is, the key knowledge points are obtained by automatically or manually annotating the candidate content in the database in advance.

[0113] It is understood that search results can be in the form of articles, web pages, or other media content. The following examples use articles as an example for illustration, but are not intended to be limiting.

[0114] Specifically, through the configuration of key knowledge points, the query process in this embodiment is as follows: Figure 4 As shown, Figure 4 This is a schematic diagram illustrating a data content query method provided in this application embodiment. First, the user enters a query term; then, multiple search results (result 1 to result N) are returned and sorted. Further, based on the identified key knowledge points, the user first learns about the solutions contained in the article and can choose to click to enter the details page to read the detailed content (for example, if the user prefers solution 2 and solution 4, they can directly select result 2 and result N to view the detailed content), making information acquisition more efficient. Additionally, the user can enter the details page to read detailed information and can directly initiate a new search to learn about the relevant content of the solution through the details page entry.

[0115] As can be seen, by configuring key knowledge points, key viewpoints in an article can be identified and key knowledge points can be extracted. These points can then be presented in a structured manner before the user reads the content, providing a reference for the user. This allows the user to understand the content explained in the article in advance, intuitively saving reading costs and improving information acquisition efficiency.

[0116] 303. In response to the target object's selection of search results, display the data content corresponding to the search results in the target interface, and highlight key knowledge points in the data content corresponding to the search results.

[0117] In this embodiment, by highlighting key knowledge points in the data content corresponding to the search results, users can understand the content explained in the article in advance, intuitively saving reading costs and improving the accuracy of search results for users.

[0118] In one possible scenario, the above query process is as follows: Figure 5 As shown, Figure 5 This is a schematic diagram illustrating another data content query method provided in this application embodiment. The diagram shows a user entering the query term "how to slim legs" in the target interface A1, resulting in multiple search results (Result 1: What are some amazing leg slimming methods?; Result 2: How to successfully slim legs if you have uneven body fat; Result 3: How to lose leg fat drastically in one month). Each result displays key knowledge points A2 in the lower right corner, allowing the user to first understand that Result 2 involves slimming legs through correct walking posture, and then select the content (Result 2) to click for detailed reading. Furthermore, the user enters the details page to read more information and can initiate a new search to learn more about the solution by clicking on knowledge point A3 at the top of the details page.

[0119] Furthermore, to highlight the relevance of key knowledge points to users, related information can be pushed out. Specifically, on the target interface highlighting the key knowledge points, the corresponding operation information is retrieved. If the operation information indicates to stop the operation, a related window containing relevant content is displayed on the target interface. This related content is determined based on the key knowledge points. Specifically, the related content can be questions of interest to users, which can be represented by search volume. If the operation information indicates to stop the operation, candidate related content is determined based on the key knowledge points; then, the search volume of the candidate related content within a preset period is retrieved; and the content is sorted based on the search volume to determine the relevant content; finally, a related window containing the relevant content is displayed on the target interface. In one possible scenario, such as... Figure 6 As shown, Figure 6 The flowchart illustrates another data content query method provided in this application embodiment. When scrolling down to read content, the "knowledge point keywords" mentioned in the article are highlighted the first time they appear (e.g., "Walking Correctly"). Once the screen is still (operation information indicates stop operation), a recommendation pop-up titled "Correct Walking Posture: How to Apply Force" (related window B1) slides up from the bottom of the screen, displays for 3 seconds, and disappears when scrolled down without being clicked. This recommendation pop-up matches the most frequently searched results with a high volume of key knowledge points from the article, making it easier for users to understand.

[0120] The above embodiments describe the interface display process related to the query process based on key knowledge points. The configuration of key knowledge points can be obtained by pre-processing the data content in the retrieval database. The configuration process of key knowledge points is explained below. Figure 7 As shown, Figure 7This is a flowchart illustrating another data content query method provided in this application embodiment. The flowchart shows the query process on the user's search side: a user initiates a search and obtains relevant text indexes. Further, matching knowledge points and corresponding articles are retrieved based on the user's profile. Specifically, this involves calculating indicator points matching the user's profile from a set of articles related to the query, and sorting the articles associated with the knowledge points. After the user clicks on an article, they can browse the article details. When scrolling through the knowledge points in the text, they are highlighted, and related query recommendations are provided based on these knowledge points.

[0121] The process of indexing articles involves extracting articles with key knowledge points configured. This is achieved by extracting information based on key domains such as article title, paragraphs, and body text using machine learning methods (e.g., using the deep learning model BERT-MRC), with the extracted spans being the knowledge points.

[0122] Furthermore, the system retrieves matching knowledge points and corresponding articles based on the user's profile, providing personalized knowledge point recommendations. When a user initiates a search, the search results return personalized knowledge points associated with the corresponding articles. Personalization refers to recommending matching knowledge points to users based on their user profiles.

[0123] As for knowledge-point-based related query recommendations, when a user scrolls to the "knowledge point" area of ​​the main text on the landing page, popular related queries are recommended to the user based on that knowledge point.

[0124] Specifically, the annotation process for key knowledge points during article indexing can be performed using a deep learning model. First, the start and end positions of the text in the candidate content are determined to obtain input information. The search results are the data content within the candidate content. The model input can be represented as {article title, article body}, with Label representing the start and end positions of the knowledge points. Then, the input information is fed into a deep learning model to obtain the probability that the start and end positions of the text belong to a knowledge point. Knowledge point extraction can be achieved by training one or more deep learning models. Based on the probabilities of the start and end positions, the key knowledge points corresponding to the candidate content are determined. Finally, the search process for the query information is performed based on the key knowledge points corresponding to the candidate content. For example, the model outputs the start and end positions of the knowledge point text in the body text, specifically the probability P(i,j) that the position combination span is a knowledge point in the text. For instance, P(5,18)>0.5 indicates that text characters at positions 5 to 18 in the body text are knowledge points.

[0125] Furthermore, the training process for deep learning models involves sourcing the training set, requiring large-scale pre-training. The training set needed for knowledge point extraction is relatively small, estimated at 30,000 to 50,000 data points (at a level achievable through manual annotation). Since this model demands high-quality training data, manual sequence labeling can be employed. This involves first obtaining pre-defined samples; then determining the corresponding content titles; and finally, performing sequence labeling on the knowledge points within the pre-defined samples based on the content titles to obtain training samples. The deep learning model is then trained using these training samples.

[0126] Specific annotation scenarios are as follows: Figure 8 As shown, Figure 8 This is a flowchart of another data content query method provided in the embodiments of this application; the figure shows the problem (article title): how to lose weight locally, so the knowledge points are manually marked from the text: "effective exercise", "liposuction", "cryolipolysis".

[0127] In this embodiment, the loss function of the deep learning model can support the loss of multiple knowledge point extractions. Specifically, the general MRC model information extraction method predicts the start and end positions of a span, equivalent to two n-class classifications, predicting the start and end positions among n characters respectively. In this embodiment, the text may contain many knowledge points, and they may even be nested (overlapping), so this method is not suitable. Based on this, two binary classifications can be used: for each character, there are two prediction results, representing the probability that each character is the start and end position, i.e., whether it is "possible" to be the start position and whether it is "possible" to be the end position.

[0128] In addition, deep models can be a single model, such as BERT, or a combination of multiple models, such as LSTM+CRF, Bert-BiLSTM-CRF, etc. The specific model configuration depends on the actual scenario.

[0129] The following explains the personalized knowledge point recommendation process, which is based on step 302. When a user initiates a search, the search results return personalized knowledge points related to the corresponding articles. This personalization refers to recommending matching knowledge points to users based on their profiles, thus achieving a unique experience for each user. For the same query, different users will see different knowledge points and articles displayed.

[0130] Specifically, the implementation of the personalized recommendation process is as follows: Figure 9 As shown, Figure 9This is a schematic diagram illustrating another data content query method provided in this application embodiment. The diagram shows the process of determining multiple search results that match a target object. First, knowledge point matching is performed based on the query information to determine candidate knowledge points. Then, the object tags corresponding to the target object are obtained. The object tags are input into the recommendation model to obtain recommended knowledge points. Then, the recommended knowledge points are matched with the candidate knowledge points to determine key knowledge points. Finally, multiple search results that match the target object are determined based on the key knowledge points and displayed on the target interface.

[0131] Specifically, for the training process of the recommendation model, the query records corresponding to the target object can be obtained first. The specific queries can be obtained through collaborative filtering, Wide & Deep, etc. Then, the click operations of knowledge points in the query records are statistically analyzed to obtain the click data corresponding to the target object. Finally, the recommendation model is trained based on the click data.

[0132] Specifically, based on the articles related to the user's "query", the system sorts out the knowledge points (corresponding to a set of articles) that match the user's profile.

[0133] In one possible scenario, due to the lack of a large-scale training set in the early stages, a two-phase deployment approach can be adopted. Phase 1: Cold start deployment phase, collecting user clicks as the training set for phase 2. In this phase, knowledge point recommendations are not personalized but primarily employ a random exposure strategy. Phase 2 is the deployment phase of the recommendation strategy. Based on the user click logs collected in the cold start phase, click data such as "<user, clicked knowledge points>" can be obtained. Using mining algorithms and strategies, data such as "<user, knowledge points of interest>" can be obtained as the training set for the recommendation model, thereby improving the quality of the training samples and the accuracy of knowledge point identification.

[0134] In addition, regarding the query and recommendation process based on knowledge points, that is... Figure 6 When a user scrolls to the "Knowledge Points" section of the main text on the landing page, popular related queries are recommended to the user based on that knowledge point (related window B1).

[0135] Specifically, the determination of related content can also be done through an association recommendation model, the implementation process of which is as follows: Figure 10 As shown, Figure 10This is a flowchart illustrating another data content query method provided in this application embodiment. Specifically, the input to the association recommendation model is knowledge points (text fragments), and the output is articles (retrieved from historical queries). The use of the association recommendation model first involves building an inverted index and a semantic vector index using historical queries from the search engine. The inverted index is built by segmenting the query, calculating word importance, etc. Then, the semantic vector index is performed. The query is vectorized using the semantic representation model of the search engine, transforming the retrieval problem into a nearest neighbor query problem in space (e.g., using HNSW). Further, based on the index from the previous step, retrieval and ranking are performed.

[0136] Specifically, for the recall process, the results of the inverted index and semantic vector index are merged into a candidate set; while for the ranking process, the recalled candidate set is ranked in one or more rounds. The ranking model can be trained by manually annotating and clicking on data results.

[0137] By configuring the association recommendation model, users can directly click to learn more about the relevant content of the knowledge points, which increases the frequency of user interaction with data content.

[0138] As described in the above embodiments, the system obtains query information input by the target object; then, based on the query information, it performs a content search to display multiple search results matching the target object on the target interface. Each search result is configured with corresponding key knowledge points for display. These search results are obtained by matching the key knowledge points with the query information, and these key knowledge points are derived from information extraction from the data content corresponding to the search results. Furthermore, in response to the target object's selection of search results, the system displays the data content corresponding to the search results on the target interface, highlighting the key knowledge points within the data content. This achieves a targeted data query process. Because key knowledge points are used for data content querying, and the display of these key knowledge points reflects the query basis of the search results, the relevance of the retrieved content is improved, allowing users to quickly obtain the query content and thus improving the accuracy of data content querying.

[0139] To better implement the above-described solutions of the embodiments of this application, related apparatus for implementing the above solutions is also provided below. Please refer to... Figure 11 , Figure 11 This is a schematic diagram of the structure of a data content query device provided in an embodiment of this application. The data content query device 1100 includes:

[0140] The acquisition unit 1101 is used to acquire the query information input by the target object;

[0141] The query unit 1102 is used to perform content search based on the query information to display multiple search results that match the target object in the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information. The key knowledge points are obtained by extracting information from the data content corresponding to the search results.

[0142] The query unit 1102 is further configured to, in response to the target object's selection of the search result, display the data content corresponding to the search result in the target interface, and highlight the key knowledge points in the data content corresponding to the search result.

[0143] Optionally, in some possible implementations of this application, the search result is the data content in the candidate content, and the query unit 1102 is specifically used to determine the text start position and text end position in the candidate content in order to obtain input information;

[0144] The query unit 1102 is specifically used to input the input information into the deep model to obtain the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0145] The query unit 1102 is specifically used to determine the key knowledge points corresponding to the candidate content based on the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point.

[0146] The query unit 1102 is specifically used to perform a content search process for the query information based on the key knowledge points corresponding to the candidate content.

[0147] Optionally, in some possible implementations of this application, the query unit 1102 is specifically used to obtain a preset sample;

[0148] The query unit 1102 is specifically used to determine the content title corresponding to the preset sample;

[0149] The query unit 1102 is specifically used to perform sequence labeling on the knowledge points in the preset sample based on the content title in order to obtain training samples;

[0150] The query unit 1102 is specifically used to train the deep model based on the training samples.

[0151] Optionally, in some possible implementations of this application, the query unit 1102 is specifically used to perform knowledge point matching based on the query information to determine candidate knowledge points;

[0152] The query unit 1102 is specifically used to obtain the object tag corresponding to the target object;

[0153] The query unit 1102 is specifically used to input the object tag into the recommendation model to obtain recommended knowledge points;

[0154] The query unit 1102 is specifically used to match the recommended knowledge points with the candidate knowledge points to determine the key knowledge points;

[0155] The query unit 1102 is specifically used to determine multiple search results that match the target object based on the key knowledge points, and to display them on the target interface.

[0156] Optionally, in some possible implementations of this application, the query unit 1102 is specifically used to obtain the query record corresponding to the target object;

[0157] The query unit 1102 is specifically used to perform statistics based on the click operations of knowledge points in the query record in order to obtain the click data corresponding to the target object;

[0158] The query unit 1102 is specifically used to train the recommendation model based on the click data.

[0159] Optionally, in some possible implementations of this application, the method further includes:

[0160] The query unit 1102 is specifically used to obtain the operation information corresponding to the target object in the target interface of the highlighted key knowledge points;

[0161] The query unit 1102 is specifically used to display an associated window containing related content in the target interface if the operation information indicates to stop the operation, and the associated content is determined based on the key knowledge points.

[0162] Optionally, in some possible implementations of this application, the query unit 1102 is specifically used to determine candidate related content based on the key knowledge points if the operation information indicates to stop the operation;

[0163] The query unit 1102 is specifically used to obtain the search volume of the candidate related content within a preset period;

[0164] The query unit 1102 is specifically used to sort based on the search volume in order to determine the associated content;

[0165] The query unit 1102 is specifically used to display the associated window containing the associated content in the target interface.

[0166] By acquiring the query information input by the target user, and then performing a content search based on that query information, multiple search results matching the target user are displayed on the target interface. Each search result is configured with corresponding key knowledge points for display. These key knowledge points are obtained by matching the query information with the key knowledge points, which are extracted from the data content corresponding to the search results. Furthermore, in response to the target user's selection of search results, the corresponding data content is displayed on the target interface, with the key knowledge points highlighted within the data content. This achieves a targeted data query process. Because the data content is queried using key knowledge points, and the display of these key knowledge points reflects the basis of the search results, the relevance of the retrieved content is improved, allowing users to quickly obtain the query content and thus improving the accuracy of data content retrieval.

[0167] This application also provides a terminal device, such as... Figure 12 The diagram shown is a structural schematic of another terminal device provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiment of this application. The terminal can be any terminal device including mobile phones, tablet computers, personal digital assistants (PDAs), point-of-sale (POS) terminals, in-vehicle computers, etc. Taking a mobile phone as an example:

[0168] Figure 12 This is a block diagram illustrating a portion of the structure of a mobile phone related to the terminal provided in the embodiments of this application. (Reference) Figure 12 The mobile phone includes components such as a radio frequency (RF) circuit 1210, a memory 1220, an input unit 1230, a display unit 1240, a sensor 1250, an audio circuit 1260, a wireless fidelity (WiFi) module 1270, a processor 1280, and a power supply 1290. Those skilled in the art will understand that... Figure 12 The mobile phone structure shown does not constitute a limitation on the mobile phone and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0169] The following is combined Figure 12 A detailed introduction to each component of a mobile phone:

[0170] RF circuit 1210 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink information from the base station and processes it with processor 1280; additionally, it transmits uplink data to the base station. Typically, RF circuit 1210 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier (LNA), a duplexer, etc. Furthermore, RF circuit 1210 can also communicate wirelessly with networks and other devices. The aforementioned wireless communication can use any communication standard or protocol, including but not limited to Global System for Mobile Communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Message Service (SMS), etc.

[0171] The memory 1220 can be used to store software programs and modules. The processor 1280 executes various functions and data processing of the mobile phone by running the software programs and modules stored in the memory 1220. The memory 1220 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 1220 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0172] The input unit 1230 can be used to receive input numerical or character information, and generate key signal inputs related to user settings and function control of the mobile phone. Specifically, the input unit 1230 may include a touch panel 1231 and other input devices 1232. The touch panel 1231, also known as a touch screen, can collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel 1231, as well as air touch operations within a certain range on the touch panel 1231), and drive the corresponding connection devices according to a pre-set program. Optionally, the touch panel 1231 may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, and transmits the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, sends it to the processor 1280, and can receive and execute commands sent by the processor 1280. In addition, the touch panel 1231 can be implemented using various types of sensors, such as resistive, capacitive, infrared, and surface acoustic wave sensors. Besides the touch panel 1231, the input unit 1230 may also include other input devices 1232. Specifically, these other input devices 1232 may include, but are not limited to, one or more of the following: a physical keyboard, function keys (such as volume control buttons, power buttons, etc.), a trackball, a mouse, and a joystick.

[0173] The display unit 1240 can be used to display information input by the user or information provided to the user, as well as various menus of the mobile phone. The display unit 1240 may include a display panel 1241, which may optionally be configured as a liquid crystal display (LCD), organic light-emitting diode (OLED), or similar form. Further, a touch panel 1231 may cover the display panel 1241. When the touch panel 1231 detects a touch operation on or near it, it transmits the information to the processor 1280 to determine the type of touch event. Subsequently, the processor 1280 provides corresponding visual output on the display panel 1241 according to the type of touch event. Although in Figure 12 In this embodiment, the touch panel 1231 and the display panel 1241 are two separate components to realize the input and output functions of the mobile phone. However, in some embodiments, the touch panel 1231 and the display panel 1241 can be integrated to realize the input and output functions of the mobile phone.

[0174] The mobile phone may also include at least one sensor 1250, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 1241 according to the ambient light level, and the proximity sensor can turn off the display panel 1241 and / or the backlight when the phone is moved to the ear. As a type of motion sensor, an accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used for applications that recognize the phone's posture (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition-related functions (such as pedometer, taps), etc. Other sensors that may be configured in the mobile phone, such as gyroscopes, barometers, hygrometers, thermometers, and infrared sensors, will not be described in detail here.

[0175] Audio circuit 1260, speaker 1261, and microphone 1262 provide an audio interface between the user and the mobile phone. Audio circuit 1260 converts received audio data into electrical signals and transmits them to speaker 1261, where speaker 1261 converts them into sound signals for output. On the other hand, microphone 1262 converts collected sound signals into electrical signals, which are received by audio circuit 1260, converted into audio data, and then processed by processor 1280 before being transmitted via RF circuit 1210 to, for example, another mobile phone, or the audio data can be output to memory 1220 for further processing.

[0176] WiFi is a short-range wireless transmission technology. Through the WiFi module 1270, mobile phones can help users send and receive emails, browse web pages, and access streaming media, providing users with wireless broadband internet access. Although Figure 12 The WiFi module 1270 is shown, but it is understood that it is not an essential component of a mobile phone and can be omitted as needed without changing the essence of the invention.

[0177] The processor 1280 is the control center of the mobile phone, connecting various parts of the phone through various interfaces and lines. It executes software programs and / or modules stored in the memory 1220, and calls data stored in the memory 1220 to perform various functions and process data, thereby providing overall monitoring of the phone. Optionally, the processor 1280 may include one or more processing units; optionally, the processor 1280 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the aforementioned modem processor may also not be integrated into the processor 1280.

[0178] The mobile phone also includes a power supply 1290 (such as a battery) that supplies power to various components. Optionally, the power supply can be logically connected to the processor 1280 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system.

[0179] Although not shown, mobile phones may also include a camera, Bluetooth module, etc., which will not be described in detail here.

[0180] In this embodiment of the application, the processor 1280 included in the terminal also has the function of performing the various steps of the page processing method described above.

[0181] This application also provides a server; please refer to [link / reference]. Figure 13 , Figure 13 This is a schematic diagram of a server structure provided in an embodiment of this application. The server 1300 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 1322 (e.g., one or more processors) and memory 1332, and one or more storage media 1330 (e.g., one or more mass storage devices) for storing application programs 1342 or data 1344. The memory 1332 and storage media 1330 can be temporary or persistent storage. The program stored in the storage media 1330 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the server. Furthermore, the CPU 1322 may be configured to communicate with the storage media 1330 and execute the series of instruction operations in the storage media 1330 on the server 1300.

[0182] Server 1300 may also include one or more power supplies 1326, one or more wired or wireless network interfaces 1350, one or more input / output interfaces 1358, and / or one or more operating systems 1341, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0183] The steps performed by the management device in the above embodiments can be based on this Figure 13 The server structure shown.

[0184] This application also provides a computer-readable storage medium storing query instructions for data content. When executed on a computer, the computer causes the computer to perform the aforementioned actions. Figures 3 to 10 The steps performed by the data content querying device in the method described in the illustrated embodiment.

[0185] This application also provides a computer program product that includes query instructions for data content. When run on a computer, it causes the computer to perform the aforementioned actions. Figures 3 to 10 The steps performed by the data content querying device in the method described in the illustrated embodiment.

[0186] This application embodiment also provides a data content query system, which may include... Figure 11 The data content query device described in the embodiments, or Figure 12 The terminal device in the described embodiments, or Figure 13 The server described.

[0187] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0188] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0189] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0190] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0191] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a data content query device, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0192] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for querying data content, characterized in that, include: Retrieve the query information input by the target object; Based on the query information, knowledge points are matched to determine candidate knowledge points; Obtain the object tag corresponding to the target object; Input the object labels into the recommendation model to obtain recommended knowledge points; Key knowledge points are determined by matching the recommended knowledge points with the candidate knowledge points. Based on the key knowledge points, multiple search results matching the target object are determined and displayed on the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information. The key knowledge points are obtained by extracting information from the data content corresponding to the search results. The key knowledge points refer to the summary of the author's core viewpoints in the article. In response to the target object's selection of the search result, the data content corresponding to the search result is displayed on the target interface, and the key knowledge points are highlighted in the data content corresponding to the search result.

2. The method according to claim 1, characterized in that, The search results are the data content among the candidate content, and the method further includes: Determine the start and end positions of the text in the candidate content to obtain the input information; The input information is fed into a deep model to obtain the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point. The key knowledge points corresponding to the candidate content are determined based on the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point. The process of searching for content corresponding to the query information is based on the key knowledge points corresponding to the candidate content.

3. The method according to claim 2, characterized in that, The method further includes: Obtain preset samples; Determine the content title corresponding to the preset sample; Based on the content title, the knowledge points in the preset sample are sequentially labeled to obtain training samples; The deep model is trained based on the training samples.

4. The method according to claim 1, characterized in that, The method further includes: Retrieve the query records corresponding to the target object; Statistical analysis is performed on the click operations of knowledge points in the query records to obtain the click data corresponding to the target object; The recommendation model is trained based on the click data.

5. The method according to claim 1, characterized in that, The method further includes: In the target interface that highlights the key knowledge points, obtain the operation information corresponding to the target object; If the operation information indicates to stop the operation, an associated window containing related content will be displayed on the target interface. The associated content is determined based on the key knowledge points.

6. The method according to claim 5, characterized in that, If the operation information indicates to stop the operation, then a related window containing related content is displayed on the target interface, including: If the operation information indicates to stop the operation, then candidate related content is determined based on the key knowledge points; Obtain the search volume of the candidate related content within a preset period; The relevant content is determined by sorting based on the search volume. The associated window containing the associated content is displayed in the target interface.

7. A data content query device, characterized in that, include: The acquisition unit is used to acquire the query information input by the target object; A query unit is used to perform knowledge point matching based on the query information to determine candidate knowledge points; Obtain the object tag corresponding to the target object; Input the object labels into the recommendation model to obtain recommended knowledge points; Key knowledge points are determined by matching the recommended knowledge points with the candidate knowledge points. Based on the key knowledge points, multiple search results matching the target object are determined and displayed on the target interface. Each search result is configured with corresponding key knowledge points for display. The search results are obtained by matching the key knowledge points with the query information. The key knowledge points are obtained by extracting information from the data content corresponding to the search results. The key knowledge points refer to the summary of the author's core viewpoints in the article. The query unit is further configured to respond to the target object's selection of the search result, display the data content corresponding to the search result in the target interface, and highlight the key knowledge points in the data content corresponding to the search result.

8. The apparatus according to claim 7, characterized in that, The search result is the data content among the candidate content, and the query unit is specifically used for: Determine the start and end positions of the text in the candidate content to obtain the input information; The input information is fed into a deep model to obtain the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point. The key knowledge points corresponding to the candidate content are determined based on the probability that the start position of the text belongs to a knowledge point and the probability that the end position of the text belongs to a knowledge point. The process of searching for content corresponding to the query information is based on the key knowledge points corresponding to the candidate content.

9. A computer device, characterized in that, The computer device includes a processor and memory: The memory is used to store program code; the processor is used to execute the data content query method according to any one of claims 1 to 6 according to the instructions in the program code.

10. A computer program product comprising a computer program / instructions stored in a computer-readable storage medium, characterized in that, When the computer program / instructions in the computer-readable storage medium are executed by a processor, they implement the steps of the data content query method according to any one of claims 1 to 6.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the data content query method according to any one of claims 1 to 6.

Images