Page display method and device, electronic equipment and computer readable storage medium

Abstract

The application provides a page display method and device, electronic equipment and a computer readable storage medium. The method comprises: displaying a first page; detecting an operation on the displayed first page; when the detected operation comprises a jump trigger operation on a target object in the first page, determining a second page in a page set corresponding to the target object based on a mapping relationship between the object and a jump page, wherein the page set comprises reading pages of a plurality of electronic reading materials and graph pages of a knowledge graph constructed based on the plurality of electronic reading materials; and displaying the second page. Through the application, accurate and rapid page positioning and page jump display of electronic reading materials and a knowledge graph can be achieved.

Description

Technical Field

[0001] This application relates to computer technology, and more particularly to a page display method, apparatus, electronic device, and computer-readable storage medium. Background Technology

[0002] In related technologies, individual e-books are usually displayed separately for users to browse individually. In some scenarios, such as teaching scenarios, knowledge graphs are also provided to help users comprehensively understand the knowledge in the e-books. The e-books and knowledge graphs are also displayed separately.

[0003] However, in real-world scenarios, when users are browsing e-books or knowledge graphs, they may encounter objects that require more explanations for better understanding. In such cases, users need to manually consult other e-books or knowledge graphs to obtain the explanations for the objects. However, the manual consultation process is cumbersome and inconvenient, and it is difficult to accurately locate the required page, resulting in low accuracy and efficiency in page location. Summary of the Invention

[0004] To address the related technical issues, embodiments of this application provide a page display method, apparatus, electronic device, and computer-readable storage medium, which can accurately and quickly locate and display electronic publications and knowledge graphs.

[0005] The technical solution of this application embodiment is implemented as follows:

[0006] This application provides a page display method, including:

[0007] Display the first page;

[0008] Detect actions performed on the first page displayed.

[0009] When the detected operation includes a jump trigger operation targeting a target object in the first page, the second page in the page set corresponding to the target object is determined based on the mapping relationship between the object and the jump page; the page set includes reading pages of multiple e-books and graph pages of the knowledge graph constructed based on the multiple e-books;

[0010] The second page will be displayed.

[0011] In the above scheme, determining the second page in the page set corresponding to the target object based on the mapping relationship between the object and the redirected page includes: determining at least two candidate pages in the page set corresponding to the target object based on the mapping relationship between the object and the redirected page; generating selection information; the selection information is used to indicate the selection operation for the at least two candidate pages; receiving the selection operation for the at least two candidate pages, and using the selected candidate page as the second page.

[0012] The above scheme further includes: obtaining the mapping relationship between the saved object and the redirected page.

[0013] In the above scheme, the method further includes: obtaining multiple objects extracted from the reading pages of the multiple electronic reading materials; determining the jump page corresponding to each of the multiple objects from the reading pages of the multiple electronic reading materials and the graph page corresponding to the knowledge graph; constructing a mapping relationship between the objects and the jump pages based on the multiple objects and the jump pages corresponding to each object, and saving the constructed mapping relationship between the objects and the jump pages.

[0014] In the above scheme, determining the jump page corresponding to each object from the reading pages of the multiple electronic books and the graph pages corresponding to the knowledge graph includes: for each object, determining the reading page that satisfies the association condition with the corresponding object from the reading pages of the multiple electronic books, and using the determined reading page as the jump page corresponding to the corresponding object; and / or, for each object, determining the graph page whose page content contains the corresponding object from the graph pages corresponding to the knowledge graph, and using the determined graph page as the jump page corresponding to the corresponding object.

[0015] In the above scheme, when the corresponding object contains entities, determining the reading pages that satisfy the association condition with the corresponding object from the reading pages of the plurality of electronic reading materials includes: counting the number of times the corresponding object appears in each reading page of the plurality of electronic reading materials to obtain the number of entities of the corresponding object contained in each reading page; selecting reading pages from the reading pages of the plurality of electronic reading materials whose number of entities satisfies the quantity condition, and using the selected reading pages as the reading pages that satisfy the association condition with the corresponding object.

[0016] In the above scheme, when the object contains entities, determining the pages from the plurality of electronic publications that satisfy the association conditions with the object includes: counting the number of times the entity appears in each page of the plurality of electronic publications to obtain the number of entities of the entity contained in each page; obtaining the weight corresponding to each of the plurality of electronic publications; scoring the corresponding pages based on the weight of each electronic publication and the number of entities of the entity contained in each page to obtain the score of the corresponding page; and selecting the pages from the plurality of electronic publications whose scores meet the scoring conditions, and using the selected pages as the pages that satisfy the association conditions with the corresponding object.

[0017] In the above scheme, when the corresponding object contains an image, determining the page that satisfies the association condition with the corresponding object from the pages of the plurality of electronic publications includes: extracting the page image from each page of the plurality of electronic publications to obtain at least one page image; determining the similarity between the corresponding object and the corresponding page image for each page image; and taking the page image whose similarity satisfies the similarity condition as the page that satisfies the association condition with the corresponding object.

[0018] In the above scheme, the method further includes: obtaining the page content corresponding to the reading pages of the plurality of electronic reading materials; extracting entities from the page content to obtain a plurality of entities; extracting entity relationships from the page content to obtain entity relationships between the plurality of entities; constructing the knowledge graph based on the plurality of entities and the entity relationships between the plurality of entities; and generating corresponding graph pages based on the knowledge graph.

[0019] In the above scheme, the extraction of entity relationships from the page content includes: extracting semantic features from the page content using a bidirectional encoder representing a BERT model through a converter to obtain corresponding semantic features; and extracting entity relationships from the page content based on the semantic features using an entity relationship extraction model.

[0020] This application also provides a page display device, including:

[0021] The first display module is used to display the first page;

[0022] The detection module is used to detect operations performed on the first page of the display.

[0023] The determination module is used to determine the second page in the page set corresponding to the target object based on the mapping relationship between the object and the page when the detected operation includes a jump trigger operation for the target object in the first page; the page set includes reading pages of multiple electronic books and graph pages of the knowledge graph constructed based on the multiple electronic books;

[0024] The second display module is used to display the second page.

[0025] This application also provides an electronic device, including:

[0026] Memory, used to store executable instructions;

[0027] The processor, when executing executable instructions stored in the memory, implements the page display method provided in the embodiments of this application.

[0028] This application also provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, implement the page display method provided in this application.

[0029] In this embodiment, during the display of the first page, operations on the first page are detected. When the detected operation includes a jump trigger operation on a target object in the first page, the second page corresponding to the target object is determined and displayed based on the mapping relationship between the object and the jump page. Through the mapping relationship between the object and the jump page, the reading page of the electronic book and the graph page of the knowledge graph are associated, realizing accurate and fast location and jump display of the pages of the electronic book and the knowledge graph, thereby facilitating users to jump and look up information. Attached Figure Description

[0030] Figure 1 This is an optional architecture diagram of the page display system 100 provided in this application embodiment;

[0031] Figure 2 This is an optional structural schematic diagram of the electronic device 200 provided in the embodiments of this application;

[0032] Figure 3 This is an optional flowchart illustrating the page display method provided in the embodiments of this application;

[0033] Figure 4 This is an optional schematic diagram of a reading page provided in an embodiment of this application;

[0034] Figure 5 This is an optional schematic diagram of the atlas page provided in an embodiment of this application;

[0035] Figure 6This is an optional flowchart illustrating the page display method provided in the embodiments of this application;

[0036] Figure 7 This is an optional schematic diagram of the entity relationship extraction model provided in the embodiments of this application;

[0037] Figure 8 This is an optional schematic diagram of the BERT model provided in an embodiment of this application;

[0038] Figure 9 This is an optional structural diagram of a combination of the BERT model and the entity relationship extraction model provided in an embodiment of this application;

[0039] Figure 10 This is an optional flowchart illustrating the page display method provided in the embodiments of this application;

[0040] Figure 11 This is an optional flowchart illustrating the page display method provided in the embodiments of this application;

[0041] Figure 12 This is an optional interface diagram of page navigation provided in the embodiments of this application;

[0042] Figure 13 This is an optional interface diagram of page navigation provided in the embodiments of this application;

[0043] Figure 14 This is an optional interface diagram of page navigation provided in the embodiments of this application;

[0044] Figure 15 This is an optional schematic diagram of the page display method provided in the embodiments of this application;

[0045] Figure 16 This is an optional schematic diagram illustrating the knowledge graph construction process provided in this application embodiment;

[0046] Figure 17 This is an optional schematic diagram of the Canopy+K-means algorithm provided in the embodiments of this application. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0049] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0051] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0052] 1) An entity refers to an objectively existing object or thing in the real world that can be distinguished from one another. It can be a specific person or thing, or an abstract concept or relationship.

[0053] 2) Knowledge graph, known in the library and information science field as knowledge domain visualization or knowledge domain mapping map, is a series of different graphics that show the development process and structural relationships of knowledge. It uses visualization technology to describe knowledge resources and their carriers, and to mine, analyze, construct, draw and display knowledge and the interrelationships between them.

[0054] 3) Term Frequency-Inverse Document Frequency (TF-IDF): This is a statistical method, a commonly used weighting technique in information retrieval and data mining, used to assess the importance of a word to a document within a document set or corpus. The importance of a word increases proportionally to the number of times it appears in the document, but decreases inversely proportionally to its frequency in the corpus.

[0055] This application provides a page display method, apparatus, electronic device, and computer-readable storage medium, which can accurately and quickly locate and display electronic publications and knowledge graphs.

[0056] First, the page display system provided in the embodiments of this application will be described, see [link to relevant documentation]. Figure 1 , Figure 1 This is an optional architecture diagram of the page display system 100 provided in this application embodiment. To support a page display application, the terminal 103 connects to the server 101 via the network 102. In some embodiments, the terminal 103 may be a laptop, tablet, desktop computer, smartphone, dedicated messaging device, portable gaming device, smart speaker, smartwatch, etc., but is not limited to these. The server 101 may be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides 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, content delivery network (CDN) services, and big data and artificial intelligence platforms. The network 102 may be a wide area network or a local area network, or a combination of both. The terminal 103 and the server 101 may be directly or indirectly connected via wired or wireless communication, which is not limited in this application embodiment.

[0057] Terminal 103 is used to send a download request carrying the electronic book identifier to server 101 in response to the download operation of the electronic book.

[0058] Server 101 is used to receive download requests sent by terminal 103, obtain the electronic publication identifier carried in the download request, download the corresponding electronic publication based on the electronic publication identifier, and send the downloaded electronic publication to terminal 103.

[0059] The electronic device for implementing the above-described page display method, as provided in the embodiments of this application, will now be described. See [link to relevant documentation]. Figure 2 , Figure 2 This is an optional structural diagram of the electronic device 200 provided in this application embodiment. In practical applications, the electronic device 200 can be implemented as follows: Figure 1 Terminal 103 or server 101 in the middle, with electronic devices as Figure 1 Taking the terminal 103 shown as an example, the electronic device implementing the page display method of this application embodiment will be described. Figure 2 The illustrated electronic device 200 includes at least one processor 201, a memory 205, at least one network interface 202, and a user interface 203. The various components in the electronic device 200 are coupled together via a bus system 204. It is understood that the bus system 204 is used to implement communication between these components. In addition to a data bus, the bus system 204 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 2 The general labeled all buses as Bus System 204.

[0060] Processor 201 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0061] User interface 203 includes one or more output devices 2031 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 203 also includes one or more input devices 2032, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0062] The memory 205 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 205 may optionally include one or more storage devices physically located away from the processor 201.

[0063] The memory 205 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 205 described in this application embodiment is intended to include any suitable type of memory.

[0064] In some embodiments, the memory 205 can store data to support various operations. Examples of such data include programs, modules, and data structures, or subsets or supersets thereof. In this embodiment, the memory 205 stores an operating system 2051, a network communication module 2052, a presentation module 2053, an input processing module 2054, and a page display device 2055. Specifically...

[0065] Operating system 2051 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, and driver layer, for implementing various basic business functions and handling hardware-based tasks.

[0066] The network communication module 2052 is used to reach other computing devices via one or more (wired or wireless) network interfaces 202, such as Bluetooth, WiFi, and Universal Serial Bus (USB).

[0067] The presentation module 2053 is configured to enable the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 2031 (e.g., a display screen, a speaker, etc.) associated with the user interface 203;

[0068] The input processing module 2054 is used to detect and translate one or more user inputs or interactions from one or more input devices 2032.

[0069] In some embodiments, the page display device provided in this application can be implemented in software. Figure 2 A page display device 2055 stored in memory 205 is shown. This device can be software in the form of programs and plug-ins, and includes the following software modules: a first display module 20551, a detection module 20552, a determination module 20553, and a second display module 20554. These modules are logically connected and can therefore be arbitrarily combined or further separated according to their implemented functions. The functions of each module will be described below.

[0070] In other embodiments, the page display device provided in this application can be implemented in hardware. As an example, the page display device provided in this application can be a processor in the form of a hardware decoding processor, which is programmed to execute the page display method provided in this application. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0071] The following will be based on Figure 1 Taking the aforementioned terminal as an example, and combining the exemplary applications and implementations provided in the embodiments of this application, the page display method applied to electronic devices provided in the embodiments of this application is described.

[0072] See Figure 3 , Figure 3 This is an optional flowchart illustrating the page display method provided in the embodiments of this application, which will be combined with... Figure 3 The steps shown are explained.

[0073] Step 301: The terminal displays the first page;

[0074] Step 302: Detect operations performed on the first displayed page;

[0075] Step 303: When the detected operation includes a jump trigger operation for the target object in the first page, determine the second page in the page set corresponding to the target object based on the mapping relationship between the object and the jump page;

[0076] Step 304: Display the second page.

[0077] It should be noted that the first page is a page within a page collection. This collection can include pages from multiple e-books and graph pages representing knowledge graphs built upon these e-books. Here, e-books can be learning materials, and their file types can be, but are not limited to, text documents (txt), Word documents, Excel spreadsheets, Portable Document Format (PDF) documents, Microsoft Office PowerPoint presentations (PPT), and HyperText Markup Language (HTML) files. The number of knowledge graphs built upon multiple e-books can be one or more. It should be understood that there cannot be two entities with an entity relationship between two knowledge graphs.

[0078] It should be noted that a page in a page collection can include multiple objects, and the types of objects can be various, including but not limited to text, images, audio, and video. For example, see [link to example]. Figure 4 and Figure 5 , Figure 4 This is an optional schematic diagram of a reading page provided in an embodiment of this application. Figure 5 This is an optional schematic diagram of the atlas page provided in an embodiment of this application. Wherein, Figure 4 The shown page contains text 401 and image 402. Figure 5The knowledge graph shown on the graph page includes entities 501 to 504. Entity 501 contains four attributes. For example, attribute 505 indicates that the unit attribute of entity 501's "friction force" is "Newtons (N)". The entity relationship 506 between entities 501 and 502 is "Ff = μFn", where Ff is entity 501 ("friction force"), μ is entity 502 ("coefficient of kinetic friction"), and Fn is "pressure". Entities, entity attributes, and entity relationships are all represented using text.

[0079] In some embodiments, the graph pages in the page set are generated based on the reading pages in the page set. Specifically, the terminal generates a knowledge graph based on the reading pages and generates corresponding graph pages based on the generated knowledge graph, so that the graph pages in the page set correspond to the content of the reading pages, so that the corresponding content can be directly obtained when jumping between reading pages and graph pages. See [link to specific documentation]. Figure 6 , Figure 6 This is an optional flowchart illustrating the page display method provided in this application embodiment. Before step 301, the following steps may also be performed:

[0080] Step 601: The terminal obtains the page content corresponding to the reading pages of multiple electronic books;

[0081] Step 602: Extract entities from the page content to obtain multiple entities;

[0082] Step 603: Extract entity relationships from the page content to obtain entity relationships between multiple entities;

[0083] Step 604: Construct a knowledge graph based on multiple entities and the entity relationships between them;

[0084] Step 605: Generate the corresponding knowledge graph page based on the knowledge graph.

[0085] In practical implementation, the terminal extracts page content from multiple e-book pages to obtain the corresponding page content. In this embodiment, the terminal extracts text from the e-book pages and uses the extracted text as the page content. In some embodiments, the terminal also performs image recognition on images in the e-book pages to identify text in the images and uses the identified text as the page content. In some embodiments, the terminal also performs speech recognition on audio in the e-book pages, converts the audio into text, and uses the obtained text as the page content. In some embodiments, the terminal also performs speech and image recognition on video in the e-book pages. Specifically, the terminal extracts audio from the video, performs speech recognition on the audio, and obtains the text after speech recognition. The terminal extracts image frames from the video, performs image recognition on the extracted image frames, and obtains the text in the image frames. The terminal uses the text identified from the video as the page content.

[0086] In practical applications, the terminal performs entity extraction on the obtained page content to obtain all entities contained in all pages of multiple e-books. In this embodiment, the terminal can use regular expression matching to extract entities from the page content. Specifically, it uses a regular expression containing multiple entities to match the page content, obtains the text that matches the regular expression, and uses this text as the extracted entity. In some embodiments, the terminal can also use an entity extraction model to extract entities from the page content. This entity extraction model is pre-trained and can be trained using sample text carrying entity labels. For example, the sample text is input into the entity extraction model, and the entity extraction model extracts entities from the sample text to obtain the corresponding predicted entities. Based on the difference between the predicted entities and the entity labels, the model parameters of the entity extraction model are updated, thereby completing the training of the entity extraction model.

[0087] For example, see Figure 4 ,from Figure 4 The entities extracted include "friction", "object", "relative motion", "contact surface", "direction", "static friction", "velocity", "mechanical energy", "thermal energy", etc.

[0088] After extracting multiple entities, the terminal needs to extract entity relationships from the page content to obtain the entity relationships between the multiple entities. For example, there is a relationship between the entities of friction and pressure: the frictional force Ff = μFn (μ is the coefficient of friction, and Fn is the pressure). In this embodiment, the terminal needs to extract the entity relationship "Ff = μFn" between friction and pressure from the page content involving entities such as friction and pressure and the relationship between them.

[0089] Based on this, in some embodiments, based on Figure 6 Step 203 can also be implemented as follows: The terminal extracts semantic features from the page content using the Bidirectional Encoder Representation from Transformers (BERT) model of the converter to obtain the corresponding semantic features; and extracts entity relationships from the page content based on the semantic features using the entity relationship extraction model.

[0090] In practice, the terminal combines the BERT model and the entity relationship extraction model to extract entity relationships from the page content. See also Figure 7 , Figure 7 This is an optional schematic diagram of the entity relation extraction model provided in this application embodiment. Here, the entity relation extraction model adopts the attention-based convolutional neural network (ATT-CNN) model. The ATT-CNN model is an improvement on the convolutional neural network (CNN) model. Specifically, the ATT-CNN model introduces an attention layer between the input layer and the convolutional layer of the CNN. It creates a context vector for each word. After concatenating the context vector with the word vector, the concatenated vector is used as the new vector representation of the word, and then it is input into the convolutional layer. The idea of ​​the attention mechanism is to learn to focus attention on specific important words when deriving the context vector gi of the word vector Xi. The attention mechanism is another multi-layer perceptron neural network (MLP neural network), which is trained together with all layers of the ATT-CNN model. When predicting the sentence category, the attention mechanism will determine which words should receive more attention than other words in the sentence. The scored words are combined by weighted summation, and the weighted summation method is shown in formula (1):

[0091] g j =∑ j′＝i α i,j ·x j (1)

[0092] Where, α i,j For attention weights, g i It is a context vector, x j It refers to word vectors. In practical implementation, α is required. i,j ≥0, and ∑ j α i,j ·x j=1, here α can be normalized using softmax. i,j ·x j The values ​​are mapped to (0,1) and the sum of these values ​​is made equal to 1. Furthermore, the attention mechanism is described by equations (2) and (3):

[0093]

[0094]

[0095] The score is calculated by an MLP. Simply put, an MLP is used to simulate word pairs (x... j ,x j,j′＝i The correlation between x and y. And those x with larger score values. j,j′＝i In the context vector g i It holds more weight in the middle.

[0096] Reference Figure 8 , Figure 8 This is an optional schematic diagram of the BERT model provided in this application embodiment. The BERT model structure consists of an encoder and decoder composed of multiple layers of transformers. Here, the transformers... Figure 8 The `Trm` in the text indicates that the Transformer used here is implemented based on a multi-head attention mechanism. The BERT model is bidirectional, using a bidirectional Transformer Encoder to generate a language model. The Transformer Encoder reads the entire text sequence at once, rather than reading it sequentially from left to right or right to left. This bidirectional structure allows the model to learn based on both sides of a word. That is, the BERT model can use the preceding and following information of a word to obtain a better word vector representation. Through this bidirectional structure, the model can comprehensively learn the contextual information of a word in the sentence in which the word is located, thereby learning the semantics of the word more accurately.

[0097] In this embodiment, entity relations are extracted by combining the BERT model and the entity relation extraction model. The following will describe the process. Figure 9 This section explains the process of extracting entity relationships by combining the BERT model and an entity relationship extraction model. (See also...) Figure 9 , Figure 9This is an optional structural diagram illustrating the combination of a BERT model and an entity relation extraction model provided in this application embodiment. Here, a BERT model is introduced before the entity relation extraction model. In actual implementation, the terminal inputs the page content and the entities extracted in step 202 into the BERT model. The BERT model extracts semantic features from the sentences in the page content, obtaining the semantic features of the sentences and the semantic features of the entities. It should be understood that the semantic features of the sentences and the semantic features of the entities obtained after processing the page content through the BERT model are more accurate. Next, the terminal inputs the semantic features of the sentences and the semantic features of the entities output by the BERT model into the entity relation extraction model. The entity relation extraction model extracts the entity relations between different entities, thus completing the extraction of entity relations.

[0098] It should be noted that before using the BERT model and entity relationship extraction model to extract entity relationships in this embodiment, both need to be trained. In this embodiment, the BERT model used is a pre-trained model. Downstream tasks can be implemented through two methods: fine-tuning (changing the model parameters of the pre-trained model) or feature extraction (not changing the model parameters of the pre-trained model, but using the output of the pre-trained model as features input to the downstream task). In this embodiment, the terminal fine-tunes the BERT model to better adapt to the application scenario of this embodiment. Specifically, the terminal collects a large amount of page content from reading materials and inputs it into the BERT model, enabling the BERT model to learn the textual characteristics of the reading materials and generate word embedding vectors. These word embedding vectors are stored in a list format, resulting in a word embedding table. The word embedding table is essentially a dictionary for a large number of reading materials, containing the mapping relationship between words and their corresponding word embedding vectors. Here, the word embedding vector is a representation of the word's features. When encoding input text, the BERT model can encode the text by querying the word embedding vectors corresponding to the words in the text based on the word embedding table. In this embodiment, the BERT model is fine-tuned by collecting page content corresponding to a large number of e-book pages, making the BERT model more suitable for e-book application scenarios. This improves the accuracy of semantic feature extraction from page content using the BERT model, that is, it improves the accuracy of entity relation extraction.

[0099] Next, the terminal simultaneously trains the fine-tuned BERT model and the entity relation extraction model. Here, the terminal uses two sample entities, the sample sentences to which the two sample entities belong, and the entity relation between the two sample entities as training samples to train the BERT model and the entity relation extraction model. Specifically, the terminal inputs the two sample entities and the sample sentences into the BERT model, and the BERT model and the entity relation extraction model extract the entity relation between the two sample entities to obtain the predicted entity relation. The terminal then calculates the error between the predicted entity relation and the sample entity relation, and updates the model parameters of the BERT model and the entity relation extraction model based on this error. Here, the sample entity relation is the standard entity relation between the two sample entities. This standard entity relation can be manually determined and used as the sample entity relation, so that the BERT model and the entity relation extraction model can be trained using the sample entity relation and the sample entities as training samples.

[0100] In practice, the terminal constructs a knowledge graph based on multiple entities and the relationships between them. For example, see [link to example]. Figure 5 , Figure 5 The entity relationship between "friction" and "pressure" is "Ff = μFn", and the entity relationship between "friction" and "Newton" is that the concept of "friction" was proposed by Newton. Here, arrows are used to link the two entities with this relationship. In actual implementation, after constructing the knowledge graph, the terminal generates corresponding graph pages based on the knowledge graph. For example, see [link to example]. Figure 5 ,here, Figure 5 The page content is a knowledge graph. In practice, the terminal can also generate corresponding graph pages based on page parameters. Page parameters may include page size parameters, font of page text, etc. It should be noted that in practical applications, the content of independent knowledge graphs generated based on multiple e-books is often very large, and the terminal's display interface may find it difficult to fully accommodate the entire knowledge graph while displaying it clearly. In this embodiment, the terminal can segment a complete knowledge graph to generate multiple graph pages. In some embodiments, the terminal can also generate a corresponding graph page based on a single knowledge graph, and achieve a complete display of the knowledge graph through page scaling and panning.

[0101] In the above embodiments, by combining the BERT model and the entity relationship extraction model to extract entity relationships from the page content, the extracted entity relationships can be made more accurate.

[0102] In this embodiment, the terminal displays multiple selection options for electronic publications and knowledge graphs on its display interface. In response to a selection operation on a selection option, the selected electronic publication or knowledge graph is displayed. In actual implementation, the terminal displays a first page on the display interface. Specifically, the terminal displays the first page within a first display area of ​​the display interface. The size and position of the first display area can be preset, and this embodiment does not impose specific limitations on this.

[0103] Here, the second page refers to a page in the page set. The mapping relationship between objects and redirected pages can be stored in the form of a database table. This table can be pre-stored locally or in an external storage device connected to the terminal. In practice, the terminal detects operations on the displayed first page. Specifically, the terminal periodically detects operations on the first page according to a detection cycle. This detection cycle can be set to a short duration to achieve real-time detection.

[0104] In this embodiment, when the terminal detects an operation that includes a jump-triggered operation targeting a target object on a first page, a second page mapped to the target object is determined based on the mapping relationship between the object and the jump page. In real-world scenarios, the jump-triggered operation can be clicking on an object on the page, or swiping on the object, etc. This embodiment does not specifically limit the specific method of the jump-triggered operation.

[0105] In some embodiments, before displaying a page, the terminal also constructs a mapping relationship between objects and redirected pages. This allows the terminal to obtain the redirected page (i.e., the second page) based on the mapping relationship when a redirection trigger operation is detected during page display. Specifically, see [link to documentation]. Figure 10 , Figure 10 This is an optional flowchart illustrating the page display method provided in the embodiments of this application, and it can also be executed as follows:

[0106] Step 1001: The terminal obtains multiple objects extracted from the pages of multiple electronic publications;

[0107] Step 1002: Determine the jump page corresponding to each object among multiple objects from the reading pages of multiple electronic books and the graph pages corresponding to the knowledge graph;

[0108] Step 1003: Based on multiple objects and the corresponding jump page for each object, construct a mapping relationship between objects and jump pages, and save the constructed mapping relationship between objects and jump pages.

[0109] In practical implementation, the terminal extracts objects from the pages of multiple electronic publications to obtain multiple extracted objects. In this embodiment, the objects include entities and images. The terminal extracts entities from the pages of multiple electronic publications to obtain the corresponding entities, and also extracts images from the pages of the electronic publications.

[0110] Here, the knowledge graph is constructed based on entities from the pages of multiple e-books. The corresponding graph pages are generated based on the knowledge graphs themselves; it should be understood that the content of each graph page is the content of the corresponding knowledge graph. In practice, the terminal determines the corresponding page for each object from the pages of the multiple e-books and the graph pages corresponding to the knowledge graphs.

[0111] In some embodiments, based on Figure 10 Step 302 can also be implemented in the following ways: For each object, the terminal determines the reading page that meets the association condition with the corresponding object from the reading pages of multiple electronic reading materials, and uses the determined reading page as the jump page corresponding to the corresponding object; and / or, for each object, the terminal determines the graph page whose page content contains the corresponding object from the graph page corresponding to the knowledge graph, and uses the determined graph page as the jump page corresponding to the corresponding object.

[0112] In some embodiments, when the corresponding object contains entities, determining the reading pages that satisfy the association condition with the corresponding object from the reading pages of multiple electronic reading materials can also be achieved in the following way: the terminal counts the number of times the corresponding object appears in each reading page of the multiple electronic reading materials to obtain the number of entities of the corresponding object contained in each reading page; from the reading pages of the multiple electronic reading materials, the reading pages whose number of entities meets the quantity condition are selected, and the selected reading pages are regarded as the reading pages that satisfy the association condition with the corresponding object.

[0113] In this embodiment, objects containing entities are denoted as entity objects, and objects containing images are denoted as image objects. Here, the terminal counts the number of times all entity objects appear on each reading page to obtain the number of entities for each entity object on the reading page. For example, see... Figure 4 , Figure 4 The entity "friction" appears 7 times on the page shown, meaning there are 7 entities of the entity "friction" on the page.

[0114] In practice, for a given entity object, the terminal obtains the number of entities of that entity object in each reading page through the aforementioned statistics, and selects the reading page whose entity number meets the quantity condition from all reading pages as the reading page that satisfies the association condition with the entity object. In some embodiments, the terminal selects the reading page with the largest number of entities as the reading page that satisfies the association condition with the entity object. In some embodiments, the terminal can also sort the number of entities of that entity object in multiple reading pages, and select the reading page whose entity number sorting meets the sorting condition as the reading page that satisfies the association condition with the entity object. The sorting condition could be, for example, the top 3 reading pages. Then, the reading page determined by the terminal is used as the jump page corresponding to the entity object.

[0115] In some embodiments, when the corresponding object contains an entity, determining the reading pages that satisfy the association condition with the corresponding object from the reading pages of multiple electronic reading materials can also be achieved in the following manner: the terminal counts the number of times the entity appears in each reading page of the multiple electronic reading materials to obtain the number of entities of the entity contained in each reading page; obtains the weight corresponding to each electronic reading material; scores the corresponding reading page based on the weight corresponding to each electronic reading material and the number of entities of the entity contained in each reading page to obtain the score of the corresponding reading page; selects the reading pages whose scores satisfy the scoring condition from the reading pages of the multiple electronic reading materials, and uses the selected reading pages as the reading pages that satisfy the association condition with the corresponding object.

[0116] In practice, the terminal also categorizes e-books, determines their weights based on the categorization, and selects pages from the e-books as the redirect pages associated with the physical objects based on these weights. Here, the weights of e-books can be set according to their categories; for example, textbooks could have a weight of 50%, PowerPoint presentations 30%, and other materials 20%.

[0117] In practical implementation, the terminal can also score the reading page based on the number of entities of the entity objects contained in the reading page and the weight of the electronic reading book to which the reading page belongs, and obtain a score for the reading page. Based on the score, the terminal selects the corresponding reading page from multiple reading pages as the reading page that meets the association condition with the entity objects. For example, the terminal can score the reading page using formula (4):

[0118] score(p,x)=f(w p ,q p,x (4)

[0119] Where score(p,x) is the score given by entity x to page p of the reading material, and w p The weight of the e-book to which page p belongs, q p,x Let f be the number of entity objects x contained in page p of the reading material. Here, the function f can be set according to the actual scenario; for example, it can be set as a linear function, such as f(w). p ,q p,x )=a*w p +b*q p,x +c, where a, b, and c are all real numbers, and a and b are both greater than or equal to 0. This application does not impose specific limitations on the function f.

[0120] In practice, when a terminal obtains ratings for a given entity from multiple electronic reading pages, it selects the reading page whose rating meets the rating criteria from among the multiple reading pages as the reading page that satisfies the association criteria with the entity.

[0121] In some embodiments, the terminal further determines, for each entity object, a graph page whose content contains the entity object from the graph page corresponding to the knowledge graph, and uses the determined graph page as the jump page corresponding to the entity object.

[0122] In some embodiments, when the corresponding object contains an image, determining the page that satisfies the association condition with the corresponding object from the pages of the plurality of electronic publications can also be achieved in the following manner: the terminal extracts the page image from each page of the plurality of electronic publications to obtain at least one page image; for each page image, the similarity between the corresponding object and the corresponding page image is determined; and the page image whose similarity satisfies the similarity condition is taken as the page that satisfies the association condition with the corresponding object.

[0123] In practical implementation, the terminal calculates the similarity between page images in different reading materials to obtain the similarity between the page images. Here, the terminal can calculate the similarity between two page images by calculating the similarity of their pixels, or by using histogram matching, or by using an image matching model to classify the similarity between two page images and determine whether the similarity between the two page images meets the similarity condition. The image matching model is a classifier that can be pre-trained, and the classification type can include satisfying the similarity condition and not satisfying it. The terminal identifies reading materials pages that satisfy the similarity condition as reading materials pages that satisfy the association condition with the image objects. Here, the similarity condition can be the maximum similarity. In some embodiments, the similarity condition can also be set to a certain range, such as ranking the top 3 in similarity, etc. This application embodiment does not specifically limit the similarity condition.

[0124] In some embodiments, when the corresponding object contains an image, determining the page from the plurality of electronic publications that satisfies the association condition with the corresponding object can also be achieved as follows: the terminal extracts page images from each page of the plurality of electronic publications to obtain at least one page image; for each page image, the similarity between the corresponding object and the corresponding page image is determined; the weight corresponding to each electronic publication is obtained; based on the weight corresponding to each electronic publication and the similarity between the corresponding object and the corresponding page image, the page corresponding to the corresponding page image is scored to obtain a score for the corresponding page; from the pages of the plurality of electronic publications, the page whose score satisfies the scoring condition is selected, and the selected page is taken as the page that satisfies the association condition with the corresponding object.

[0125] In practical implementation, after obtaining the similarity between the image object and the corresponding page image, the terminal scores the page to which the corresponding page image belongs based on the similarity and the weight of the e-book to which the page image belongs, thus obtaining a score for the page. The terminal then selects the appropriate page based on the score. The specific scoring process is detailed in the embodiments described above and will not be repeated here.

[0126] In practice, the terminal constructs a mapping relationship between objects and corresponding redirect pages based on multiple objects and the redirect page corresponding to each object. Here, one object can correspond to one or more redirect pages.

[0127] For example, if the e-books include e-book 1 and e-book 2, and the knowledge graph pages constructed based on e-book 1 and e-book 2 include knowledge graph page 1 and knowledge graph page 2, for a certain entity object, among all the pages in e-book 1 and e-book 2, the pages where the number of entities satisfies the quantity condition are e-book page 3 of e-book 1 and e-book page 2 of e-book 2, and the knowledge graph page containing this entity object is knowledge graph page 1, then the terminal constructs and saves the mapping relationship between the entity object and e-book page 3 of e-book 1, e-book page 2 of e-book 2, and knowledge graph page 1. If for a certain image object, the page corresponding to the page image that satisfies the similarity condition is e-book page 1 of e-book 1, then the terminal constructs the mapping relationship between the image and e-book page 1 of e-book 1.

[0128] In step 303, when the terminal detects an operation that includes a jump trigger operation targeting a target object on the first page, the terminal obtains the saved mapping relationship between the object and the jump page, and determines the second page mapped to the target object based on this mapping relationship. Specifically, the terminal obtains the saved mapping relationship between the object and the jump page according to the storage address of the mapping relationship between the object and the jump page.

[0129] In some embodiments, in the mapping relationship between objects and redirected pages, one object may correspond to one or more pages. In this embodiment, the pages corresponding to objects in the mapping relationship are described as candidate pages. When the terminal detects an operation that includes a redirection trigger operation for a target object in the first page, the terminal uses the target object as an index to query the mapping relationship between the object and the redirected page. When there is one and only one candidate page that maps to the target object, the terminal uses that candidate page as the second page.

[0130] In some embodiments, when the target object corresponds to multiple candidate pages, it is also necessary to further determine the second page to be redirected from the multiple candidate pages. The process of determining the second page from the multiple candidate pages is described in [link to documentation]. Figure 11 , Figure 11 This is an optional flowchart illustrating the page display method provided in this application embodiment. The specific implementation of step 303 may include:

[0131] Step 1101: When the terminal detects an operation that includes a jump trigger operation for a target object in the first page, at least two candidate pages in the page set corresponding to the target object are determined based on the mapping relationship between the object and the jump page.

[0132] Step 1102: Generate selection information, which is used to indicate the selection operation for the at least two candidate pages;

[0133] Step 1103: Receive a selection operation for at least two candidate pages, and use the selected candidate page as the second page.

[0134] Here, the target object has a mapping relationship with at least two candidate pages. In this embodiment, the at least two candidate pages may simultaneously include a reading material page and a map page.

[0135] In practice, the terminal generates selection information for at least two candidate pages. This selection information can include page selection function items, which indicate the user's choice between the candidate pages. Users can select candidate pages based on these selection function items. These selection function items can be displayed as a dropdown list, showing the page information of at least two candidate pages. The number of selection function items can also be the same as the number of candidate pages, with one selection function item corresponding to one candidate page. Users can select the corresponding candidate page by clicking the appropriate selection function item.

[0136] For example, see Figure 12 , Figure 12 This is an optional interface diagram of page navigation provided in the embodiments of this application. Figure 12 This illustrates an example of navigating from one reading page to another. When the terminal detects a navigation trigger operation for target object 121, a page selection interface 122 is displayed, along with page selection function items. Here, the target object corresponds to two candidate pages: reading page 1 and map page 1. The page selection function items include page selection function item 123 for navigation to reading page 1 and page selection function item 124 for navigation to map page 1. The selection information is "Jump to reading page 1" displayed in page selection function item 123 and "Jump to map page 1" displayed in page selection function item 124. The user can then select a candidate page based on this selection information.

[0137] For example, see Figure 12 When the terminal receives a selection operation that triggers the selection function item 123, it means that the reading page 1 is selected as the second page. When the terminal receives a selection operation that triggers the selection function item 124, it means that the map page 1 is selected as the second page.

[0138] In practice, the terminal displays the second page in the second display area of ​​the display interface. Here, the second display area can be the same as the first display area; the terminal cancels the display of the first page in the first display area and displays the second page in the second display area. In some embodiments, the second display area can also be a different display area from the first display area; the terminal displays the first page in the first display area and the second page in the second display area.

[0139] For example, see Figure 12 Here, the terminal displays page 1 of the reading material associated with target object 121, and the terminal directly displays the second page within the display interface that displays the first page. In some embodiments, the terminal also presents a return function item within the second page, for example... Figure 12 The terminal responds to the trigger operation of the return function 125 by returning to the first page, that is, by re-displaying the first page between page jumps in the display interface.

[0140] For example, see Figure 13 , Figure 13 This is an optional interface diagram of page navigation provided in the embodiments of this application. Figure 13 This example illustrates a jump from a reading material page to a graph page. Here, in response to a jump trigger operation on target object 131, the terminal presents a page selection interface 132, and within this interface, page selection function items 133 and 134 are displayed. Page selection function item 134 is used to jump to graph page 1. In response to a selection operation on page selection function item 134, the terminal determines graph page 1 as the second page and displays it. During the display of the graph page, the terminal can respond to movement operations on the graph page to move it for easier browsing, and can also respond to zoom-in or zoom-out operations to zoom in or out of the graph page.

[0141] For example, see Figure 14 , Figure 14 This is an optional interface diagram of page navigation provided in the embodiments of this application. Figure 14 The example shown illustrates a jump from the graph page to the reading material page. Here, in response to a jump trigger operation for target object 141, the terminal queries the mapping relationship between the object and the jump page. If there is only one candidate page that maps to target object 141, then the candidate page is directly used as the second page and displayed.

[0142] In this embodiment, during the display of the first page, operations on the first page are detected. When the detected operation includes a jump trigger operation on a target object in the first page, the second page corresponding to the target object is determined and displayed based on the mapping relationship between the object and the jump page. Through the mapping relationship between the object and the jump page, the reading page of the electronic book and the graph page of the knowledge graph are associated, realizing accurate and fast location and jump display of the pages of the electronic book and the knowledge graph, thereby facilitating users to jump and look up information.

[0143] In some embodiments, operations on the displayed first page also include page turning, display adjustment, and editing operations. When the detected operation includes a page turning operation on the first page, the terminal displays the next page on the display interface. When the detected operation includes a display adjustment operation, the terminal adjusts the display format of the first page. Here, display adjustment operations include, but are not limited to, adjusting display brightness, font size, font color, and page background, etc. This application embodiment does not specifically limit the type of display adjustment operation.

[0144] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario. The page display method provided in the embodiments of this application is implemented on a server. See also... Figure 15 , Figure 15 This is an optional schematic diagram of the page display method provided in this application embodiment. In actual implementation, the page display method provided in this application embodiment may include the following operations:

[0145] Step 1500: The server receives the login system request.

[0146] Step 1501: Determine if the user is an administrator. If yes, proceed to step 1502; otherwise, proceed to step 1503.

[0147] Step 1502: Configure the learning materials categories and category weights, then proceed to step 1503;

[0148] Step 1503: Receive learning materials and proceed to step 1504;

[0149] Step 1504: Determine if it is text information. If it is, proceed to step 1505; otherwise, proceed to step 1507.

[0150] Step 1505: Construct a knowledge graph based on text information, then proceed to step 1506;

[0151] Step 1506: Perform word frequency analysis on the learning materials, construct a jump index table, and then proceed to step 1509;

[0152] Step 1507: Store the page location of the non-text information, then proceed to step 1508;

[0153] Step 1508: Build the page index table, then proceed to step 1509;

[0154] Step 1509: Determine whether to continue improving the knowledge graph. If yes, proceed to step 1503; otherwise, proceed to step 1510.

[0155] Step 1510: Determine whether the jump information is text information. If it is, proceed to step 1511; otherwise, proceed to step 1512.

[0156] Step 1511: Redirect the page according to the redirection index table;

[0157] Step 1512: Redirect the page based on image similarity and the page index table.

[0158] Here, before displaying the page, the server first builds an index table to navigate between pages. In actual implementation, the server receives login requests from the page display system. The login request carries the login account information, which corresponds to various user permissions, such as administrator and regular user permissions. The login account information includes the username and password.

[0159] In practice, the server determines whether the logged-in user has administrator privileges. If the user does, the server receives a configuration operation and configures the categorization and weighting of the learning materials to be uploaded. These learning materials include courseware, textbooks, supplementary materials, and practice questions. Administrators can categorize the learning materials according to their needs, such as by importance, and set different weightings for different categories. The server responds to the configuration operation, obtains the configuration information, and then categorizes and weights the learning materials accordingly. For example, it might set textbooks to 50%, PPTs to 30%, and other supplementary materials to 20%, providing a reference for subsequent knowledge graph construction and content positioning.

[0160] In practice, users can categorize and upload learning materials based on the categories configured by the administrator. The server receives the upload operation and obtains the learning materials carried in the upload operation. Specifically, when the server receives the upload operation, it can also obtain the category information of the learning materials carried in the upload operation, obtain the corresponding category weight based on the category information, and assign a category weight to the uploaded learning materials.

[0161] See Figure 16 , Figure 16This is an optional schematic diagram illustrating the knowledge graph construction process provided in this application embodiment. In actual implementation, the server extracts knowledge from the text information in the original data (learning materials). Here, the original data includes structured data, semi-structured data, and unstructured data. The server first integrates the original data, collecting, organizing, cleaning, and transforming data from different data sources to form a new data source. Next, the server extracts knowledge from the text information in the original data, including entity extraction, relation extraction, and attribute extraction, to obtain a preliminary knowledge representation. Here, the extracted knowledge includes entities, entity relations, and attributes. The extraction of entities and entity relations is described above in the embodiments of this application and will not be repeated here. After knowledge extraction, the server obtains multiple knowledge bases, and then further integrates the knowledge in the knowledge bases to form a unified knowledge base. Since different knowledge bases have different focuses in knowledge collection, it is necessary to align entities to their relations and attributes in the knowledge in different knowledge bases. Here, similarity calculation (cosine similarity, TFIDF similarity), aggregation (setting a threshold for each component of the similarity vector, and connecting the two entities if the threshold is exceeded), and clustering (hierarchical clustering: dividing knowledge at different levels by calculating the similarity between different knowledge, and finally forming a tree-like clustering structure) can be used to achieve relation alignment and attribute alignment.

[0162] In this embodiment, the server can use the Canopy+K-means clustering algorithm for relation alignment and attribute alignment. See also Figure 17 , Figure 17This is an optional schematic diagram of the Canopy+K-means algorithm provided in this application embodiment. Specifically, the server obtains distance thresholds T1 and T2, randomly selects target knowledge from multiple knowledge bases, calculates the distance D between the target knowledge and each cluster center. If the distance D is less than T1, it indicates that the target knowledge belongs to that cluster and is added to the cluster list. If the distance D is less than T2, it indicates that the target knowledge not only belongs to that cluster but is also very close to the current cluster center, so the target knowledge is deleted from the knowledge base. Here, knowledge can be entities and their attributes, the cluster center is an entity, and the distance D is the similarity. In this way, the server continuously aggregates similar entities and attributes in the knowledge base and removes duplicate entities and attributes. Next, based on their entity relationships and attributes, the server further constructs a knowledge graph and data model, supplements data according to the framework of the data model, and completes the construction of the knowledge graph. In practical implementation, the server further performs knowledge reasoning. Based on existing data models and data, and using logic programming methods, it sets rules that the answers must conform to in order to solve problems. ProLog language is used for rule reasoning to obtain new knowledge or conclusions, further refining the knowledge graph. For example, the frictional force on an object on a plane is Ff = μFn (μ is the coefficient of friction, Fn is the pressure), and the pressure exerted by the object on the plane is Fn = mg (m is the object's weight, g is the local gravitational acceleration). Through knowledge reasoning, we can further derive: the frictional force on an object on a plane is Ff = μmg.

[0163] In step 1506, the server performs word frequency statistics on the entities in the learning materials and establishes a correspondence between the entities and the pages of the learning materials with the highest word frequency to construct a jump index table. Here, the jump index table includes the mapping relationship between entities and jump pages. The process of constructing the jump index table based on word frequency (number of entities) is described in the above embodiments of this application and will not be repeated here.

[0164] In steps 1507 and 1508, the server stores the page positions of non-text information within the learning materials, establishing a correspondence between non-text information and pages in the learning materials to construct a page index table. In actual implementation, the server extracts images from the non-text information, stores the page positions of the images, and establishes a correspondence between the images and pages in the learning materials to construct a corresponding page index table.

[0165] In step 1509, the server determines whether it needs to continue improving the knowledge graph. Specifically, the server determines whether it has received an instruction to continue improving the knowledge graph. If it has received such an instruction, it returns to step 1503 to continue uploading learning materials; otherwise, it executes step 1510.

[0166] In step 1510, when the terminal displays the first page, it receives a jump trigger operation for the jump information and sends the jump trigger operation to the server. The server responds to the jump trigger operation by determining whether the jump information is text information. In this embodiment, the jump information is an object on the page.

[0167] In step 1511, when the server determines that the redirection information is text information, it uses the redirection information as an index to query the redirection index table, obtains the page position corresponding to the redirection information, and sends the obtained page position to the terminal so that the terminal displays the second page corresponding to the page position.

[0168] In step 1512, when the server determines that the jump information is an image, it calculates the similarity between the image and other page images, obtains the page image with the highest similarity, queries the page position corresponding to the page image with the highest similarity based on the page index table, and displays the page at that page position as the second page, so as to realize the jump from the first page to the second page.

[0169] The following description continues to illustrate the exemplary structure of the page display device 2055 provided in the embodiments of this application as a software module. In some embodiments, such as Figure 2 As shown, the software modules stored in the page display device 2055 of the memory 205 may include:

[0170] The first display module 20551 is used to display the first page;

[0171] Detection module 20552 is used to detect operations on the first page being displayed;

[0172] The determination module 20553 is used to determine the second page in the page set corresponding to the target object based on the mapping relationship between the object and the page when the detected operation includes a jump trigger operation for the target object in the first page; the page set includes reading pages of multiple electronic books and graph pages of the knowledge graph constructed based on the multiple electronic books.

[0173] The second display module 20554 is used to display the second page.

[0174] In some embodiments, the determining module 20553 is further configured to determine at least two candidate pages in the page set corresponding to the target object based on the mapping relationship between the object and the jump page; generate selection information; the selection information is used to indicate the selection operation for the at least two candidate pages; receive the selection operation for the at least two pages, and use the selected page as the second page.

[0175] In some embodiments, the page display device further includes: an acquisition module, configured to acquire the mapping relationship between the saved object and the page to be redirected.

[0176] In some embodiments, the page display device further includes: a mapping relationship construction module, configured to acquire multiple objects extracted from the reading pages of the multiple electronic reading materials; determine the jump page corresponding to each of the multiple objects from the reading pages of the multiple electronic reading materials and the graph page corresponding to the knowledge graph; construct a mapping relationship between the objects and the jump pages based on the multiple objects and the jump pages corresponding to each object, and save the constructed mapping relationship between the objects and the jump pages.

[0177] In some embodiments, the mapping relationship construction module is further configured to, for each object, determine from the reading pages of the plurality of electronic reading materials that satisfy the association conditions with the corresponding object, and use the determined reading pages as the jump pages corresponding to the corresponding object; and / or, for each object, determine from the graph pages corresponding to the knowledge graph that the page content contains the corresponding object, and use the determined graph pages as the jump pages corresponding to the corresponding object.

[0178] In some embodiments, when the corresponding object contains entities, the mapping relationship construction module is further configured to count the number of times the corresponding object appears in each page of the plurality of electronic publications to obtain the number of entities of the corresponding object contained in each page; select pages from the pages of the plurality of electronic publications whose number of entities meets the quantity condition, and use the selected pages as pages that meet the association condition with the corresponding object.

[0179] In some embodiments, when the object contains entities, the mapping relationship construction module is further configured to: count the number of times the entity appears in each page of the plurality of electronic publications to obtain the number of entities contained in each page; obtain the weight corresponding to each electronic publication in the plurality of electronic publications; score the corresponding page based on the weight corresponding to each electronic publication and the number of entities contained in each page to obtain the score of the corresponding page; select the page whose score meets the scoring conditions from the pages of the plurality of electronic publications, and use the selected page as the page that meets the association conditions with the corresponding object.

[0180] In some embodiments, when the corresponding object contains an image, the mapping relationship construction module is further configured to extract page images from each page of the plurality of electronic publications to obtain at least one page image; for each page image, determine the similarity between the corresponding object and the corresponding page image; and use the page image whose similarity satisfies the similarity condition as the publication page that satisfies the association condition with the corresponding object.

[0181] In some embodiments, the page display device further includes: a graph page generation module, configured to acquire page content corresponding to the reading pages of the plurality of electronic reading materials; extract entities from the page content to obtain a plurality of entities; extract entity relationships from the page content to obtain entity relationships between the plurality of entities; construct the knowledge graph based on the plurality of entities and the entity relationships between the plurality of entities; and generate corresponding graph pages based on the knowledge graph.

[0182] In some embodiments, the graph page generation module is further configured to extract semantic features from the page content by representing the BERT model through the bidirectional encoder of the converter, and obtain corresponding semantic features; and extract entity relationships from the page content based on the semantic features by using an entity relationship extraction model.

[0183] This application provides a computer program product or computer program that includes 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 page display method provided in this application.

[0184] This application provides a computer-readable storage medium storing executable instructions, wherein the executable instructions are stored and when executed by a processor, the processor will execute the page display method provided in this application.

[0185] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0186] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0187] As an example, executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple collaborating files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0188] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0189] In summary, the embodiments of this application enable accurate and rapid location and navigation of electronic publications and knowledge graph pages.

[0190] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A page display method, characterized in that, include: Display the first page; Detect actions performed on the first page displayed. When the detected operation includes a jump trigger operation targeting a target object in the first page, the second page in the page set corresponding to the target object is determined based on the mapping relationship between the object and the jump page; the page set includes reading pages of multiple e-books and graph pages of the knowledge graph constructed based on the multiple e-books; Display the second page; The method further includes: Obtain multiple objects extracted from the pages of the multiple electronic publications; From the reading pages of the multiple electronic books and the graph pages corresponding to the knowledge graph, determine the jump page corresponding to each of the multiple objects; based on the multiple objects and the jump page corresponding to each object, construct the mapping relationship between the objects and the jump pages, and save the constructed mapping relationship between the objects and the jump pages.

2. The method according to claim 1, characterized in that, The step of determining the second page in the set of pages corresponding to the target object based on the mapping relationship between the object and the redirected page includes: Based on the mapping relationship between objects and redirected pages, at least two candidate pages are determined from the set of pages corresponding to the target object; Generate selection information; the selection information is used to indicate the selection operation for the at least two candidate pages; Receive a selection operation for the at least two candidate pages, and use the selected candidate page as the second page.

3. The method according to claim 1, characterized in that, The method further includes: Retrieve the mapping relationship between the saved object and the page to be redirected.

4. The method according to claim 1, characterized in that, The step of determining the jump page corresponding to each of the multiple objects from the reading pages of the multiple electronic books and the graph pages corresponding to the knowledge graph includes: For each object, from the pages of the multiple electronic publications, determine the page that satisfies the association condition with the corresponding object, and use the determined page as the jump page corresponding to the object; And / or, For each object, determine the graph page whose content contains the corresponding object from the graph pages corresponding to the knowledge graph, and use the determined graph page as the jump page corresponding to the corresponding object.

5. The method according to claim 4, characterized in that, When the corresponding object contains an entity, determining the page from the plurality of electronic publications that satisfies the association condition with the corresponding object includes: The number of times the corresponding object appears in each page of the multiple electronic publications is counted to obtain the number of entities of the corresponding object contained in each page; From the pages of the multiple electronic publications, select the pages whose number of entities meets the quantity condition, and use the selected pages as the pages that meet the association condition with the corresponding objects.

6. The method according to claim 4, characterized in that, When the object contains an entity, determining the pages from the plurality of electronic publications that satisfy the association conditions with the object includes: The number of times the entity appears in each page of the multiple electronic publications is counted to obtain the number of entities of the entity contained in each page; Obtain the weight corresponding to each of the plurality of electronic publications; Each e-book is scored based on its corresponding weight and the number of entities contained in each e-book page, resulting in a score for the corresponding e-book page. From the pages of the multiple electronic publications, select the pages whose ratings meet the rating criteria, and use these selected pages as the pages that satisfy the association criteria with the corresponding objects.

7. The method according to claim 4, characterized in that, When the corresponding object contains an image, determining the page from the plurality of electronic publications that satisfies the association condition with the corresponding object includes: Extract page images from each page of the plurality of electronic publications to obtain at least one page image; For each page image, determine the similarity between the corresponding object and the corresponding page image; and designate the page images whose similarity meets the similarity criteria as reading pages that satisfy the association criteria with the corresponding object.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes: Obtain the page content corresponding to the reading pages of the multiple electronic publications; Entity extraction is performed on the page content to obtain multiple entities; Extract entity relationships from the page content to obtain the entity relationships between the multiple entities; The knowledge graph is constructed based on the multiple entities and the entity relationships between them. Based on the knowledge graph, a corresponding graph page is generated.

9. The method according to claim 8, characterized in that, The extraction of entity relationships from the page content includes: The BERT model is represented by a bidirectional encoder of the converter, and semantic features are extracted from the page content to obtain the corresponding semantic features. The entity relationship extraction model extracts entity relationships from the page content based on the semantic features.

10. A page display device, characterized in that, include: The first display module is used to display the first page; The detection module is used to detect operations performed on the first page of the display. The first determining module is used to determine the second page in the page set corresponding to the target object based on the mapping relationship between the object and the page when the detected operation includes a jump trigger operation for the target object in the first page; the page set includes reading pages of multiple electronic books and graph pages of the knowledge graph constructed based on the multiple electronic books. The second display module is used to display the second page; The acquisition module is used to acquire multiple objects extracted from the pages of the multiple electronic publications; The second determining module is used to determine the jump page corresponding to each of the multiple objects from the reading pages of the multiple electronic reading materials and the graph pages corresponding to the knowledge graph; the constructing module is used to construct the mapping relationship between the objects and the jump pages based on the multiple objects and the jump pages corresponding to each object, and save the constructed mapping relationship between the objects and the jump pages.

11. An electronic device, characterized in that, include: Memory, used to store executable instructions; A processor, when executing executable instructions stored in the memory, implements the page display method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, It stores executable instructions for use by a processor to implement the page display method according to any one of claims 1 to 9.

Images