Display method and device of floating window, computer equipment and storage medium

Abstract

The invention relates to a floating window display method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps of displaying a target page, and displaying current display content and a suspension component in the target page; in response to a trigger operation for the suspension component, displaying the suspension window; and displaying at least one of the content summary corresponding to the current display content and the recommendation information related to the current display content in the floating window. By adopting the method, the resource utilization rate can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer device, and storage medium for displaying a floating window. Background Technology

[0002] In the internet age, with the development of internet information technology, internet information display technology has emerged. This technology allows for the presentation of information; for example, users can read text or watch videos through information display applications. Specifically, these applications can be text reading apps or short video apps. Users can read text through text reading apps, or watch short videos through short video apps.

[0003] Information display applications can be either full applications or mini-programs. For example, if the information display application is a mini-program, users can launch it from an instant messaging application and view information through it. However, currently, information display applications are often only used for text reading or video playback; for instance, text reading applications are only used for text reading, and short video applications are only used for video playback, resulting in low resource utilization. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for displaying floating windows that can improve resource utilization, in order to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for displaying a floating window, the method comprising:

[0006] Display the target page, and on the target page display the currently displayed content and the floating component;

[0007] In response to a trigger operation on the floating component, a floating window is displayed;

[0008] The floating window displays at least one of the following: a summary of the currently displayed content and recommended information related to the currently displayed content.

[0009] Secondly, this application also provides a floating window display device, the device comprising:

[0010] The floating component display module is used to display the target page, showing the currently displayed content and the floating component on the target page.

[0011] The floating window display module is used to display the floating window in response to a trigger operation on the floating component.

[0012] An information display module is used to display, in the floating window, at least one of the following: a content summary corresponding to the currently displayed content and recommended information related to the currently displayed content.

[0013] In one embodiment, the floating window display module is further configured to, in response to a trigger operation on the floating component, switch the floating component to a floating window and display the floating window on the target page; and, in response to a closing operation on the floating window, close the floating window and redisplay the floating component.

[0014] In one embodiment, the floating window display module is further configured to move the floating window according to a drag trajectory in response to a drag operation on the floating window; or to scale the floating window according to a scaling ratio corresponding to a scaling operation on the floating window in response to a scaling operation on the floating window.

[0015] In one embodiment, the recommendation information is a recommended text link related to the currently displayed content; the information display module is further configured to display a content summary corresponding to the currently displayed content and the recommended text link in the floating window; the text displayed on the recommended text link is a content summary of specific recommended content related to the currently displayed content; in response to a trigger operation on the recommended text link, specific recommended content related to the currently displayed content is displayed.

[0016] In one embodiment, the floating window display device further includes a setting module for displaying a content summary option and a recommendation information option in response to a content setting operation for the floating window; the content summary option is used to trigger the floating window to display a content summary when selected; the recommendation information option is used to trigger the floating window to display recommendation information when selected.

[0017] In one embodiment, the floating window includes a session display area; the information display module is further configured to display at least one of the content summary corresponding to the currently displayed content and the recommendation information related to the currently displayed content as a session message of the machine learning model in the session display area.

[0018] In one embodiment, the floating window further includes a conversation input area; the conversation input area displays a conversation input box; the information display module is further configured to respond to a trigger operation on the conversation input box by displaying the question entered through the conversation input box as a user's conversation message in the conversation display area; and displaying the answer corresponding to the question as a conversation message of the machine learning model in the conversation display area.

[0019] In one embodiment, the machine learning model's session message also displays at least one question button; each question button displays a predicted question based on the currently displayed content, and different question buttons display different predicted questions; the information display module is also used to respond to a trigger operation on the question button by displaying the answer corresponding to the predicted question displayed on the triggered question button as a session message of the machine learning model in the session display area.

[0020] In one embodiment, the information display module is further configured to, in response to a progress adjustment operation on the currently displayed content, switch the currently displayed content to the content after progress adjustment; and display at least one of a content summary and related recommendation information corresponding to the content after progress adjustment in the floating window.

[0021] In one embodiment, the information display module is further configured to, in response to an article sentence selection operation in the currently displayed content, distinguish and display the selected article sentence in the currently displayed content, and distinguish and display a summary sentence generated based on the selected article sentence in the content summary, when the target page is an article reading page.

[0022] In one embodiment, the information display module is further configured to, in response to an operation of selecting an article paragraph in the currently displayed content, distinguish and display the selected article paragraph in the currently displayed content, and add a paragraph summary of the selected article paragraph in the floating window when the target page is an article reading page.

[0023] In one embodiment, the information display module is further configured to close the floating window and display a content selection page in response to an exit operation on the target page; the content selection page displays multiple content titles; in response to a trigger operation on the content title, the floating window is re-displayed, and a summary of the content corresponding to the triggered content title is displayed in the re-displayed floating window.

[0024] In one embodiment, the recommendation information is a recommended video related to the currently displayed content; the recommended video is generated based on the already delivered content related to the currently displayed content.

[0025] In one embodiment, the floating window display device further includes an archiving module, which, in response to an archiving operation for the content summary, displays a successful archiving prompt and adds the summary information of the content summary to the archive list; and, in response to a sharing operation for the summary information in the archive list, shares the content summary corresponding to the summary information that triggered the sharing operation to the target sharing object specified by the sharing operation.

[0026] In one embodiment, the floating window display device further includes a summary generation module, which is used to determine key statements based on the currently displayed content; generate a content summary corresponding to the currently displayed content based on the key statements; and display the content summary corresponding to the currently displayed content in the floating window.

[0027] In one embodiment, the floating window display device further includes a summary generation module, used to determine the frequency of occurrence of each word in the currently displayed content and the context when the currently displayed content is the currently read content, and to determine the position information, length information, and sentence similarity of each sentence in the currently displayed content and the context; and to filter out key sentences from the currently displayed content and the context by combining the frequency of occurrence of each word, the position information of each sentence, the length information of each sentence, and the sentence similarity of each sentence.

[0028] In one embodiment, the floating window display device further includes a link generation module, configured to determine the current content features of the currently displayed content, and determine the similarity between the current content features and candidate content features in a preset feature library; filter at least one candidate content feature from the preset feature library based on the similarity; determine specific recommended content based on the filtered at least one candidate content feature; generate a content summary of the specific recommended content, and generate a recommended text link based on the content summary of the specific recommended content.

[0029] In one embodiment, the floating window display device further includes a link generation module, used to obtain interest tags corresponding to the currently logged-in account on the target page; filter target content features from at least one selected candidate content features based on the interest tags; and generate specific recommended content based on the target content features.

[0030] Thirdly, this application also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0031] Display the target page, and on the target page display the currently displayed content and the floating component;

[0032] In response to a trigger operation on the floating component, a floating window is displayed;

[0033] The floating window displays at least one of the following: a summary of the currently displayed content and recommended information related to the currently displayed content.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0035] Display the target page, and on the target page display the currently displayed content and the floating component;

[0036] In response to a trigger operation on the floating component, a floating window is displayed;

[0037] The floating window displays at least one of the following: a summary of the currently displayed content and recommended information related to the currently displayed content.

[0038] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0039] Display the target page, and on the target page display the currently displayed content and the floating component;

[0040] In response to a trigger operation on the floating component, a floating window is displayed;

[0041] The floating window displays at least one of the following: a summary of the currently displayed content and recommended information related to the currently displayed content.

[0042] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for displaying floating windows can display currently displayed content and a floating component on a target page. By displaying the floating component, a floating window can be displayed in response to a trigger operation on the floating component, and at least one of the following can be displayed: a summary of the currently displayed content and related recommendations. Since it can display not only the currently displayed content but also a summary of the currently displayed content and related recommendations, this application improves resource utilization compared to traditional methods that can only display the currently displayed content.

[0043] Furthermore, by displaying a floating component on the target page, users can view real-time content summaries based on this component while reading or watching videos, improving information retrieval efficiency, especially suitable for long articles, long videos, or time-sensitive scenarios. Moreover, displaying a floating component on the target page also allows users to view real-time recommendations related to the currently displayed content while reading or watching videos. Because the displayed recommendations are relevant to the currently displayed content, they are highly matched to the user's current needs, thus enhancing the effectiveness of information recommendations. Attached Figure Description

[0044] Figure 1 This is an application environment diagram of a floating window display method in one embodiment;

[0045] Figure 2 This is a flowchart illustrating a method for displaying a floating window in one embodiment;

[0046] Figure 3 This is a schematic diagram of the target page in one embodiment;

[0047] Figure 4A This is a schematic diagram of a floating window in one embodiment;

[0048] Figure 4B This is a schematic diagram of a floating window in another embodiment;

[0049] Figure 5 This is a schematic diagram of a floating window in yet another embodiment;

[0050] Figure 6 This is a schematic diagram of recommendation information in one embodiment;

[0051] Figure 7 This is a schematic diagram illustrating recommended text links in one embodiment;

[0052] Figure 8 This is a schematic diagram of a floating window in yet another embodiment;

[0053] Figure 9 This is a schematic diagram of the problematic button in one embodiment;

[0054] Figure 10 This is a schematic diagram of a question response in one embodiment;

[0055] Figure 11 This is a schematic diagram illustrating the differences between statements in one embodiment;

[0056] Figure 12 This is a schematic diagram illustrating the generation of paragraph summaries in one embodiment;

[0057] Figure 13This is a schematic diagram illustrating the summary of the article in one embodiment;

[0058] Figure 14 This is an interactive schematic diagram of a floating window display method in one embodiment;

[0059] Figure 15 This is a structural block diagram of a floating window display device in one embodiment;

[0060] Figure 16 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] The floating window display method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated on server 104 or placed in the cloud or on other servers. Both terminal 102 and server 104 can be used independently to execute the floating window display method provided in this embodiment. Terminal 102 and server 104 can also work together to execute the floating window display method provided in this embodiment. Taking the example of terminal 102 and server 104 working together to execute the floating window display method provided in this embodiment, a user can read text through terminal 102, thereby terminal 102 can display a target page, and display the currently displayed content and floating component on the target page. When the user triggers the floating component, terminal 102 can send an information retrieval request to server 106, thereby server 106 can generate a content summary and recommendation information corresponding to the currently displayed content, and send the content summary and recommendation information to terminal 106, so that terminal 106 can display the floating window and display it within the floating window. The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0063] It should be noted that the terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Unless the context clearly indicates otherwise, the singular forms "a," "one," or "the," etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The quantities of "multiple" or "multiple copies" mentioned in the embodiments of this application all refer to a quantity of "at least two," for example, "multiple" means "at least two," and "multiple copies" means "at least two copies."

[0064] In one embodiment, such as Figure 2 As shown, a method for displaying a floating window is provided, which can be applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps:

[0065] Step 202: Display the target page, showing the currently displayed content and floating components.

[0066] The target page can be an article reading page or a video playback page. When the target page is an article reading page, the currently displayed content is the currently displayed reading content on the article reading page; when the target page is a video playback page, the currently displayed content is the currently playing video on the video playback page.

[0067] Specifically, when a user wishes to read or watch a video, the target page can be displayed on the device. For example, a user can launch a communication application on the device, then launch a reading mini-program through the communication application, and finally display the target page through the reading mini-program. Furthermore, when the target page is displayed on the device, the currently displayed content and floating components can also be shown on the target page. For example, see [reference needed]. Figure 3 , Figure 3 The illustration shows a schematic diagram of a target page in one embodiment. When the target page is an article reading page, it can be... Figure 3 The target page shown displays the article title, article text, article images, and a floating component 301 that can be displayed on the target page. The floating component refers to a component used to display a floating window.

[0068] Specifically, the currently displayed content can refer to the content currently displayed on the target page. If the target page is an article reading page, it's easy to understand that, due to varying article lengths, the complete article content may not be displayed on the same page. Therefore, when reading a short article, the currently displayed content on the target page may be the entire article; when reading a long article, the currently displayed content on the target page may be a portion of that long article. In this case, users can adjust their reading progress using scrollbars, such as those displayed on the target page.

[0069] In one embodiment, when the target page is an article reading page, after the terminal launches the reading mini-program, the mini-program can display the titles of multiple articles. The user can select an article of interest to read, and then the reading mini-program can display the content of the selected article as the currently displayed content, and display a floating component at a preset position on the target page. When the target page is a video playback page, after the terminal launches the short video mini-program, the short video mini-program can play a video. The user can change the currently playing video by swiping up or down, and the currently playing video will be the currently displayed content.

[0070] In one embodiment, the user can change the position of the floating component on the target page. For example, the terminal can respond to the user's drag operation on the floating component, move the floating component according to the drag route, and fix it at the end of the drag route.

[0071] Step 204: In response to a trigger action on the floating component, display the floating window.

[0072] Specifically, users can trigger the display of a floating window on the terminal at any time during reading based on the floating component. For example, a user can trigger the floating component while reading, and the terminal can then respond to the trigger operation by displaying the floating window. Trigger operations include, but are not limited to, click operations, long press operations, press-and-drag operations, double-click operations, etc. For instance, a user can click the floating component while reading, and the terminal can then display the floating window in response to that click operation.

[0073] In one embodiment, the floating window can be displayed while floating on the target page, or it can be embedded in the target page. For example, see reference... Figure 4A and Figure 4B , Figure 4A A schematic diagram of a floating window in one embodiment is shown. Figure 4BA schematic diagram of a floating window in another embodiment is shown. When the target page is an article reading page, when the floating window 401 is displayed floating on the target page, the currently displayed content of the target page does not change; the floating window simply floats on top of the currently displayed content. When the target page is a video playback page, when the floating window 402 is displayed floating on the target page, the target page can continue playing the current video; the floating window simply floats on top of the currently displayed content.

[0074] For example, refer to Figure 5 , Figure 5 A schematic diagram of a floating window is shown in yet another embodiment. When the target page is an article reading page, the floating window 501 can also be embedded within the target page for display, so that the currently displayed content is shown in the remaining areas of the target page excluding the floating window.

[0075] Step 206: Display at least one of the following in the floating window: a summary of the content currently displayed and recommended information related to the content currently displayed.

[0076] Specifically, when a floating window is displayed, it can also display at least one of the following: a summary of the currently displayed content and related recommendations. For example, see [reference]. Figure 5 , Figure 5 The floating window can display a summary (502) and recommendations (503). It's easy to understand that you can display only the summary or recommendations, or both.

[0077] In one embodiment, the summary of the currently displayed content can be a summary text generated by summarizing the currently displayed content. When the target page is an article reading page, the currently displayed content can also be a summary text generated by summarizing the currently displayed content and its context, or it can be a summary text generated by summarizing the entire article containing the currently displayed content.

[0078] In one embodiment, the recommendation information may be information related to the currently displayed content recommended to the user. Specifically, the recommendation information may be a recommended text link, such as a reference. Figure 5 It can be displayed in a floating window, such as Figure 5The recommended text link 503 shown allows users to click on specific recommended content when they wish to view it. The terminal then responds to this click by displaying the recommended content. For example, if the currently displayed content is a car introduction article or video, the recommendation information could be a recommended text link to a car advertisement. When the user clicks this link, the terminal can redirect to the car advertisement page and display the recommended car advertisement.

[0079] In one embodiment, the recommendation information may also be other text that the user might read based on predictions of the currently displayed content. For example, references Figure 6 It can be displayed in a floating window, such as Figure 6 The text content 601 shown allows users to expand their reading through the displayed text. For example, when the currently displayed content is a car introduction article or video, recommended information could include car driving methods, methods for obtaining a driver's license, etc. Figure 6 A schematic diagram of recommendation information in one embodiment is shown.

[0080] In the above-described method for displaying a floating window, by displaying a target page, the currently displayed content and a floating component can be shown on the target page. By displaying the floating component, in response to a trigger operation on the floating component, a floating window can be displayed, showing at least one of the following: a summary of the currently displayed content and related recommendation information. Since it can display not only the currently displayed content but also a summary of the currently displayed content and related recommendation information, compared to traditional methods that can only display the currently displayed content, this application improves resource utilization.

[0081] Furthermore, by displaying a floating component on the target page, users can view content summaries in real time while reading or watching videos, improving information retrieval efficiency, especially suitable for long articles, long videos, or time-sensitive scenarios. Moreover, displaying a floating component on the target page also allows users to view related recommendations in real time while viewing information. Because these recommendations are relevant to the currently displayed content, they are highly matched to the user's current needs, thus enhancing the effectiveness of information recommendations.

[0082] In one embodiment, displaying a target page includes: displaying the target page and displaying currently displayed content on the target page; and displaying a floating window in response to a trigger operation on the floating component includes: embedding the floating window in the currently displayed content in response to a trigger operation on the floating component.

[0083] Specifically, after a user selects an article they wish to read, the terminal can display the target page and its content, such as the article title, body text, and accompanying images. The terminal can also display a floating component on the target page. When the user clicks this component, the terminal can respond by embedding a floating window within the displayed content. For example, the floating window could be displayed in the center of the target page, allowing the content to be displayed on the top and bottom edges. By embedding the floating window within the content, it avoids obstructing the view of the current content, allowing the user to read the current content while viewing the content in the floating window.

[0084] In one embodiment, displaying a floating window in response to a trigger operation on the floating component includes: switching the floating component to a floating window in response to the trigger operation on the floating component, and displaying the floating window floating on the target page; the method further includes: closing the floating window and redisplaying the floating component in response to a closing operation on the floating window.

[0085] Specifically, the terminal can display a target page, including the currently displayed content and a floating component. When the user triggers the floating component, the terminal can de-display the floating component and display a floating window on the target page. The floating window displays a close button, which the user can use to close it. The terminal can then respond to this close operation by closing the floating window and redisplaying the floating component.

[0086] In this embodiment, displaying a floating component allows users to easily trigger the display of a floating window on the terminal, simplifying the display process and improving efficiency. A close button on the floating window allows users to close it. Re-displaying the floating component after closing the window allows users to reopen it based on the re-displayed component. This achieves the goal of allowing users to view content summaries and recommendations at any time during reading.

[0087] In one embodiment, the method further includes: moving the floating window along a drag trajectory in response to a drag operation on the floating window; or scaling the floating window according to the scaling ratio corresponding to the scaling operation in response to a scaling operation on the floating window.

[0088] Specifically, when a floating window is displayed on the target page, the terminal can display at least one of the following: a summary of the currently displayed content and recommended information associated with the currently displayed content. Furthermore, the user can move and zoom the displayed floating window. For example, when the user wants to move the floating window, the user can drag it, and the terminal can respond to the drag operation by moving the floating window along the drag trajectory. Similarly, when the user wants to zoom the floating window, the user can trigger a zoom operation on the floating window, and the terminal can respond to the zoom operation by scaling the floating window according to the zoom ratio indicated by the zoom operation.

[0089] In a straightforward manner, when a floating window is displayed on a target page, if it obscures the content the user wishes to read, the user can drag and / or zoom the floating window to simultaneously view the content within it and read or watch the currently displayed content. When the target page is an article reading page, if the user finds the size or position of the floating window unsuitable, they can drag and / or zoom it. During and after these actions, the floating window remains embedded within the currently displayed content. For example, if initially the floating window is in the center of the target page and the currently displayed content is on the top and bottom edges, dragging the floating window to the top half of the target page will change the display location from the top and bottom edges to the bottom half.

[0090] In the above embodiments, users can freely adjust the position and size of the floating window based on drag and zoom operations as needed, which not only improves the flexibility of the floating window adjustment, but also enhances the user experience.

[0091] In one embodiment, the recommendation information is a recommended text link; displaying at least one of the following in a floating window: a content summary corresponding to the currently displayed content and recommendation information related to the currently displayed content, includes: displaying a content summary corresponding to the currently displayed content and a recommended text link in a floating window; the text displayed on the recommended text link is a content summary of specific recommended content related to the currently displayed content; the method further includes: in response to a trigger operation on the recommended text link, displaying specific recommended content related to the currently displayed content.

[0092] Specifically, the recommended information can be recommended text links. When the terminal displays a floating window, it can obtain a content summary and recommended text links generated for the currently displayed content, and then display these in the floating window. The recommended text link is a text chain, which contains only text. The recommended text link displays a summary of the specific recommended content related to the currently displayed content. For example, see [reference] Figure 7 If the current content being displayed is an introduction to book A, two recommended text links can be displayed. Recommended text link 701 can display the text "Jump to store B to buy book A", and recommended text link 702 can display the text "Jump to reading software C to read book A". Figure 7 A schematic diagram of recommended text links is shown in one embodiment.

[0093] Furthermore, since the text displayed on the recommended text link is a summary of the specific recommended content, users can roughly understand the specific recommended content based on the text on the recommended text link, thus determining whether to click on the recommended text link. When a user wants to view the specific recommended content, the user can click on the recommended text link, and the terminal can respond to the user's click operation by displaying specific recommended content related to the currently displayed content. For example, in the above example, when the user clicks recommended text link 701, the terminal can jump to the purchase page of book A in store B, and display the detailed introduction of book A on the purchase page of book A. This detailed introduction is the specific recommended content related to the currently displayed content mentioned above. When the user clicks recommended text link 702, the terminal can jump to the reading page of book A in reading software C, and display the book content of book A on the reading page of book A. This book content is the specific recommended content related to the currently displayed content mentioned above.

[0094] In the above embodiments, displaying a content summary allows users to quickly understand the main points of the currently displayed content, thereby improving information retrieval efficiency. Displaying recommended text links allows users to quickly understand the main points of the specific recommended content based on the text displayed on the links, further improving information retrieval efficiency. Furthermore, since a recommended text link is displayed instead of directly showing the specific recommended content, the simplicity of the content displayed in the floating window is enhanced, allowing users to quickly access important content through a concise floating window, greatly improving the user experience.

[0095] Furthermore, since the recommended text links can display text, users can quickly grasp the summary of the recommended content based on the displayed text, and then decide whether they need to read the specific recommended content, thus improving the user experience. Because users can determine whether they need to read the specific recommended content based on the summary, the specific recommended content can only be displayed when reading it is required. This reduces the unnecessary display of specific recommended content and saves the computer and other resources consumed by displaying unnecessary content.

[0096] In one embodiment, the recommended information is supplementary reading content; displaying at least one of the following in the floating window: a summary of the content corresponding to the currently displayed content and recommended information related to the currently displayed content; the supplementary reading content is text that supplements or expands upon the currently displayed content.

[0097] Specifically, the recommended information can also include supplementary reading content. When the terminal displays a floating window, it can obtain a summary and supplementary reading content generated for the currently displayed content, and then display these in the floating window. The supplementary reading content can be text that supplements or expands upon the currently displayed content. For example, when the currently displayed content is a car introduction, the supplementary reading content could be car information not mentioned in the current content, or information about a specific car component mentioned in the current content. For instance, the supplementary text could be, "Car A also includes engine model C, which uses XX technology to achieve…". Similarly, when the currently displayed content is an introduction to book A, the supplementary reading text could be information about book A not mentioned in the current content, or text that further describes a section of book A mentioned in the current content. For instance, the supplementary text could be, "The preface to book A was written by author XX…".

[0098] In this embodiment, since the recommended information can be supplementary reading content, users can supplement the currently displayed content based on the supplementary reading content, thus eliminating the need to search for relevant supplementary reading text on the network. This not only improves the user experience but also increases the efficiency of obtaining supplementary reading text.

[0099] In one embodiment, the method further includes: in response to a content setting operation for the floating window, displaying a content summary option and a recommendation information option; the content summary option is used to trigger the floating window to display a content summary when selected; the recommendation information option is used to trigger the floating window to display recommendation information when selected.

[0100] Specifically, users can customize the content displayed in the floating window through content settings. For example, a settings button can be displayed on the target page, allowing users to adjust the displayed content by clicking the button. The terminal responds to this content settings action by displaying a settings page. This page displays a content summary option and a recommendation information option. When a user selects the content summary option, it's assumed they expect to see the summary in the floating window; therefore, the terminal can trigger the display of the summary in the floating window based on this selection. Similarly, when a user selects the recommendation information option, it's assumed they expect to see recommended information in the floating window; therefore, the terminal can trigger the display of recommended information in the floating window based on this selection.

[0101] In the above embodiments, users can freely set the content displayed in the floating window through the content display settings, thereby increasing the freedom of content display and improving the user experience.

[0102] In one embodiment, the floating window includes a session display area; displaying at least one of the content summary corresponding to the currently displayed content and the recommendation information related to the currently displayed content in the floating window includes: displaying at least one of the content summary corresponding to the currently displayed content and the recommendation information related to the currently displayed content as a session message of the machine learning model in the session display area.

[0103] Specifically, the floating window includes a session display area and a session input area, for example, reference Figure 8 , Figure 8 801 in the diagram is the session display area. Figure 8 802 in the diagram represents the conversation input area. The conversation display area can show at least one of a content summary and recommendation information, while the conversation input area can display a conversation input box. When the user clicks the floating component, the terminal can display a floating window and generate a content summary and recommendation information based on the currently displayed content using a machine learning model. This content summary and recommendation information generated by the machine learning model is then displayed as a conversation message in the conversation display area. For example, refer to... Figure 8 The conversation display area can display the avatar of the machine learning model, and next to the machine learning model's avatar, the conversation message 803 is displayed. The content of the conversation message is a summary of the content corresponding to the currently displayed content and related recommendation information. Figure 8A schematic diagram of a floating window in one embodiment is shown.

[0104] In one embodiment, the floating window further includes a conversation input area; a conversation input box is displayed in the conversation input area; the method further includes: in response to a trigger operation on the conversation input box, displaying the question entered through the conversation input box as a user's conversation message in the conversation display area; and displaying the answer corresponding to the question as a conversation message of the machine learning model in the conversation display area.

[0105] Specifically, when a user has questions about the displayed content summary or recommendations, or expects to obtain more information about the content summary or recommendations, the user can enter a question through the conversation input box in the conversation input area. The terminal will then display the user's input question as a conversation message in the conversation display area. For ease of description, the questions entered by the user through the conversation input box will be referred to as inquiry questions below. For example, when the content summary includes "Hyundai cars are equipped with intelligent driving assistance systems that use sensors, cameras, and radar to determine the surrounding environment," the user can enter "How exactly do sensors, cameras, and radar use to determine the surrounding environment?" through the conversation input box. The terminal will then display this inquiry question, "How exactly do sensors, cameras, and radar use to determine the surrounding environment?" as the user's conversation message 804 in the conversation display area.

[0106] Furthermore, the terminal can input the user's query into a machine learning model, which will then output an answer to the query. The terminal can then display this answer as a conversational message from the machine learning model in the conversation display area. For example, refer to... Figure 8 In the example above, the terminal can display "obtain sensor information through sensors, obtain point cloud information through radar... determine the surrounding environment" as a session message 805 of the machine learning model in the session display area.

[0107] In one embodiment, if the user still has doubts about the answer output by the terminal, the user can continue to enter questions through the conversation input box, and the terminal can display the answers to the continued questions in the conversation display area until the user gets a satisfactory answer.

[0108] In the above embodiments, by displaying a conversation input box, users can easily input questions through the box, and the terminal can then display answers to those questions. Compared to finding answers from other channels, this embodiment significantly improves the efficiency of answer retrieval, that is, it greatly improves the efficiency of information acquisition. Furthermore, displaying information in a conversational manner makes it easier for users to distinguish between their own input and the output of the machine learning model, further enhancing the user experience.

[0109] In one embodiment, the machine learning model's conversation message also displays at least one question button; each question button displays a predicted question based on the currently displayed content, and different question buttons display different predicted questions; the method further includes: in response to a trigger operation on a question button, displaying the answer corresponding to the predicted question displayed on the triggered question button as a conversation message of the machine learning model in the conversation display area.

[0110] Specifically, when a floating window is displayed on the terminal, the terminal can input the currently displayed content into a machine learning model. The machine learning model then predicts the questions the user might ask regarding that content. Based on the predicted questions, the terminal generates a question button and displays the question button, a summary of the currently displayed content, and related recommendations in the conversation display area. For ease of description, the predicted questions based on the currently displayed content will be referred to as predicted questions. For example, refer to... Figure 9 The question button, the summary of the content corresponding to the currently displayed content, and the recommended information related to the currently displayed content are displayed as the machine learning model's conversation message 901 in the conversation display area.

[0111] The question button can display a predicted question, and different question buttons can display different predicted questions. For example, refer to... Figure 9 When the current displayed content is a car introduction, the predicted question displayed on question button 902 could be "What are the significant features of modern cars?", and the question displayed on question button 903 could be "How to remotely start a vehicle via mobile phone?". Furthermore, when a user clicks a question button, the terminal can display the answer to the predicted question on that question button as a conversation message from the machine learning model in the conversation display area. For example, in the above example, when the user clicks question button 903, the terminal can display... Figure 10 The session message 1001 shown reads, "First, the user can enable the vehicle startup application...". Figure 9 A schematic diagram of the problematic button in one embodiment is shown. Figure 10 A schematic diagram of a question response in one embodiment is shown.

[0112] In one embodiment, the predicted question related to the currently displayed content can be a question output by a machine learning model based on the currently displayed content, or a question output based on a summary of the content corresponding to the currently displayed content, or a question output by combining the currently displayed content and the summary of the content corresponding to the currently displayed content. For example, the currently displayed content can be input into a machine learning model, which can predict the questions the user might ask about the currently displayed content. The terminal can then generate and display a question button based on the predicted question from the machine learning model. Alternatively, the terminal can concatenate the currently displayed content and the summary of the content corresponding to the currently displayed content to obtain concatenated text, and input the concatenated text into a machine learning model. The machine learning model can then predict the questions the user might ask about the concatenated text, and the terminal can then generate and display a question button based on the predicted question from the machine learning model.

[0113] In the above embodiments, by predicting in advance the questions that users may ask related to the currently displayed content, the predicted questions can be displayed on the question button. Thus, users can get answers simply by clicking the question button, which not only improves the user experience but also greatly improves the efficiency of obtaining answers.

[0114] In one embodiment, when a user finishes viewing the content displayed in the session display area and wishes to return to continue reading the currently displayed content, the user can click the floating window close button. The terminal can then respond to the click operation of the floating window close button, close the floating window, and redisplay the floating component.

[0115] In one embodiment, the method further includes: in response to a progress adjustment operation on the currently displayed content, switching the currently displayed content to the content after progress adjustment; and displaying at least one of a summary of the content after progress adjustment and related recommendation information in a floating window.

[0116] Specifically, the target page can also display progress adjustment buttons, allowing users to adjust the reading progress. For example, if the article being read is long, users can adjust the reading progress using the progress bar, or they can adjust the video playback progress using a gesture. The terminal can then respond to progress adjustment operations on the currently displayed content, changing the displayed content to the content with the adjusted progress. For instance, if the currently displayed content is the first and second paragraphs of an article, after the user finishes reading the first and second paragraphs, they can use the reading progress adjustment operation to change the first and second paragraphs displayed on the target page to the third and fourth paragraphs of the article. The third and fourth paragraphs displayed on the target page at this point represent the content with the adjusted progress.

[0117] Furthermore, when a user wants to view a summary or recommendation information for the content after the progress adjustment, the user can click on the floating component. The terminal can then respond to the user's click on the floating component by displaying a floating window, and in the floating window, display at least one of the content summary corresponding to the content after the progress adjustment and the recommendation information related to the content after the progress adjustment.

[0118] In one embodiment, at least one of the content summary and recommendation information corresponding to the content before the progress adjustment can be used as a historical conversation message. This historical conversation message can then be displayed in the floating window. At least one of the content summary and recommendation information corresponding to the content after the progress adjustment can also be displayed as a new conversation message in the conversation display area. By retaining historical conversation messages, users can review the content summary and recommendation information corresponding to the content before the progress adjustment, which not only improves the user experience but also increases the efficiency of obtaining the content summary and recommendation information.

[0119] In one embodiment, when the length of the article currently being read by the user is less than or equal to a preset length threshold, the summary of the reading content before the progress adjustment is consistent with the summary of the reading content after the progress adjustment. In this case, the summary can be a text output summarizing the entire article. When the length of the article currently being read by the user is greater than the preset length threshold, the summary of the reading content before the progress adjustment may differ from the summary of the reading content after the progress adjustment.

[0120] In one embodiment, it is easy to understand that when the progress is adjusted, not only can the content summary and recommendation information be changed, but the questions displayed on the question control can also be changed. This allows the content displayed in the floating window to change in real time with the progress, making the content displayed in the floating window more relevant to the user's current reading or viewing, greatly improving the user experience.

[0121] In the above embodiments, by adjusting the content displayed in the floating window accordingly after the progress is adjusted, the content displayed in the floating window can be more closely related to the content currently being read or viewed by the user. This makes it easier for the user to gain a deeper and more comprehensive understanding of the content currently being read or viewed through the content displayed in the floating window, which not only improves the efficiency of information acquisition but also enhances the user experience.

[0122] In one embodiment, the method further includes: when the target page is an article reading page, in response to the selection operation of an article sentence in the currently displayed content, distinguishing and displaying the selected article sentence in the currently displayed content, and distinguishing and displaying a summary sentence generated based on the selected article sentence in the content summary.

[0123] Specifically, when the target page is an article reading page, to facilitate users in matching statements in the content summary with statements in the currently displayed content, the terminal can differentiate between the statements in the content summary and their corresponding statements in the currently displayed content. For example, a user can select some article statements from the currently displayed content, and the terminal can then differentiate between the selected article statements in the currently displayed content. When a floating window is present, the terminal can also differentiate between the summary statements obtained by summarizing the selected article statements in the content summary. Here, "article statements" refers to statements in the currently displayed content, and "summary statements" refers to statements in the content summary.

[0124] For example, refer to Figure 11 When a user selects the phrase "A notable feature of Hyundai cars is their intelligent driver assistance systems...enhancing driving comfort" from the currently displayed content, the terminal will highlight this phrase within the displayed content. Since the summary of the corresponding content, "Hyundai cars have intelligent driver assistance systems...enabling comfortable driving," is a summary of "A notable feature of Hyundai cars is their intelligent driver assistance systems...enhancing driving comfort," it will be highlighted within the summary of the content corresponding to the currently displayed content. Figure 11 A schematic diagram illustrating the distinction between statements in one embodiment is shown.

[0125] In this embodiment, the user only needs to select a portion of the article statements, and the terminal can then distinguish and display the summary statements corresponding to the selected article statements in the content summary. This simplifies the process of distinguishing and displaying summary statements and improves the efficiency of the display. Furthermore, by distinguishing and displaying the selected article statements and the summary statements, users can easily associate the article statements with the summary statements, thus obtaining the correspondence between the statements.

[0126] In one embodiment, the method further includes: when the target page is an article reading page, in response to the selection operation of an article paragraph in the currently displayed content, distinguishing and displaying the selected article paragraph in the currently displayed content, and adding a paragraph summary of the selected article paragraph in the floating window.

[0127] Specifically, when the target page is an article reading page, the floating window can display a summary of the currently displayed content, as well as a summary of the paragraph selected by the user. For example, references... Figure 12When a user wants a summary of one or more paragraphs in the currently displayed content, the user can select paragraphs from the displayed content. The terminal can then distinguish and display the selected paragraphs within the currently displayed content. The terminal inputs the selected paragraphs into a machine learning model, which outputs a summary of the selected paragraphs. The terminal can then add a summary of the selected paragraphs to a floating window. For example, the terminal can display the summary of the selected paragraphs as a conversation message from the machine learning model in the conversation display area. Figure 12 A schematic diagram illustrating the generation of paragraph summaries in one embodiment is shown.

[0128] In this embodiment, users can freely choose the content to be generated as a summary, which not only increases the freedom of summary generation but also enhances the user experience.

[0129] In one embodiment, the method further includes: in response to an exit operation on the target page, closing the floating window and displaying a content selection page; the content selection page displays multiple content titles; in response to a trigger operation on a content title, redisplaying the floating window and displaying a summary of the content corresponding to the triggered article title in the redisplaying floating window.

[0130] Specifically, when a user triggers an exit action for a target page, the terminal can respond by displaying a content selection page and closing the floating window if it is currently open. The content selection page can display multiple content titles; for example, it can display the titles of multiple articles or the titles of multiple videos. The user can trigger an action on a content title, and the terminal will respond by redisplaying the floating window and displaying a summary of the content corresponding to the triggered title. For example, a user can long-press on an article title, and the terminal will respond by redisplaying the floating window and displaying a summary of the article corresponding to the triggered title. For example, refer to... Figure 13 When a user long-presses an article title, the terminal responds to the user's long-press action by displaying a floating window, which then shows a summary of the article linked to the long-pressed article title (1301). Figure 13 The illustration shows a diagram of an article summary in one embodiment. As is easily understood, a close button is displayed on the floating window. When the user clicks the close button, the terminal closes the floating window. The user can then select a new content title and long-press it. The terminal can then re-display the floating window, displaying a summary of the content linked to the newly selected title.

[0131] In one embodiment, a floating window can be displayed to summarize the content corresponding to the triggered content title. This allows users to preview the general content before reading or watching it. When a user becomes interested in the article or video based on this preview, they can click on the content title. The terminal then responds to the click, displaying the target page and showing the content linked to the clicked title. If the user is not interested in reading or watching the content, they can close the floating window, select a different content title, and trigger the terminal to re-display the floating window, showing a summary of the content linked to the newly selected title.

[0132] In the above embodiments, since the floating window can display a content summary, users can quickly understand the outline of the article or video before reading it or watching it, thus improving the efficiency of summary retrieval. Furthermore, users can freely choose which article or video they want to understand the summary of, increasing the flexibility of summary retrieval. In addition, since the terminal can display the content summary through a floating window, users can determine whether they need to read the article or watch the video based on the displayed summary. The target page is only displayed when it is determined that reading or watching is necessary, thereby reducing unnecessary target page display processes and saving computer and other resources consumed when displaying target pages.

[0133] In one embodiment, the recommendation information is a recommended video related to the currently displayed content; the recommended video is generated based on the already delivered content related to the currently displayed content.

[0134] Specifically, recommendations can include not only text links but also videos related to the currently displayed content. For example, if the currently displayed content is a car introduction, the recommended videos could be brand introduction videos for that car. Alternatively, the recommended videos could be advertisements for that car.

[0135] In one embodiment, the recommended video can be generated based on previously delivered content related to the currently displayed content. For example, a content library can store previously delivered content. The terminal can perform relevance matching between the currently displayed content and the content features of each delivered content in the content library, and based on the matching results, filter out target delivered content related to the currently displayed content from the library. For example, one or a preset number of target delivered content can be filtered out from the library in descending order of relevance. Then, the terminal can generate a recommended video related to the currently displayed content based on the filtered target delivered content. For example, if the target delivered content is a video and only one target delivered content is filtered out, then that target delivered content can be directly used as the recommended video related to the currently displayed content. For example, if the target delivered content is an advertisement video delivered by an advertiser and only one advertisement video is filtered out, then that filtered advertisement video is used as the recommended video related to the currently displayed content, and the filtered advertisement video is played in a floating window.

[0136] When the target content is a video, but multiple target content options are selected, the terminal can generate recommended videos related to the currently displayed content based on these multiple target content options. For example, when the target content is a video and multiple target content options are selected, the terminal can display information about these multiple target content options, allowing the user to choose one. The terminal then uses the target content corresponding to the user's selected option as a recommended video related to the currently displayed content. Alternatively, the terminal can rate the selected target content options based on their information and use the highest-rated target content as a recommended video related to the currently displayed content. For instance, the target content could be an ad video placed by an advertiser. The terminal can rate each ad video based on its historical views, placement value, etc., and use the ad video with the highest rating as a recommended video related to the currently displayed content.

[0137] When the target content is text, the terminal can generate recommended videos related to the currently displayed content based on the text. For example, the terminal can input the target content as video description text into a diffusion model, which then generates recommended videos. Similarly, when multiple target content items are selected, each item can be rated, and the highest-rated item can be input into the diffusion model to generate recommended videos. For instance, the target content could be an image / text ad placed by an advertiser. The terminal can rate each ad based on factors such as view count and ad placement value, and input the highest-rated ad into the diffusion model to generate recommended videos related to the currently displayed content.

[0138] In the above embodiments, by generating a recommended video and playing it in a floating window, users can intuitively view the information recommended by the terminal based on the currently displayed content through the played recommended video. Furthermore, due to the strong visual impact, high interactivity, and rich information delivery of the video, the recommendation effect can be improved. In addition, regardless of whether the content is video or text / images, recommended videos can be generated based on the content, further enhancing the flexibility of recommended video generation. In one embodiment, the method further includes: in response to an archiving operation on the content summary, displaying a successful archiving prompt and adding the summary information of the content summary to the archive list; and in response to a sharing operation on the summary information in the archive list, triggering the sharing of the content summary corresponding to the summary information of the sharing operation to the target sharing pair specified by the sharing operation.

[0139] Specifically, content summaries can be archived, and the terminal can respond to the archiving operation by archiving the content summary. For example, when a user long-presses a content summary, the terminal can display an archive button. When the user clicks the archive button, the terminal can archive the content summary. Furthermore, the terminal can display a successful archiving message, confirm the summary information of the content summary, and add the summary information to the archive list. The summary information may include the archiving time, the title of the currently displayed content corresponding to the summary, the summary's identifier, etc. The archive list is a list that records the summary information of successfully archived content summaries.

[0140] Furthermore, users can trigger the display of an archive list on the terminal. When a user wishes to share a summary of content, they can initiate a sharing action. For example, a user can long-press on a summary in the archive list, and the terminal will treat this long-press as a sharing action. Alternatively, each summary in the archive list can have a corresponding share button. Clicking the share button for a summary confirms that the user has triggered a sharing action. Furthermore, in response to the sharing action, the terminal can display candidate users for sharing. The user can select one or more users from these candidates as target sharing users, and the terminal will then share the summary content corresponding to the summary that triggered the sharing action to the target sharing user.

[0141] In an easy-to-understand way, unarchived content summaries can also be shared. For example, when a content summary is displayed in a floating window, the user can long-press the content summary, and then the terminal can display a share button. When the user clicks the share button, the terminal can display sharing candidate users. The user can select one or more users as target sharing users from the sharing candidate users, and then the terminal will share the content summary displayed in the floating window to the target sharing users.

[0142] In the above embodiments, archiving content summaries facilitates subsequent viewing by users. For example, users can trigger the display of an archive list on the terminal, and then select the content summary they wish to view again from the summary information displayed in the archive list. This not only improves the utilization rate of content summaries but also enhances the user experience. Furthermore, since content summaries can be shared, they can be viewed by more users, further improving their utilization rate. In one embodiment, displaying the content summary corresponding to the currently displayed content in a floating window includes: determining key statements based on the currently displayed content; generating a content summary corresponding to the currently displayed content based on the key statements; and displaying the content summary corresponding to the currently displayed content in a floating window.

[0143] Specifically, the terminal can determine the currently displayed content and extract key sentences from it. If the currently displayed content is reading material, the terminal can directly extract key sentences; if the currently displayed content is a currently playing video, the terminal can convert the video into text and extract key sentences from the converted text. For example, the terminal can perform text recognition on the audio of the currently playing video to obtain the text. Furthermore, the terminal generates a content summary based on the key sentences and displays the summary corresponding to the currently displayed content in a floating window.

[0144] In one embodiment, when the currently displayed content is the currently being read content, after the user clicks on the floating component in the target page, the terminal can display the target page and show the currently displayed content on the target page. The terminal can determine the structural layout of the article containing the currently displayed content and determine whether the structural layout of the article containing the currently displayed content is a preset target layout. The structural layout of the article refers to the organization and arrangement of the various parts within the article, which determines the overall framework and logical order of the article. The structural layout of the article includes a general-to-specific layout, a specific-to-general layout, or a parallel layout, etc. A parallel layout means that the several parts of the article do not have a hierarchical relationship but are independent of each other. The preset target layout can specifically be a general-to-specific layout, a specific-to-general layout, or a general-to-specific-to-general layout.

[0145] Furthermore, when the structure and layout of the article containing the currently displayed content are within a preset target layout, the terminal can obtain the currently displayed content and its context, such as the previous and next paragraphs. The terminal then concatenates the currently displayed content and its context according to the reading order to obtain a concatenated text. This concatenated text is then input into a content summarization machine learning model, which outputs a summary of the currently displayed content. For example, the model can filter out key sentences from the concatenated text and concatenate them according to the reading order to obtain a summary of the currently displayed content, which the terminal can then display in a floating window.

[0146] In one embodiment, when the currently displayed content is the currently read content, the structural layout of the article containing the currently displayed content can be determined by examining the various headings within the article. For example, if all headings in the article are at the same level, the article's structural layout can be determined to be a parallel layout; if the headings in the article include both main headings and subheadings, and the main headings contain subheadings, the article's structural layout can be determined to be a general-to-specific layout.

[0147] In one embodiment, when the currently displayed content is the currently read content, semantic analysis can be performed on the first sentence or the first few sentences of each paragraph in the article containing the currently displayed content. The structural layout of the article can be determined based on the semantic analysis results. For example, if the article comprises three paragraphs, semantic analysis can be performed on the first sentence of the first paragraph, the first sentence of the second paragraph, and the first sentence of the third paragraph. If the semantics of the first sentence of the first paragraph includes the semantics of the first sentences of the second and third paragraphs, then the structural layout of the article is determined to be a general-to-specific layout. If the semantics of the first sentences of the first, second, and third paragraphs are independent and have no relation, then the structural layout of the article is determined to be a parallel layout.

[0148] In one embodiment, when the currently displayed content is the currently read content, the structural layout of the article containing the currently displayed content may not be known. The currently displayed content can be directly input into the content summary machine learning model, and the content summary machine learning model can output the content summary corresponding to the currently displayed content.

[0149] In the above embodiments, by determining the structural layout of the article, it can be assumed that the content summary output based solely on the currently displayed content would be inaccurate if the structural layout of the article is determined to be the target layout. Therefore, by obtaining the context of the currently displayed content, the content summary corresponding to the currently displayed content can be generated with the help of the context of the currently displayed content, thereby improving the accuracy of the content summary.

[0150] In one embodiment, when the currently displayed content is the currently read content, key sentences are filtered from the currently displayed content and its context, including: determining the frequency of occurrence of each word in the currently displayed content and its context, and determining the position information, length information, and sentence similarity of each sentence in the currently displayed content and its context; and combining the frequency of occurrence of each word, the position information of each sentence, the length information of each sentence, and the sentence similarity of each sentence to filter key sentences from the currently displayed content and its context.

[0151] Specifically, the content summarization machine learning model can segment the concatenated text into multiple sentences, resulting in a sentence set. For each sentence in the sentence set, the content summarization machine learning model can perform word segmentation on each sentence, dividing it into multiple words. Based on the segmented words, the model can determine the frequency of each word in the concatenated text, i.e., the word frequency, and also determine the inverse document frequency (IDF) of each word. The IDF represents the rarity of a word in the concatenated text. The IDF can be obtained using the formula IDF(word) = log(total number of sentences / number of sentences containing the word). Furthermore, for each word in the concatenated text, the model can multiply the word frequency and the IDF to obtain the word's importance within the sentence, i.e., the word's significance.

[0152] For each sentence in the concatenated text, the content summarization machine learning model determines the importance of the sentence based on the importance of each word in that sentence. For example, the importance of each word in the sentence can be summed to obtain the overall importance of the sentence.

[0153] Furthermore, the content summarization machine learning model can also determine the position of each sentence in the concatenated text, the length of each sentence, and the similarity between sentences in the concatenated text. By combining the importance, position, length, and similarity of each sentence in the concatenated text, key sentences can be filtered out. For example, the first and last sentences in the concatenated text, or the first and last sentences of each paragraph, often contain important information. Therefore, each sentence can be scored based on its position to obtain a position score. For instance, if a sentence is the first or last sentence in the concatenated text, or the first or last sentence of a paragraph, its position score is higher than that of sentences that are not the first or last sentences. Sentences in the concatenated text can also be scored based on their length to obtain a length score. For example, sentences of moderate length are usually more likely to contain complete ideas, while very short or very long sentences may contain less key information or redundant information. Therefore, when the number of words contained in a sentence is within a preset range, its length score is higher than that of other sentences. Each sentence can be scored based on its similarity to others, resulting in a similarity score. For example, the more similar sentences a sentence has, the more important it is considered, and thus its similarity score can be increased. After obtaining the importance, position, length, and similarity scores of each sentence in the concatenated text, key sentences can be identified based on these scores. For instance, the importance, position, length, and similarity scores of each sentence can be summed to obtain a summed score, and key sentences can be selected in descending order of summed score.

[0154] In the above embodiments, key sentences are selected by comprehensively considering the frequency of occurrence of each word in the spliced ​​text, as well as the position information, length information, and sentence similarity of each sentence in the spliced ​​text, making the selected key sentences more accurate.

[0155] In one embodiment, the current content features of the currently displayed content are determined, and the similarity between the current content features and candidate content features in a preset feature library is determined; at least one candidate content feature is selected from the preset feature library based on the similarity, and specific recommended content is determined based on the selected at least one candidate content feature; a content summary of the specific recommended content is generated, and a recommended text link is generated based on the content summary of the specific recommended content, and the recommended text link is displayed in a floating window.

[0156] Specifically, when a user clicks on a floating component on a target page, the terminal can display the target page and show the currently displayed content. Further, the terminal can input the currently displayed content into a recommendation information generation model to generate recommendations. For example, the recommendation information generation model can perform convolution processing on the currently displayed content to obtain its content features. For ease of description, the content features of the currently displayed content will be referred to as current content features. Further, the terminal can pre-store a feature library; for ease of description, the content features stored in the feature library will be referred to as candidate content features. The recommendation information generation model can calculate the similarity between the current content features and the candidate content features in the feature library. For example, the recommendation information generation model can perform a dot product operation between the current content features and the candidate content features to obtain the similarity between the current content features and the candidate content features.

[0157] Furthermore, the recommendation information generation model can select at least one candidate content feature from a pre-defined feature library based on similarity. For example, the model can extract a predetermined number of candidate content features in descending order of similarity. After obtaining the predetermined number of candidate content features, the model can determine the specific recommended content based on these features. For instance, the model obtains the specific recommended content corresponding to each selected candidate content feature and randomly selects one from these selected candidate content features as the specific recommended content.

[0158] Furthermore, the recommendation information generation model can generate a summary of the specific recommended content and a recommended text link. The text in the recommended text link serves as the summary of the specific recommended content. For example, the recommendation information generation model can use the title of the specific recommended content as its summary. Alternatively, the model can obtain a pre-set summary for the specific recommended content. Or, the model can summarize the content of the specific recommended content to generate its own summary. Then, based on the webpage address corresponding to the specific recommended content and the summary of the specific recommended content, the recommended text link generated by the model can be displayed on the terminal in a floating window. When the user clicks the recommended text link, the terminal can be redirected to the page containing the specific recommended content.

[0159] In one embodiment, determining specific recommended content based on at least one selected candidate content feature includes: obtaining the interest tags of the currently logged-in account on the target page; selecting target content features from the at least one selected candidate content features based on the interest tags; and generating specific recommended content based on the target content features.

[0160] Specifically, the recommendation information generation model can also obtain the interest tags of the currently logged-in account on the target page. These interest tags can be tags selected by the user through the interest tag selection page and authorized by the recommendation information generation model. Further, the recommendation information generation model can determine the interest tags corresponding to each of the selected candidate content features based on the preset correspondence between content features and interest tags. For ease of description, the interest tags corresponding to candidate content features are referred to as candidate interest tags; the interest tags corresponding to the currently logged-in account are referred to as current interest tags. The recommendation information model determines whether there is a target interest tag among the candidate interest tags that matches the current interest tag. If so, the candidate content feature corresponding to the target interest tag is used as the target content feature. If no matching target interest tag exists, the candidate interest tag with the highest similarity to the current interest tag is used as the target interest tag, and the candidate content feature corresponding to the target interest tag is used as the target content feature.

[0161] Furthermore, the recommendation information generation model determines the specific recommended content corresponding to the target content features based on the pre-defined correspondence between content features and specific recommended content, and generates recommended text links based on the specific recommended content.

[0162] In the above embodiments, by combining interest tags to determine specific recommended content, the generated specific recommended content is more in line with the user's reading interests, thereby improving the effectiveness of information recommendation.

[0163] In one embodiment, reference Figure 14 , Figure 14 An interactive schematic diagram of a floating window display method in one embodiment is shown. The floating window display method is implemented through a floating window display system, which includes a reading module, a widget module, a summary generation module, an information recommendation module, and an interaction module. The reading module displays article content and supports user scrolling; it is responsible for receiving user reading behavior data, such as scroll position and dwell time. The widget module is embedded in the reading module and is used to generate article summaries in real time and push relevant recommendation information; it includes a summary generation module and an information recommendation module. The summary generation module is used for...

[0164] Based on natural language processing technology, the system analyzes article content in real time and generates a concise summary. Machine learning algorithms are used to perform semantic analysis on the article content and extract key information. The ad push module generates relevant recommended text links based on article content and user interests, displaying them along with the summary. Recommendation algorithms match relevant recommended content based on user reading behavior and interest tags. The user interaction module allows users to click on the summary and recommended text links to obtain more information or jump to relevant pages; it records user click behavior and feedback data to optimize the information recommendation algorithm. The specific workflow is as follows: 1. User enters the reading page: The user opens the instant messaging application, selects an article to start reading; the reading module displays the article content and launches the widget module. 2. Real-time generation of article summary: The widget module monitors the user's reading progress and passes the currently displayed content to the summary generation module. The summary generation module analyzes the currently displayed content and generates a concise article summary. 3. Display of relevant recommended text links: The information recommendation module generates recommended text links for relevant content based on article content and user interests; the widget module displays the generated article summary and recommended text links in a floating window on the target page. 4. Users can view summaries and advertisements: During reading, users can click on the floating window at any time to view article summaries and recommended text links. Users can receive replies to their questions through the chat message input box. 5. Users can click on recommended text links for more information: Clicking on a recommended text link will redirect the user to a related recommendation display page, showing detailed recommended content. 6. Continue reading or return: After viewing the summary and recommended content, users can continue reading the article or return to the article list.

[0165] In one specific embodiment, a method for displaying a floating window is provided, including:

[0166] The terminal displays the target page; the target page displays the currently displayed content and floating components.

[0167] The terminal responds to the triggered operation on the floating component by switching the floating component to a floating window and displaying the floating window on the target page.

[0168] The terminal responds to a drag operation on the floating window by moving the floating window along the drag trajectory; or responds to a zoom operation on the floating window by scaling the floating window according to the zoom ratio corresponding to the zoom operation; the floating window includes a session display area.

[0169] The terminal determines the frequency of each word in the text corresponding to the currently displayed content, as well as the position, length, and similarity of each sentence in the text. The text corresponding to the currently displayed content can be the currently displayed content and its context, or it can be video text. The terminal combines the frequency of each word, the position of each sentence, the length of each sentence, and the similarity of each sentence to filter out key sentences from the text corresponding to the currently displayed content; and generates a summary of the content corresponding to the currently displayed content based on the key sentences.

[0170] The terminal determines the current content features of the currently displayed content and the similarity between the current content features and the candidate content features in the preset feature library; based on the similarity, it selects at least one candidate content feature from the preset feature library to obtain the interest tags corresponding to the currently logged-in account on the target page.

[0171] The terminal selects target content features from at least one candidate content feature based on interest tags, and generates specific recommended content based on the target content features. It then generates a content summary of the specific recommended content and a recommended text link based on the content summary.

[0172] The terminal will display at least one of the following in the conversation display area: a summary of the content currently displayed and recommendation information related to the currently displayed content. This summary will be used as a conversation message from the machine learning model.

[0173] When the recommended information is a text link containing specific recommended content, the terminal responds to the trigger operation of the recommended text link by redirecting to a redirect page and displaying specific recommended content related to the currently displayed content on the redirect page; in response to the closing operation of the redirect page, the terminal closes the redirect page, redisplays the target page, and displays a floating window on the target page.

[0174] The terminal responds to the trigger operation of the conversation input box by displaying the question entered through the conversation input box as the user's conversation message in the conversation display area; and displays the answer to the question as the conversation message of the machine learning model in the conversation display area.

[0175] The machine learning model's session messages also display at least one question button; in response to the triggering operation of the question button, the terminal displays the answer corresponding to the predicted question displayed on the triggered question button as the machine learning model's session message in the session display area.

[0176] In response to a progress adjustment operation on the currently displayed content, the terminal switches the currently displayed content to the content with the adjusted progress; and displays at least one of the following in a floating window: a summary of the content with the adjusted progress and related recommendations.

[0177] In response to an exit operation on the target page, the terminal closes the floating window and displays the content selection page; the article selection page displays multiple content titles; in response to a trigger operation on a content title, the floating window is displayed again, and a summary of the content corresponding to the triggered content title is displayed in the displayed floating window.

[0178] The beneficial effects of the above embodiments are as follows: 1. Provides real-time summary function: By embedding a small component module, the content currently being read or viewed by the user is analyzed in real time, generating a concise and clear summary to help users quickly obtain core information. Users can view the summary at any time during reading or viewing, improving the efficiency of information acquisition, especially suitable for long articles, long videos, or time-sensitive scenarios. 2. Achieves accurate push of recommended information: Through the information recommendation module, relevant recommended text links or recommended videos are generated based on the specific content and interests currently being read or viewed by the user, ensuring that the recommended content is highly matched with the user's current needs; improving the click-through rate of recommended text links, and enhancing the effectiveness and value of information recommendations. 3. Improves user experience: Displaying a floating window in the reading or viewing module will not interrupt the user's reading or video viewing experience. Users can view the summary and recommended information at any time; by clicking on the summary and recommended text links, users can obtain more information or jump to relevant pages, enhancing the interactivity and convenience of the reading or video viewing process; providing personalized services and information recommendations to meet the diverse needs of users during the reading or video viewing process, and improving user satisfaction. In summary, this embodiment addresses the problems of lack of real-time summarization, inaccurate information recommendations, and poor user experience in existing technologies by providing real-time summary functionality and accurate information recommendations. By generating content summaries and related recommendations in real time, it improves the user's reading experience, video viewing experience, and information recommendation effectiveness, demonstrating broad application prospects.

[0179] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0180] This application also provides an application scenario in which the above-described floating window display method is applied. Specifically, the application of the floating window display method in this scenario is as follows:

[0181] Users can access a public account through a communication application and view its published articles. Users can select one article to read, and the terminal will then display the article content (i.e., the currently displayed content) and a floating component. When a user wants to view a summary of the article's content, they can click the floating component, which will then display a floating window summarizing the article's content. Simultaneously, the terminal can also display recommended text links related to the article within this floating window. When a user clicks a recommended text link, the terminal will redirect to the corresponding page, and the terminal can then provide relevant services through this redirected page.

[0182] This application also provides another application scenario in which the above-described floating window display method is applied. Specifically, the floating window display method is applied in this application scenario as follows:

[0183] The above application scenarios are merely illustrative. It is understood that the application of the floating window display method provided in the various embodiments of this application is not limited to the above scenarios.

[0184] Game applications sometimes display target pages. For example, puzzle games might display articles introducing the game's background on the target page. A floating widget can be displayed on the target page, allowing users to click on it when they want to view a summary of the content displayed on the target page. The device can then respond to this click by displaying a floating window that shows a summary of the content and related recommendations.

[0185] Based on the same inventive concept, this application also provides a floating window display device for implementing the floating window display method described above. The solution provided by this device is similar to the implementation solution described in the above method. Therefore, the specific limitations of one or more floating window display device embodiments provided below can be found in the limitations of the floating window display method above, and will not be repeated here.

[0186] In one embodiment, such as Figure 15 As shown, a floating window display device 1500 is provided, including: a floating component display module 1502, a floating window display module 1504, and an information display module 1506, wherein:

[0187] The floating component display module 1502 is used to display the target page, showing the currently displayed content and the floating component on the target page.

[0188] The floating window display module 1504 is used to display the floating window in response to a trigger operation on the floating component.

[0189] The information display module 1506 is used to display at least one of the following in a floating window: a summary of the currently displayed content and recommended information related to the currently displayed content.

[0190] In one embodiment, the floating window display module 1504 is further configured to, in response to a trigger operation on the floating component, switch the floating component to a floating window and display the floating window on the target page; and, in response to a closing operation on the floating window, close the floating window and redisplay the floating component.

[0191] In one embodiment, the floating window display module 1504 is further configured to move the floating window according to the drag trajectory in response to a drag operation on the floating window; or to scale the floating window according to the scaling ratio corresponding to the scaling operation in response to a scaling operation on the floating window.

[0192] In one embodiment, the recommendation information is a text link of specific recommended content; the information display module 1506 is also used to display a content summary corresponding to the currently displayed content and a text link related to the currently displayed content in a floating window; the text link displays a content summary of specific recommended content related to the currently displayed content; in response to a trigger operation on the text link, the specific recommended content related to the currently displayed content is displayed.

[0193] In one embodiment, the floating window display device 1500 further includes a setting module for responding to a display content setting operation for the floating window, displaying a content summary option and a recommendation information option; the content summary option is used to trigger the floating window to display a content summary when selected; the recommendation information option is used to trigger the floating window to display recommendation information when selected.

[0194] In one embodiment, the floating window includes a session display area; the information display module 1506 is further configured to display at least one of the content summary corresponding to the currently displayed content and the recommendation information related to the currently displayed content as a session message of the machine learning model in the session display area.

[0195] In one embodiment, the floating window further includes a conversation input area; a conversation input box is displayed in the conversation input area; the information display module 1506 is also used to respond to a trigger operation on the conversation input box, display the question entered through the conversation input box as a user's conversation message in the conversation display area; and display the answer corresponding to the question as a conversation message of the machine learning model in the conversation display area.

[0196] In one embodiment, the machine learning model's session message also displays at least one question button; each question button displays a predicted question based on the currently displayed content, and different question buttons display different predicted questions; the information display module 1506 is also used to respond to a trigger operation on a question button by displaying the answer corresponding to the predicted question displayed on the triggered question button as a session message of the machine learning model in the session display area.

[0197] In one embodiment, the information display module 1506 is further configured to, in response to a progress adjustment operation on the currently displayed content, switch the currently displayed content to the content after progress adjustment; and display at least one of the content summary and related recommendation information corresponding to the content after progress adjustment in a floating window.

[0198] In one embodiment, the information display module 1506 is further configured to, in response to an operation of selecting an article sentence in the currently displayed content, distinguish and display the selected article sentence in the currently displayed content, and distinguish and display a summary sentence generated based on the selected article sentence in the content summary, when the target page is an article reading page.

[0199] In one embodiment, the information display module 1506 is further configured to, in response to an operation of selecting an article paragraph in the currently displayed content, distinguish and display the selected article paragraph in the currently displayed content, and add a paragraph summary of the selected article paragraph in the floating window when the target page is an article reading page.

[0200] In one embodiment, the information display module 1506 is further configured to close the floating window and display the content selection page in response to an exit operation on the target page; the article selection page displays multiple content titles; in response to a trigger operation on the content title, the floating window is re-displayed, and a summary of the content corresponding to the triggered content title is displayed in the re-displayed floating window.

[0201] In one embodiment, the recommendation information is a recommended video related to the currently displayed content; the recommended video is generated based on the already delivered content related to the currently displayed content.

[0202] In one embodiment, the floating window display device 1500 further includes an archiving module, which, in response to an archiving operation for a content summary, displays a successful archiving prompt and adds the summary information of the content summary to the archive list; and, in response to a sharing operation for the summary information in the archive list, shares the content summary corresponding to the summary information that triggered the sharing operation to the target sharing object specified by the sharing operation.

[0203] In one embodiment, the floating window display device 1500 further includes a summary generation module, which is used to determine key sentences based on the currently displayed content; generate a content summary corresponding to the currently displayed content based on the key sentences; and display the content summary corresponding to the currently displayed content in the floating window.

[0204] In one embodiment, the floating window display device 1500 further includes a summary generation module, which, when the currently displayed content is the currently read content, determines the frequency of occurrence of each word in the currently displayed content and the context, and determines the position information, length information, and sentence similarity of each sentence in the currently displayed content and the context; and, by combining the frequency of occurrence of each word, the position information of each sentence, the length information of each sentence, and the sentence similarity of each sentence, selects key sentences from the currently displayed content and the context.

[0205] In one embodiment, the floating window display device 1500 further includes a link generation module, which is used to determine the current content features of the currently displayed content and determine the similarity between the current content features and candidate content features in a preset feature library; filter at least one candidate content feature from the preset feature library based on the similarity; determine specific recommended content based on the filtered at least one candidate content feature; generate a content summary of the specific recommended content and generate a recommended text link based on the content summary of the specific recommended content.

[0206] In one embodiment, the floating window display device 1500 further includes a link generation module, which is used to obtain the interest tags corresponding to the currently logged-in account on the target page; select target content features from at least one selected candidate content features based on the interest tags; and generate specific recommended content based on the target content features.

[0207] The various modules in the aforementioned floating window display device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0208] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 16As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a floating window display method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0209] Those skilled in the art will understand that Figure 16 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0210] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0211] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0212] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including 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 steps in the above method embodiments.

[0213] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0214] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0215] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0216] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for displaying a floating window, characterized in that, The method includes: Display the target page, and on the target page display the currently displayed content and the floating component; In response to a trigger operation on the floating component, a floating window is displayed; The floating window displays at least one of the following: a summary of the currently displayed content and recommended information related to the currently displayed content.

2. The method according to claim 1, characterized in that, The step of displaying a floating window in response to a trigger operation on the floating component includes: In response to a trigger operation on the floating component, the floating component is switched to a floating window, and the floating window is displayed on the target page. The method further includes: In response to the closing operation of the floating window, the floating window is closed and the floating component is displayed again.

3. The method according to claim 2, characterized in that, The method further includes: In response to a drag operation on the floating window, move the floating window along the drag trajectory; or In response to a scaling operation on the floating window, the floating window is scaled according to the scaling ratio corresponding to the scaling operation.

4. The method according to claim 1, characterized in that, The recommended information is a recommended text link related to the currently displayed content; the step of displaying at least one of the content summary corresponding to the currently displayed content and the recommended information related to the currently displayed content in the floating window includes: The floating window displays a summary of the currently displayed content and a recommended text link; the text displayed on the recommended text link is a summary of the specific recommended content related to the currently displayed content. The method further includes: In response to a triggering action on the recommended text link, specific recommended content related to the currently displayed content is displayed.

5. The method according to claim 1, characterized in that, The method further includes: In response to the operation of setting the display content for the floating window, a content summary option and a recommendation information option are displayed; the content summary option is used to trigger the floating window to display a content summary when selected; the recommendation information option is used to trigger the floating window to display recommendation information when selected.

6. The method according to claim 1, characterized in that, The floating window includes a session display area; displaying at least one of the following in the floating window: a content summary corresponding to the currently displayed content and recommendation information related to the currently displayed content: At least one of the content summary corresponding to the currently displayed content and the recommendation information related to the currently displayed content is displayed as a conversation message of the machine learning model in the conversation display area.

7. The method according to claim 6, characterized in that, The floating window also includes a conversation input area; the conversation input area displays a conversation input box; the method further includes: In response to a trigger operation on the conversation input box, the question entered through the conversation input box is displayed as a user's conversation message in the conversation display area; The answers to the questions are displayed as conversation messages of the machine learning model in the conversation display area.

8. The method according to claim 6, characterized in that, The machine learning model's session messages also display at least one question button; each question button displays a predicted question based on the currently displayed content, and different question buttons display different predicted questions. The method further includes: In response to the triggering operation of the question button, the answer corresponding to the predicted question displayed on the triggered question button is displayed as a session message of the machine learning model in the session display area.

9. The method according to claim 1, characterized in that, The method further includes: In response to a progress adjustment operation on the currently displayed content, the currently displayed content is switched to the content after the progress adjustment; The floating window displays at least one of the following: a summary of the content corresponding to the progress adjustment and related recommendations.

10. The method according to claim 1, characterized in that, The method further includes: When the target page is an article reading page, in response to the selection operation of an article sentence in the currently displayed content, the selected article sentence is displayed separately in the currently read content, and a summary sentence generated based on the selected article sentence is displayed separately in the content summary.

11. The method according to claim 1, characterized in that, The method further includes: When the target page is an article reading page, in response to the selection operation of an article paragraph in the currently displayed content, the selected article paragraph is displayed separately in the currently displayed content, and a paragraph summary of the selected article paragraph is added to the floating window.

12. The method according to claim 1, characterized in that, The method further includes: In response to an exit operation on the target page, the floating window is closed and a content selection page is displayed; the content selection page displays multiple content titles; In response to the trigger operation on the content title, the floating window is redisplayed, and a summary of the content corresponding to the triggered title is displayed in the redisplayed floating window.

13. The method according to claim 1, characterized in that, The recommendation information consists of recommended videos related to the currently displayed content; the recommended videos are generated based on the already delivered content related to the currently displayed content.

14. The method according to claim 1, characterized in that, The method further includes: In response to the archiving operation for the content summary, a successful archiving message is displayed, and the summary information of the content summary is added to the archive list; In response to a sharing operation for the summary information in the archive list, the content corresponding to the summary information that triggered the sharing operation is summarized and shared to the target sharing object specified by the sharing operation.

15. The method according to claim 1, characterized in that, The summary of the content corresponding to the currently displayed content shown in the floating window includes: Determine the key statements based on the currently displayed content; A summary of the content currently displayed is generated based on the key statement, and the summary of the content currently displayed is displayed in the floating window.

16. The method according to claim 15, characterized in that, The step of determining key statements based on the currently displayed content includes: When the currently displayed content is the currently read content, determine the frequency of occurrence of each word in the currently displayed content and its context, and determine the position information, length information, and sentence similarity of each sentence in the currently displayed content and its context; By combining the frequency of occurrence of each word, the position information of each statement, the length information of each statement, and the statement similarity of each statement, key statements are selected from the currently displayed content and the context.

17. The method according to claim 1, characterized in that, The recommended information consists of recommended text links related to the currently displayed content; The steps for generating the recommended text links include: Determine the current content features of the currently displayed content, and determine the similarity between the current content features and candidate content features in a preset feature library; At least one candidate content feature is selected from the preset feature library based on the similarity. The specific recommended content is determined based on at least one feature of the selected candidate content. Generate a content summary of the specific recommended content, and generate recommended text links based on the content summary of the specific recommended content.

18. The method according to claim 17, characterized in that, The step of determining specific recommended content based on at least one selected candidate content feature includes: Retrieve the interest tags corresponding to the currently logged-in account on the target page; Based on the interest tags, target content features are selected from at least one candidate content feature, and specific recommended content is generated based on the target content features.

19. A display device with a suspended window, characterized in that, The device includes: The floating component display module is used to display the target page, showing the currently displayed content and the floating component on the target page; A floating window display module is used to display a floating window in response to a trigger operation on the floating component; An information display module is used to display, in the floating window, at least one of the following: a content summary corresponding to the currently displayed content and recommended information related to the currently displayed content.

20. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 18.

21. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 18.

22. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 18.

Images