Image-text data recommendation method, device and equipment and storage medium

Abstract

The application discloses a kind of graphic information recommendation method, device, equipment and storage medium, wherein, graphic information recommendation method includes: obtaining the first demand text that user inputs in the input box of search interface, wherein, first demand text is unstructured text;First graphic information recommended for user is displayed based on first demand text in search interface.The above scheme, only need to obtain the unstructured first demand text that user inputs in the input box of search interface, i.e. first graphic information is recommended for user, without user definition, even replace keyword for information retrieval to search, improve the efficiency of user obtaining graphic information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a method and device for recommending graphic and text materials, an electronic device and a storage medium. BACKGROUND

[0002] Reading graphic and text materials is one of the main ways for people to obtain information nowadays, for example, users can understand or learn relevant information by reading books, newspapers and periodicals.

[0003] However, users often search for graphic and text materials to be read through terminal devices. For example, users usually need to define keywords such as title, publisher and author for material search, and input the keywords in the search box of the terminal device, so as to obtain the corresponding graphic and text materials. The efficiency is low by using this way to obtain graphic and text materials. SUMMARY

[0004] The technical problem solved by the present application is to provide a method and device for recommending graphic and text materials, and an electronic device and a storage medium, which can improve the efficiency of users obtaining graphic and text materials.

[0005] In order to solve the above technical problem, the first aspect of the present application provides a method for recommending graphic and text materials, comprising: obtaining a first demand text input by a user in an input box of a search interface; wherein the first demand text is an unstructured text; and displaying a first graphic and text material recommended for the user in the search interface based on the first demand text.

[0006] In order to solve the above technical problem, the second aspect of the present application provides a device for recommending graphic and text materials, comprising: an obtaining module, configured to obtain a first demand text input by a user in an input box of a search interface; wherein the first demand text is an unstructured text; and a recommending module, configured to display a first graphic and text material recommended for the user in the search interface based on the first demand text.

[0007] In order to solve the above technical problem, the third aspect of the present application provides an electronic device, comprising a human-computer interaction circuit, a memory and a processor, the human-computer interaction circuit and the memory are respectively coupled to the processor, the memory stores program instructions, and the processor is configured to execute the program instructions to realize the method for recommending graphic and text materials of the first aspect.

[0008] In order to solve the above technical problem, the fourth aspect of the present application provides a computer readable storage medium, which stores program instructions capable of being executed by a processor, and the program instructions are used to realize the method for recommending graphic and text materials of the first aspect.

[0009] The scheme is characterized in that: the first requirement text input by the user in the input box of the search interface is acquired, and the first graphic-text material recommended for the user is displayed on the search interface based on the first requirement text. The first requirement text is unstructured text. In this way, the first graphic-text material can be recommended for the user only by acquiring the unstructured first requirement text input by the user in the input box of the search interface, without the need for the user to define or even replace the keyword for material retrieval to search, thereby improving the efficiency of the user in acquiring the graphic-text material.

[0010] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, rather than limiting the present application. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the technical solutions of the present application.

[0012] Figure 1 is a flowchart of an embodiment of a graphic-text material recommendation method provided by the present application;

[0013] Figure 2 is a flowchart of another embodiment of a graphic-text material recommendation method provided by the present application;

[0014] Figure 3 is a schematic diagram of a main interface provided by the present application;

[0015] Figure 4 is a schematic diagram of a search interface provided by the present application;

[0016] Figure 5 is a schematic diagram of another search interface provided by the present application;

[0017] Figure 6 is a schematic diagram of another search interface provided by the present application;

[0018] Figure 7 is a schematic diagram of another search interface provided by the present application;

[0019] Figure 8 is a flowchart of another embodiment of a graphic-text material recommendation method provided by the present application;

[0020] Figure 9 is a flowchart of another embodiment of a graphic-text material recommendation method provided by the present application;

[0021] Figure 10 is a flowchart of another embodiment of a graphic-text material recommendation method provided by the present application;

[0022] Figure 11is a flowchart of an embodiment of the information recommendation device provided in the present application;

[0023] Figure 12 is a framework diagram of an embodiment of the electronic device provided in the present application;

[0024] Figure 13 is a framework diagram of an embodiment of the computer readable storage medium provided in the present application. DETAILED DESCRIPTION

[0025] The scheme of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0026] In the following description, specific details such as specific system structures, interfaces, techniques, etc. are presented in order to provide a thorough understanding of the present application for the sake of explanation, but not for the sake of limitation.

[0027] The term "and / or" herein is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, "multiple" herein means two or more than two. In addition, the term "at least one" herein means any combination of any one or at least two of multiple, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C. "Several" means at least one. The terms "first", "second" and the like in the specification and claims herein and the above-described drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0028] Please refer to Figure 1 , Figure 1 is a flowchart of an embodiment of the information recommendation method provided in the present application. The method can be executed by a terminal device, which can be a notebook, a mobile phone, a notebook computer, a tablet computer, etc. for example, and the present embodiment does not make specific limitation on this. It should be noted that the method of the present application is not limited to the flow order shown in Figure 1 . As shown in Figure 1 , the method comprises the following steps:

[0029] S11: obtaining a first demand text input by a user in an input box of a search interface.

[0030] The first demand text is used to describe the reading demand of the user.

[0031] It should be noted that the first demand text in this embodiment is an unstructured text. For example, the first demand text is "I recently learn to cook, please recommend some books suitable for learning" and "I often feel anxious recently, what books should I read?" and the like. Unlike unstructured text, structured text includes key information of the picture-text material, such as the name, publisher, author and the like of the picture-text material.

[0032] In this embodiment, the terminal device can display a search interface, and the search interface is provided with an input box. The terminal device can obtain the first demand text input by the user in the input box of the search interface. For example, when the user clicks the input box of the search interface, the terminal device can pop up a voice input interface, obtain and recognize the voice about reading demand input by the user in the voice input interface to obtain the recognized text. Then, the recognized text is taken as the first demand text, or the recognized text is regularized to obtain the first demand text. For another example, when the user clicks the input box of the search interface, the terminal device can pop up a text input interface, and obtain the text about reading demand manually input by the user in the text input interface as the first demand text.

[0033] S12: Based on the first demand text, display the first picture-text material recommended for the user on the search interface.

[0034] Exemplarily, the first picture-text material can include but is not limited to books, news, blog articles, public number articles, newspapers and periodicals, and the like, and the embodiment is not limited specifically in this regard.

[0035] In this embodiment, the first demand text can be specifically input into a large language model, and the large language model is used to obtain the first picture-text material related to the reading demand of the user. The large language model refers to a computer model capable of processing and generating natural language. Exemplarily, the large language model can include but is not limited to pre-trained language models such as GPT (Generative Pre-trained Transformer, generative pre-trained transformer) series (such as ChatGPT) and MOSS (Multilingual Open-Source Synthesizer, multilingual open-source synthesizer). Since the large language model has strong knowledge ability, learning ability, logical reasoning ability and understanding ability, after the first demand text is input into the large language model, the large language model can quickly and accurately understand the reading demand of the user, and recommend the first picture-text material related to the reading demand of the user.

[0036] After obtaining the first picture-text material, the first picture-text material can be displayed on the search interface to facilitate the user to read the related information of the first picture-text material.

[0037] In this embodiment, the first demand text input by the user in the input box of the search interface is acquired, and the first image-text material recommended for the user is displayed on the search interface based on the first demand text. The first demand text is unstructured text. In this way, the first image-text material can be recommended for the user by only acquiring the unstructured first demand text input by the user in the input box of the search interface, without the need for the user to define or even replace the keyword for material retrieval to search, thereby improving the efficiency of the user in acquiring image-text material.

[0038] Referring to Figure 2 , Figure 2 is a flowchart of another embodiment of the image-text material recommendation method provided in the present application. The method can be executed by a terminal device. For related content of the terminal device, refer to the embodiment shown in Figure 1 , which will not be described here again. As shown in Figure 2 , the method comprises the following steps:

[0039] S21: In response to the click operation of the user on the search box in the main interface, jump from the main interface to the search interface.

[0040] The terminal device can display a main interface for image-text material recommendation for the user, and the main interface comprises a search box. When the user clicks the search box, the main interface jumps to the search interface.

[0041] Referring to Figure 3 , Figure 3 is a schematic diagram of the main interface provided in the present application. Figure 3 The box with the number 1 in the main interface shown in Figure 3 is the search box. When the user clicks the area where the search box is located, the main interface automatically jumps to the search interface. It should be noted that Figure 3 only the book recommendation is exemplarily described, the main interface shown in

[0042] is a main interface applied to book recommendation.

[0043] The related content of the first demand text can refer to the aforementioned step S11, which will not be described here again. In this embodiment, the search interface is provided with an input box, and the first demand text input by the user in the search interface can be acquired. Referring to Figure 4 , Figure 4 is a schematic diagram of the search interface provided in the present application. Figure 4 The box with the number 2 in the search interface shown in

[0044] In an embodiment, the first demand text is obtained by acquiring and recognizing the voice input by the user in the voice input interface about the user's reading demand. Specifically, in response to the user's click operation on the input box, the voice input interface is popped up on the search interface; the first demand text is obtained by recognizing the user's voice based on the language option selected by the user in the voice input interface and the collected user's voice, and the recognized first demand text is displayed in the input box. The voice input interface is provided with a plurality of language options. For example, the language options can include Mandarin, dialect, ethnic language, simultaneous interpretation (such as English to Chinese, Chinese to audio), and foreign language, etc.

[0045] Please refer to Figure 5 , Figure 5 is another schematic diagram of the search interface provided by the present application. As shown in Figure 5 , the search interface includes an input box 2 and a voice input interface 3. When the user clicks on the input box 2 of the search interface, the voice input interface 3 is popped up. The user can select a language option and input a voice about the user's reading demand in the voice input interface 3. For example, the voice input interface 3 can display a language option button and a voice input button (not shown in the figure), and the user can select the required language option by clicking the language option button and start inputting the voice about the user's reading demand by clicking the voice input button.

[0046] In the present embodiment, the step of obtaining the first demand text based on the language option selected by the user in the voice input interface and the collected user's voice includes: identifying the user's voice based on the language option selected by the user in the voice input interface to obtain the recognized text; and taking the recognized text as the first demand text. Alternatively, the user's voice can also be identified based on the language option selected by the user in the voice input interface to obtain the recognized text; and the first demand text is obtained by normalizing the recognized text. For example, the recognized text can be normalized by deleting redundant words, correcting text errors, and rearranging the order of words, etc. For example, exclamation words in the recognized text are deleted, and inverted sentences in the text are corrected. In this example, the recognized text after normalization is taken as the first demand text, which can facilitate the large language model to more accurately understand the user's reading demand.

[0047] In another embodiment, the first demand text can also be obtained by acquiring the text input by the user in the text input interface about the reading demand. Specifically, in response to the user's click operation on the input box, the text input interface is popped up on the search interface; the first demand text is obtained based on the text input by the user in the text input interface. In the present embodiment, the text input by the user in the text input interface can be directly taken as the first demand text, or the text input by the user in the text input interface can also be normalized as described above, and the normalized text is taken as the first demand text.

[0048] In this embodiment, the text input interface may include an input method keyboard or a handwriting area. For example, when a user clicks the input box, an input method keyboard pops up in the search interface to support the user in manually inputting the first desired text using the input method keyboard. Alternatively, when a user clicks the input box, a handwriting area pops up in the search interface to support the user in inputting the first desired text by handwriting.

[0049] Optionally, in this embodiment, the acquired first requirement text is input into a large language model to obtain the first image and text data. To facilitate users quickly inputting the first requirement text in the search interface, and to facilitate the large language model understanding the user's reading needs based on the first requirement text, sample requirement text can be displayed in the search interface. For example, such as... Figure 4 and Figure 5 As shown, the sample request text displayed may include "I recently want to learn about cooking, please recommend some suitable books for me" or "I've been feeling anxious lately, what books should I read?".

[0050] S23: Based on the first requirement text, display the first image and text materials recommended to the user on the search interface.

[0051] In this embodiment, after obtaining the first requirement text entered by the user in the search interface, the first requirement text is input into the large language model, and the large language model is used to obtain the first text and image materials related to the user's reading needs. Alternatively, after receiving the user's confirmation operation on the first requirement text in the search interface, the first requirement text is input into the large language model. For example, a send button can be displayed in the search interface, such as in the input box. When the user clicks the send button, it is determined that the user's confirmation operation on the first requirement text has been received, and then the first requirement text is input into the large language model. The relevant content of the large language model can be referred to in the aforementioned step S12, and a detailed description is omitted here. The number of first text and image materials recommended to the user is at least one.

[0052] Optionally, in this embodiment, the search interface is provided with a dialogue area, which can be used to recommend the first text and image materials in the form of dialogue interaction.

[0053] In one embodiment, after receiving the first request text entered by the user in the search interface, in response to the user's confirmation of the first request text in the search interface, the first request text is displayed as the user's first dialogue text in the dialog area. Before receiving the first recommended image and text materials for the user, a first prompt text is displayed in the dialog area as the second dialogue text responding to the first dialogue text. The first prompt text prompts the user to wait for the recommended first image and text materials, and after receiving the recommended first image and text materials, the first recommended image and text materials are displayed in the dialog area as the new second dialogue text.

[0054] For example, the search interface displays a confirmation button or a send button in the input box. After obtaining the first request text entered by the user in the search interface, when the user clicks the confirmation button or send button, it is determined that the user's confirmation operation on the first request text has been received. The first request text is displayed first in the dialog area, followed by the first prompt text.

[0055] like Figure 6 As shown, Figure 6 This is another schematic diagram of the search interface provided in this application. For example... Figure 6 As shown, the first request text obtained is "I've recently started learning to cook, please recommend some suitable books for me to learn from." This first request text is displayed as the first dialogue text in the dialogue area of ​​the search interface, and the first prompt text "Thinking..." is displayed in the dialogue area of ​​the search interface.

[0056] In one embodiment, after obtaining the first text and image materials recommended to the user, the descriptive text of the first text and image materials can also be displayed in the dialog area of ​​the search interface. Specifically, this may include the following sub-steps:

[0057] Sub-step one: Extract the summary text of each of the first image and text materials.

[0058] For example, the large language model can summarize the abstract text of each first-level text based on its specific content. Alternatively, it can retrieve the abstract text of each first-level text from the internet based on its key information. For example, the key information of the first-level text includes its title, author, publisher, etc.

[0059] Sub-step two: Select the summary text and attribute text belonging to the same first image and text material as the description text of the corresponding first image and text material.

[0060] For example, the attribute text of the first image and text material includes at least one of the title of the first image and text material and the author.

[0061] Sub-step three, the introduction text of each first information is displayed in the search interface as the dialogue text responding to the first demand text.

[0062] In an example, the introduction text of each first information can be directly displayed in the dialogue area of the search interface, so as to facilitate the user to quickly understand the key information of each first information.

[0063] In another example, in order to further improve the convenience of the user reading the first information, the introduction text and the link button of each first information can be displayed in the search interface, and the user can open the detailed introduction interface of the first information by clicking the link button of the first information. If the first information exists in the preset information base, the link button corresponding to the first information is displayed in a triggerable state, indicating that the first information can be found in the preset information base. If the first information does not exist in the preset information base, the link button corresponding to the first information is displayed in a non-triggerable state, indicating that the first information cannot be found in the preset information base.

[0064] Further, after displaying the introduction text of each first information in the search interface as the dialogue text responding to the first demand text, it further includes: in response to the clicking operation of the user on the link button in the search interface, based on the triggerable state of the link button, jumping to the introduction interface of the first information to which the link button belongs. The user can read the first information in the introduction interface after jumping.

[0065] Please refer to Figure 7 , Figure 7 is another schematic diagram of the search interface provided by the present application. As shown in Figure 7 , the search interface includes a dialogue area. The first demand text displayed in the dialogue area is "I have been learning cooking recently, please recommend some books suitable for learning for me". The dialogue text in the dialogue area responding to the first demand text includes the introduction texts of three books "Shouyuan Shidan", "Cooking Complete Manual" and "Cooking Manual", and the introduction text of each book includes the abstract text, title, author and link button of each book. Exemplarily, as shown in Figure 7 , when the link button is in a triggerable state, the first link prompt text such as "go and see" is displayed on the link button, and the first link prompt text is used to prompt the user to click the link button to read the corresponding first information. When the link button is in a non-triggerable state, the second link prompt text such as "book city is not on the shelf" is displayed on the link button, and the second link prompt text is used to prompt the user that the first information of the link button cannot be obtained at present.

[0066] In yet another example, to further improve the recommendation effect of the graphic materials and enhance the user experience, the introduction text and the progress button of each first graphic material can be displayed on the search interface. If the user has read the first graphic material, the progress button corresponding to the first graphic material prompts the reading progress. For example, the progress percentage can be displayed on the progress button of each first graphic material to prompt the user's reading progress of each first graphic material. For example, 0% indicates that the user has not read the first graphic material, and 100% indicates that the user has read the first graphic material.

[0067] Further, after displaying the introduction text of each first graphic material as the dialogue text responding to the first demand text on the search interface, the method further includes: in response to the user's click operation on the progress button on the search interface, jumping to the reading interface of the first graphic material to which the progress button belongs based on the progress button representing that the first graphic material has been read. The display content of the reading interface matches the reading progress prompted by the progress button, so that the user can start reading from the unread part of the first graphic material without starting from the beginning, further improving the user experience.

[0068] Optionally, to further improve the recommendation effect of the graphic materials and enhance the user experience, in the embodiment, after obtaining the first graphic materials recommended for the user, the reading order of each first graphic material can be recommended to the user. For example, the large language model can recommend the first graphic materials to the user in the order from easy to difficult according to the reading difficulty of each first graphic material. Alternatively, the large language model can recommend the first graphic materials to the user in the order from low to high according to the reading depth of each first graphic material. Alternatively, the large language model can recommend the first graphic materials to the user in the order from early to late or from late to early according to the publication time of each first graphic material.

[0069] In the embodiment, on the one hand, the first demand text is obtained by interacting with the search interface, and the first graphic materials recommended for the user are displayed on the search interface based on the first demand text. In this process, only the unstructured first demand text input by the user in the input box on the search interface is needed to recommend the first graphic materials to the user, without the need for the user to define or even replace the keywords for material retrieval to search, improving the efficiency of the user in obtaining the graphic materials. On the other hand, the process of recommending the first graphic materials to the user in the form of dialogue interaction on the search interface can improve the recommendation effect of the graphic materials and enhance the user experience.

[0070] Please refer to Figure 8 , Figure 8 is a flowchart of another embodiment of the graphic material recommendation method provided by the present application. For example, the method can be executed by a terminal device, and the related content of the terminal device is described in detail in the foregoing embodiments.Figure 1 In the embodiment shown, the method includes the following steps: Figure 8 In the embodiment shown, the method includes the following steps:

[0071] S81: In response to the user's profile search operation after determining the to-do item, determine that the user's reading demand is related to the to-do item.

[0072] In this embodiment, the user can make notes using the terminal device, for example, taking the terminal device as an office book, the user carries the office book when attending a meeting, and the office book can automatically extract and record the to-do items mentioned in the meeting. Or the user can directly record the to-do items on the terminal device. Or record the to-do items on the terminal device. The user may need to do some material preparation work to complete the to-do items, such as searching for materials related to the to-do items. If the user performs a material search operation after determining the to-do item, it is considered that the user wants to search for materials related to the to-do item.

[0073] S82: Generate a candidate text based on the to-do item.

[0074] The candidate text is used to describe the user's reading demand for the to-do item. For example, if the user determines that the to-do item is to learn about neural network related knowledge, the candidate text generated based on the to-do item can be "I want to understand the content related to neural networks, please recommend suitable books for me."

[0075] In an embodiment, a large language model can be used to generate a candidate text based on a to-do item. Specifically, after determining that the user's reading demand is related to the to-do item, a first instruction text for instructing the large language model to generate a candidate text based on the to-do item is generated, the first instruction text is input into the large language model, and a candidate text describing the user's reading demand output by the large language model is obtained.

[0076] S83: Obtain a first demand text based on the user's target operation on the candidate text.

[0077] The target operation includes any one of a confirmation operation and an editing operation. For example, the editing operation can include adding text, deleting text, and modifying text, etc.

[0078] In one embodiment, after generating candidate text, it can be displayed in the input box of the search interface, and the candidate text displayed in the input box is in an editable state. When the user determines that the candidate text does not meet the user's reading needs or needs further improvement, the candidate text in the input box can be edited to make the edited first requirement text more in line with the user's reading needs. Based on the first requirement text, the first text and image materials can be recommended to the user more accurately. When the user determines that the candidate text meets their reading needs, they can perform a confirmation operation, for example, by clicking the confirmation button or send button on the aforementioned search interface, to input the candidate text as the first requirement text into the large language model for first text and image material recommendation.

[0079] Optionally, in this embodiment, step S83 is an optional step, or step S83 can be omitted, and the candidate text generated in step S82 can be directly used as the first required text.

[0080] S84: Based on the first text requirement, display the first image and text materials recommended to the user on the search interface.

[0081] The relevant content of step S84 can be referred to the aforementioned step S23, and will not be repeated here.

[0082] In this embodiment, after a user confirms a task and performs a data search, the system determines that the user's reading needs are related to the task. Based on the task, candidate text describing the reading needs can be automatically generated, and then, based on the candidate text, the system recommends the first set of text and image materials to the user. Compared to... Figure 2 The illustrated embodiment eliminates the need for users to manually input their first requirement text, further improving the recommendation effect of text and image materials.

[0083] Please see Figure 9 , Figure 9 This is a schematic flowchart of another embodiment of the image and text data recommendation method provided in this application. Exemplarily, this method can be executed by a terminal device; for details regarding the terminal device, please refer to [link / reference]. Figure 1 The illustrated embodiments will not be described in detail here. Figure 9 As shown, the method includes the following steps:

[0084] S91: Obtain the first request text entered by the user in the input box of the search interface.

[0085] The relevant content can be found in steps S21 and S22 above, and will not be repeated here.

[0086] S92: Retrieve user's record data.

[0087] Exemplarily, the record data of the user can include, but is not limited to, meeting records, notes, to-do lists, and the like. The record data of the user can be obtained from a local storage unit. Alternatively, the record data of the user is backed up in the cloud, and the record data of the user can also be obtained from the cloud.

[0088] Optionally, before performing step S92, further comprising: requesting the user to obtain the obtaining permission of the record data. Exemplarily, displaying a permission option on the display interface of the terminal device, the permission option including an agree-to-obtain option and a disagree-to-obtain option. When detecting that the user clicks the agree-to-obtain option, it is determined that the obtaining permission of the record data is obtained, and the record data of the user is obtained. When detecting that the user does not click the agree-to-obtain option or detecting that the user clicks the disagree-to-obtain option, it is determined that the obtaining permission of the record data is not obtained, and the record data of the user is not obtained. At this time, only the step of S93 is performed, that is, only the first graph-text material recommended for the user is recommended based on the first demand text.

[0089] In the embodiment, in order to improve the possibility of obtaining the record data, the user can also be prompted that “the authorization helps to more accurately realize the recommendation of the graph-text material” when requesting the user to obtain the permission.

[0090] S93: obtaining the first graph-text material recommended for the user based on the first demand text.

[0091] The related content can refer to the foregoing step S23, and will not be described here again.

[0092] S94: obtaining the relevance between each first graph-text material and the record data.

[0093] Exemplarily, the relevance between each first graph-text material and the record data can be analyzed by using a large language model. Alternatively, the relevance between each first graph-text material and the record data can also be predicted by using a neural network model.

[0094] S95: recommending the first graph-text material for the user in the order from high to low according to the relevance.

[0095] For example, the first graph-text material with high relevance to the record data can be recommended in the front, and the first graph-text material with low relevance to the record data can be recommended in the back.

[0096] Alternatively, step S95 can also be replaced by: recommending the first news and picture materials with the highest relevance to the record data to the user. Alternatively, step S95 can also be replaced by: recommending the first news and picture materials with the relevance greater than a preset threshold to the user. The preset threshold can be set according to actual needs. Alternatively, step S95 can also be replaced by: recommending the first news and picture materials with the ranking position of the relevance greater than a preset position to the user, for example, recommending the first news and picture materials with the ranking position of the relevance in the top three to the user. The preset position can be set according to actual needs.

[0097] In this embodiment, after the first news and picture materials recommended for the user are obtained based on the first demand text, the first news and picture materials can be recommended to the user according to the relevance of each first news and picture material to the record data of the user. In this way, the first news and picture materials with a higher relevance to the record data of the user can be preferentially recommended to the user, and the recommendation effect of the news and picture materials is further improved.

[0098] Please refer to Figure 10 , Figure 10 is a flowchart of another embodiment of the news and picture material recommendation method provided in the present application. The method can be executed by the terminal device described above, and the method is executed after the first news and picture materials are recommended for the user, for example, after the method shown in Figure 2 、 Figure 8 or Figure 9 is executed. As shown in Figure 10 , the method includes the following steps:

[0099] S101: Obtain the browsing data of the user for each first news and picture material.

[0100] In this embodiment, the browsing data can include but is not limited to the browsing time of the user for the first news and picture material, the number of times of clicking the first news and picture material, the number of times of exiting the first news and picture material, and the browsing mark, etc. Among them, the longer the browsing time of the user for the first news and picture material, the more the number of times of clicking or exiting the first news and picture material, and the more the browsing mark indicate that the reference value of the first news and picture material is higher. The browsing mark can be a mark made when reading the first news and picture material, for example, a highlight mark.

[0101] S102: Based on the first demand text and the browsing data, display the second news and picture materials recommended for the user on the search interface.

[0102] Specifically, step S102 includes: generating a second demand text that is more detailed than the first demand text based on the first demand text and the browsing data; and displaying the second news and picture materials recommended for the user on the search interface based on the second demand text.

[0103] For example, based on the first demand text, the first image-text material A, the first image-text material B and the first image-text material C are obtained, wherein the first image-text material C corresponds to the longest browsing time. The generated second demand text can be "recommend me image-text materials more similar to the first image-text material C". For another example, based on the first demand text, the first image-text material A, the first image-text material B and the first image-text material C are obtained, wherein the first image-text material B corresponds to the longest browsing time, and the first image-text material B includes the highlight mark made by the user. The generated second demand text can be "recommend me image-text materials more similar to the first image-text material B". Or, further analysis shows that the first image-text material B relates to the theme D and the highlight E, and "image-text materials related to the theme D and the highlight E" can be added as the second demand text based on the first demand text. It can be seen that in the embodiment, the second demand text description is more accurate and detailed than the first demand text.

[0104] In an embodiment, the second demand text can be generated by using a large language model. Specifically, the second indication information for instructing the large language model to generate the second demand text can be generated based on the first demand text and the browsing data, and the second indication text is input into the large language model to obtain the second demand text output by the large language model.

[0105] Further, after generating the second demand text, the second demand text can be displayed in the input box of the search interface. The displayed candidate text in the input box is in an editable state. When the user determines that the second demand text does not meet the user's reading demand or needs to be improved, the second demand text in the input box can be edited so that the edited second demand text meets the user's reading demand more accurately, and then based on the second demand text, the second image-text material can be more accurately recommended to the user. When the user determines that the second demand text meets the user's reading demand, a confirmation operation can be performed, for example, clicking the confirmation button or the send button of the search interface to input the second demand text into the large language model for first image-text material recommendation.

[0106] Optionally, in the embodiment, the process of recommending the second image-text material can be realized in the form of dialogue interaction in the dialogue area of the search interface.

[0107] In an embodiment, after the second demand text is acquired, in response to a confirmation operation of the second demand text by the user on the search interface, the second demand text is displayed as third dialogue text of the user in the dialogue area, and before the second graphic-text material recommended for the user is acquired, the second prompt text is displayed as fourth dialogue text responding to the third dialogue text in the dialogue area. The second prompt text is used to prompt the user to wait for the recommended second graphic-text material, and after the second graphic-text material recommended for the user is acquired, the second graphic-text material recommended for the user is displayed as new fourth dialogue text in the dialogue area.

[0108] Exemplarily, the input box of the search interface displays a confirmation button or a send button. After the second demand text is acquired, when it is detected that the user clicks the confirmation button or the send button, it is determined that the confirmation operation of the second demand text by the user is received, and the second demand text is displayed in the dialogue area first, and then the second prompt text is displayed.

[0109] In an embodiment, after the second graphic-text material is acquired based on the first demand text and the browsing data, introduction text of the second graphic-text material can also be displayed in the dialogue area of the search interface. Specifically, the introduction text of each second graphic-text material can be extracted respectively; the abstract text and the attribute text belonging to the same second graphic-text material are selected as the introduction text of the corresponding second graphic-text material; and the introduction text of each second graphic-text material is displayed as dialogue text responding to the second demand text in the search interface. Exemplarily, the attribute text of the second graphic-text material includes at least one of the title and the author of the second graphic-text material.

[0110] In an example, the introduction text of each second graphic-text material can be directly displayed in the dialogue area of the search interface, so as to facilitate the user to quickly understand the key information of each second graphic-text material.

[0111] In another example, in order to further improve the convenience of the user reading the second graphic-text material, the introduction text and the link button of each second graphic-text material can be displayed on the search interface, and the user can open the detailed introduction interface of the second graphic-text material by clicking the link button of the second graphic-text material. If the second graphic-text material exists in the preset database, the link button corresponding to the second graphic-text material is displayed in a triggerable state, indicating that the second graphic-text material can be found in the preset database. If the second graphic-text material does not exist in the preset database, the link button corresponding to the second graphic-text material is displayed in a non-triggerable state, indicating that the second graphic-text material cannot be found in the preset database.

[0112] Further, after displaying the introduction text of each second news and information as the dialogue text responding to the second demand text on the search interface, the method further includes: in response to a click operation of the user on the link button on the search interface, jumping to the introduction interface of the second news and information to which the link button belongs based on the link button being in a triggerable state. The user can read the second news and information in the introduction interface after the jump.

[0113] In yet another example, to further improve the recommendation effect of news and information and enhance user experience, the introduction text of each second news and information and a progress button can be displayed on the search interface. If the user has read the second news and information, the progress button corresponding to the second news and information prompts the reading progress. For example, the progress percentage can be displayed on the progress button of each second news and information to prompt the reading progress of each second news and information to the user.

[0114] Further, after displaying the introduction text of each second news and information as the dialogue text responding to the second demand text on the search interface, the method further includes: in response to a click operation of the user on the link button on the search interface, jumping to the introduction interface of the second news and information to which the link button belongs based on the link button being in a triggerable state. The user can read the second news and information in the introduction interface after the jump.

[0115] Optionally, to further improve the recommendation effect of news and information and enhance user experience, in this embodiment, after obtaining the second news and information based on the second demand text, the reading order of each second news and information can be recommended to the user. For example, the large language model can recommend the reading order of each second news and information to the user according to the reading difficulty of each second news and information from easy to difficult. Alternatively, the large language model can recommend the reading order of each second news and information to the user according to the reading depth of each second news and information from low to high. Alternatively, the large language model can recommend the reading order of each second news and information to the user according to the publication time of each second news and information from early to late or from late to early.

[0116] In this embodiment, after obtaining the first news and information based on the first demand text, the second demand text is generated based on the first demand text and the browsing data of the user on each first news and information, and the second news and information is recommended to the user based on the second demand text. Since the second demand text is more accurate and detailed than the first demand text, the second news and information recommended based on the second demand text is more accurate, further improving the recommendation effect of news and information.

[0117] Please refer to Figure 11 , Figure 11is a framework schematic diagram of an embodiment of the information recommendation device provided in the present application. As shown in Figure 11 The information recommendation device 110 includes an acquisition module 111 and a recommendation module 112. The acquisition module 111 is configured to acquire a first demand text input by a user in an input box of a search interface, the first demand text being an unstructured text. The recommendation module 112 is configured to display, based on the first demand text, a first information recommended for the user in the search interface.

[0118] Optionally, the acquisition module 111 is configured to, in response to a click operation of the user on the search box in the main interface, jump from the main interface to the search interface, and perform the step of acquiring the first demand text input by the user in the input box of the search interface; wherein the first demand text and the first information recommended for the user are displayed in a dialogue form in a dialogue area of the search interface.

[0119] Optionally, the acquisition module 111 is configured to, in response to a click operation of the user on the input box, pop up a voice input interface in the search interface; wherein the voice input interface is provided with a plurality of language options; based on the language option selected by the user in the voice input interface and the user voice collected, display the first demand text recognized in the input box.

[0120] Optionally, the information recommendation device 110 further includes a dialogue text display module 113. The dialogue text display module 113 is configured to extract an abstract text of each first information respectively; select the abstract text and an attribute text belonging to the same first information as an introduction text of the corresponding first information; wherein the attribute text includes at least one of a title and an author of the first information; and display the introduction text of each first information as dialogue text responding to the first demand text in the search interface.

[0121] Optionally, the dialogue text display module 113 is configured to display the introduction text and a link button of each first information in the search interface; wherein if the first information exists in a preset information base, the link button corresponding to the first information is displayed in a triggerable state, and if the first information does not exist in the preset information base, the link button corresponding to the first information is displayed in a non-triggerable state. The information recommendation device 110 further includes a first information display module 114, configured to, in response to a click operation of the user on the link button in the search interface, jump to an introduction interface of the first information to which the link button belongs based on the link button being in the triggerable state.

[0122] Optionally, the dialogue text display module 113 is configured to display introduction texts of the respective first newsfeed and progress buttons on the search interface; wherein if the first newsfeed has been read by the user, the progress button corresponding to the first newsfeed prompts the reading progress. The first newsfeed display module 114 is configured to, in response to the user's clicking operation on the progress button on the search interface, jump to the reading interface of the first newsfeed to which the progress button belongs based on the progress button representing that the first newsfeed has been read; wherein the display content of the reading interface matches the reading progress prompted by the progress button.

[0123] Optionally, the acquisition module 111 is further configured to acquire browsing data of the user on the respective first newsfeed; wherein the browsing data at least includes the browsing time length. The recommendation module 112 is further configured to display the second newsfeed recommended for the user on the search interface based on the first demand text and the browsing data.

[0124] Optionally, the recommendation module 112 is configured to generate a second demand text which is more detailed than the first demand text based on the first demand text and the browsing data; and display the second newsfeed recommended for the user on the search interface based on the second demand text.

[0125] Optionally, the newsfeed recommendation apparatus 110 further comprises a demand text display module 115. After the recommendation module 112 generates a second demand text which is more detailed than the first demand text based on the first demand text and the browsing data, and before the second newsfeed recommended for the user is displayed on the search interface based on the second demand text, the demand text display module 115 is configured to display the second demand text in the input box on the search interface; wherein the second demand text and the second newsfeed recommended for the user are displayed in a dialogue form on the search interface.

[0126] Optionally, the acquisition module 111 is further configured to acquire record data of the user. The recommendation module 112 is further configured to acquire the first newsfeed recommended for the user based on the first demand text; acquire the respective correlation between the first newsfeed and the record data; and recommend the first newsfeed for the user in the order from high to low according to the correlation.

[0127] Optionally, the acquisition module 111 is configured to, in response to the user's newsfeed searching operation after determining the to-do list, determine that the reading demand of the user is related to the to-do list, and generate a candidate text based on the to-do list; and obtain the first demand text based on the target operation of the user on the candidate text; wherein the target operation includes any one of the confirmation operation and the editing operation.

[0128] It should be noted that the apparatus of the present embodiment can execute the steps in the above-mentioned method, and the detailed description of the related content can be found in the above-mentioned method part, which will not be repeated here.

[0129] Please refer toFigure 12 , Figure 12 is a framework schematic diagram of an embodiment of the electronic device provided by the present application. The electronic device 120 comprises a human-computer interaction circuit 123, a memory 121 and a processor 122, the human-computer interaction circuit 123 and the memory 121 are respectively coupled to the processor 122, the memory 121 stores program instructions, and the processor 122 is configured to execute the program instructions to implement the steps in any of the above-mentioned graphic-text material recommendation method embodiments.

[0130] Specifically, the processor 122 is configured to control itself, the memory 121 and the human-computer interaction circuit 123 to implement the steps in any of the above-mentioned graphic-text material recommendation method embodiments. The processor 122 can also be referred to as a CPU (Central Processing Unit). The processor 122 can be an integrated circuit chip with a processing capability of signals. The processor 122 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 122 can be realized by integrated circuit chips together.

[0131] Please refer to Figure 13 , Figure 13 is a framework schematic diagram of an embodiment of the computer readable storage medium provided by the present application. The computer readable storage medium 130 of the embodiment of the present application stores program instructions 131, which are executed to implement the graphic-text material recommendation method provided by the present application. Wherein, the program instructions 131 can form a program file and be stored in the above-mentioned computer readable storage medium 130 in the form of a software product, so that a computer device (which can be a personal computer, a server, or a network device, etc.) executes all or part of the steps of the method of each embodiment of the present application. And the aforementioned computer readable storage medium 130 includes: a U disk, a mobile hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk or an optical disk, etc. various media that can store program codes, or a computer, a server, a mobile phone, a tablet, etc. terminal device.

[0132] The above scheme, by acquiring the first requirement text input by the user in the input box of the search interface, and based on the first requirement text, displaying the first image-text material recommended for the user in the search interface. The first requirement text is unstructured text. In this way, only the unstructured first requirement text needs to be acquired, and the first image-text material can be recommended for the user without the user defining or even replacing the keyword for material retrieval to search, improving the efficiency of the user in acquiring the image-text material.

[0133] In some embodiments, the apparatus provided by the embodiments of the present disclosure has functions or includes modules that can be used to perform the methods described in the above method embodiments, and the specific implementation can refer to the description of the above method embodiments. For brevity, it will not be repeated here.

[0134] The above description of various embodiments tends to emphasize the differences between various embodiments, and the same or similar parts can be mutually referred to. For brevity, it will not be repeated here.

[0135] In several embodiments provided in the present application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device implementation described above is only schematic; for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed mutual units can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0136] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e. can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment scheme.

[0137] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0138] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0139] The above is only the embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent flow transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for recommending textual and image resources, characterized in that, include: Obtain the first requirement text entered by the user in the input box of the search interface; wherein, the first requirement text is unstructured text; Based on the first requirement text, the first image and text materials recommended to the user are displayed on the search interface; Obtain user browsing data for each of the first text and image materials; wherein the browsing data includes at least the browsing duration; Based on the first requirement text and the browsing data, a second requirement text that is more detailed than the first requirement text is generated; The second requirement text is displayed in the input box of the search interface; Based on the second requirement text, the search interface displays second text and image materials recommended to the user; wherein, the second requirement text and the second text and image materials recommended to the user are displayed in a dialog form on the search interface.

2. The method according to claim 1, characterized in that, The step of obtaining the first request text entered by the user in the input box of the search interface includes: In response to the user's click on the search box on the main interface, the system jumps from the main interface to the search interface and performs the step of obtaining the first requirement text entered by the user in the input box of the search interface. The first request text and the first image and text materials recommended to the user are displayed in a dialogue format in the dialogue area of ​​the search interface.

3. The method according to claim 1 or 2, characterized in that, The step of obtaining the first request text entered by the user in the input box of the search interface includes: In response to a user's click action in the input box, a voice input interface pops up on the search interface; wherein, the voice input interface has several language options; Based on the language option selected by the user in the voice input interface and the collected user voice, the first required text is displayed in the input box.

4. The method according to claim 1, characterized in that, The method further includes: Extract the summary text of each of the first image and text materials; Select the summary text and attribute text belonging to the same first image and text material as the description text of the corresponding first image and text material; wherein, the attribute text includes at least one of the title and author of the first image and text material; The introductory text of each of the first graphic materials is displayed as dialogue text in response to the first request text on the search interface.

5. The method according to claim 4, characterized in that, The step of displaying the descriptive text of each of the first graphic and textual materials as dialogue text in response to the first request text on the search interface includes: The search interface displays the description text and link button for each of the first text and image materials; wherein, if the first text and image materials exist in the preset database, the link button for the first text and image materials is displayed as a triggerable state, and if the first text and image materials do not exist in the preset database, the link button for the first text and image materials is displayed as an untriggerable state. After displaying the introductory text of each of the first graphic and textual materials as dialogue text responding to the first request text on the search interface, the method further includes: In response to a user clicking the link button on the search interface, and based on the link button being in a triggerable state, the user is redirected to the introduction interface of the first graphic document to which the link button belongs.

6. The method according to claim 4, characterized in that, The step of displaying the descriptive text of each of the first graphic and textual materials as dialogue text in response to the first request text on the search interface includes: The search interface displays the description text and progress button for each of the first text and image materials; wherein, if the user has read the first text and image material, the reading progress is displayed on the progress button of the corresponding first text and image material. After displaying the introductory text of each of the first graphic and textual materials as dialogue text responding to the first request text on the search interface, the method further includes: In response to a user clicking the progress button on the search interface, based on the progress button indicating that the text has been read, the user is redirected to the reading interface of the first text and image document to which the progress button belongs; wherein the content displayed on the reading interface matches the reading progress indicated by the progress button.

7. The method according to claim 1, characterized in that, After obtaining the first demand text entered by the user in the input box of the search interface, and before displaying the first text and image materials recommended to the user on the search interface based on the first demand text, the method further includes: Retrieve user's record data; The first set of text and image materials recommended to the user on the search interface based on the first demand text includes: Based on the first requirement text, obtain the first image and text materials recommended to the user; Obtain the correlation between each of the first text and image materials and the recorded data; The first set of text and image materials is recommended to the user in descending order of relevance.

8. The method according to claim 1, characterized in that, The step of obtaining the first request text entered by the user in the input box of the search interface includes: In response to a user's data search operation after a to-do item is identified, the system determines that the user's reading needs are related to the to-do item and generates candidate text based on the to-do item. The first required text is obtained based on the user's target operation on the candidate text; wherein the target operation includes either a confirmation operation or an editing operation.

9. A device for recommending graphic and textual materials, characterized in that, include: The acquisition module is used to acquire the first demand text entered by the user in the input box of the search interface; wherein, the first demand text is unstructured text; The recommendation module is used to display recommended image and text materials to the user on the search interface based on the first demand text. The acquisition module is also used to acquire browsing data of the user on each of the first text and image materials; wherein the browsing data includes at least the browsing duration; The recommendation module is also used to generate a second demand text that is more detailed than the first demand text description based on the first demand text and the browsing data; The image and text recommendation device also includes a demand text display module, which is used to display the second demand text in the input box of the search interface; The recommendation module is further configured to display second text and image materials recommended to the user on the search interface based on the second demand text; wherein the second demand text and the second text and image materials recommended to the user are displayed in a dialog form on the search interface.

10. An electronic device, characterized in that, The method includes a human-computer interaction circuit, a memory, and a processor. The human-computer interaction circuit and the memory are respectively coupled to the processor. The memory stores program instructions, and the processor is used to execute the program instructions to implement the image and text data recommendation method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The device stores program instructions that can be executed by a processor, the program instructions being used to implement the image and text material recommendation method according to any one of claims 1 to 8.

Images