Recommendation method, recommendation device, electronic equipment and computer readable storage medium

Abstract

The invention relates to a recommendation method, a recommendation device, electronic equipment and a computer readable storage medium. The method comprises the steps of determining a to-be-recommended text of a user; obtaining a target multimedia resource associated with the to-be-recommended text based on at least one of historical dialogue records and timeliness questions of the user and the intelligent agent and the to-be-recommended text; and displaying the to-be-recommended text and the target multimedia resource on an interactive interface of the user and the intelligent agent.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a recommendation method, a recommendation device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of artificial intelligence (AI) technology and large language models, many software that can interact with AI-driven agents have emerged. Generally speaking, these agents can recommend texts such as interactive topics to users. The user's selectivity for these texts helps to improve the retention rate of new users and low-activity users, and can inspire and guide users to develop the habit of interacting with agents in different functional scenarios, including but not limited to question-and-answer, retrieval, and text-generated images. In related interactive software, how to provide users with more personalized texts to be recommended will affect the user experience. Summary of the invention

[0003] According to some embodiments of the present disclosure, a recommendation method is provided, including: determining a user's text to be recommended; acquiring target multimedia resources associated with the text to be recommended based on at least one of the historical conversation records and timeliness issues between the user and an intelligent agent, and the text to be recommended; and displaying the text to be recommended and the target multimedia resources on an interactive interface between the user and the intelligent agent.

[0004] According to other embodiments of the present disclosure, a recommendation device is provided, including: a determination module, configured to determine a user's text to be recommended; an acquisition module, configured to acquire target multimedia resources associated with the text to be recommended based on at least one of the historical conversation records and timeliness issues between the user and the intelligent agent, and the text to be recommended; and a display module, configured to display the text to be recommended and the target multimedia resources on an interaction interface between the user and the intelligent agent.

[0005] According to some embodiments of the present disclosure, an electronic device is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute a method of any embodiment described in the present disclosure based on instructions stored in the memory.

[0006] According to some embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method of any embodiment described in the present disclosure is performed.

[0007] Other features, aspects and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The following is an explanation of the embodiments of the present disclosure with reference to the accompanying drawings. It should be understood that the drawings described below only relate to some embodiments of the present disclosure and do not constitute a limitation to the present disclosure. In the accompanying drawings:

[0009] Figure 1 A flowchart of a recommended method for user-agent interaction according to some embodiments of the present disclosure is shown;

[0010] Figure 2 A schematic diagram showing an interactive interface between a user and an agent according to some embodiments of the present disclosure;

[0011] Figure 3 A flowchart of acquiring a target multimedia resource associated with a text to be recommended according to some embodiments of the present disclosure is shown;

[0012] Figure 4 A schematic block diagram of a recommendation device for user-agent interaction according to some embodiments of the present disclosure is shown;

[0013] Figure 5 A block diagram of an electronic device according to some embodiments of the present disclosure is shown;

[0014] Figure 6 A block diagram of an electronic device according to some other embodiments of the present disclosure is shown.

[0015] It should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not necessarily drawn according to the actual proportional relationship. The same or similar reference numerals are used in the various drawings to represent the same or similar parts. Therefore, once an item is defined in one drawing, it may not be further discussed in subsequent drawings. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present disclosure. It should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein.

[0017] It should be understood that the various steps recorded in the method embodiments of the present disclosure can be performed in different orders, and / or performed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown in the execution. The scope of the present disclosure is not limited in this respect. Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments should be interpreted as being merely exemplary and do not limit the scope of the present disclosure.

[0018] The term “including” and its variations used in the present disclosure are intended to be open terms that include at least the following elements / features but do not exclude other elements / features, that is, “including but not limited to.” The term “based on” means “at least partly based on.”

[0019] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0020] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0021] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0022] The embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings, but the present disclosure is not limited to these specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. In addition, in one or more embodiments, specific features, structures or characteristics can be combined in any suitable manner that will be clear from the present disclosure by a person of ordinary skill in the art.

[0023] It should be understood that the present disclosure does not limit how to obtain the image to be applied / processed. In some embodiments of the present disclosure, it can be obtained from a storage device, such as an internal memory or an external storage device. In other embodiments of the present disclosure, a photographic component can be mobilized to shoot. It should be noted that the acquired image can be a captured image or a frame of an image in a captured video, and is not particularly limited to this.

[0024] In the context of the present disclosure, an image may refer to any of a variety of images, such as a color image, a grayscale image, etc. It should be noted that in the context of the present specification, the type of image is not specifically limited. In addition, the image may be any appropriate image, such as an original image obtained by a camera device, or an image that has been subjected to specific processing, such as preliminary filtering, anti-aliasing, color adjustment, contrast adjustment, normalization, etc. It should be noted that the preprocessing operation may also include other types of preprocessing operations known in the art, which will not be described in detail here.

[0025] In the related art, the intelligent agent interaction software usually provides users with some texts to be recommended so that users can start interactive topics in different usage scenarios. When the interactive software is further combined with the visual language model VLM (Vision Language Model) technology, the personalization of the text to be recommended can be improved by recommending candidate texts related to VLM to users. To this end, the present disclosure proposes a recommendation method, in which the multimedia resources corresponding to the text to be recommended can be displayed in the interactive interface between the user and the intelligent agent, thereby optimizing the recommended content provided to the user, improving the personalization of the interaction, and giving the user a better user experience.

[0026] Specifically, Figure 1 A flowchart of a recommended method for user-agent interaction according to some embodiments of the present disclosure is shown.

[0027] like Figure 1 As shown, in step S101, the user's text to be recommended is determined; in step S102, based on at least one of the historical conversation records and timeliness issues between the user and the intelligent agent, and the text to be recommended, the target multimedia resources associated with the text to be recommended are obtained; and in step S103, the text to be recommended and the target multimedia resources are displayed on the interaction interface between the user and the intelligent agent.

[0028] The recommendation method of this embodiment may be executed on the client side, or may be partially executed on the server side.

[0029] Generally, the text to be recommended in the present disclosure is displayed on the interactive interface in the form of a list combination, and is usually presented near the user's current interactive content as a module for implementing the recommendation function. The historical conversation record mainly includes the interactive content that has been completed between the user and the agent in the previous use process, including but not limited to the text and / or multimedia resources involved in the interaction. Timeliness issues mainly include news events that occur at the current time point of the interaction or hot topics and discussion focuses on social platforms. The target multimedia resources displayed on the interactive interface are in various forms, including but not limited to static pictures, dynamic images, short videos, slides, virtual reality / augmented reality content, social media content, animated films, interactive elements, etc. The specified position of the interactive interface can be set to one or a combination of the above forms according to actual needs.

[0030] The recommendation method provided in the present disclosure generally corresponds to the recommendation module in the interactive interface. Figure 2 The embodiment shown in FIG. Figure 2A schematic diagram of an interaction interface 1 between a user and an agent according to some embodiments of the present disclosure is shown. Generally, the interaction interface 1 includes an input dialog box 11 for the user to interact with the agent. As an area for receiving user input, at least one of the display position and the area size of this area in the interaction interface 1 can be changed according to the user's usage habits or the agent's preset. Figure 2 A form of the input dialog box 11 is shown, which may include standby text for the agent to greet the user, and may also include Q&A text automatically generated by the agent using a machine learning model during the interaction with the user, etc. The interaction interface 1 further includes a recommendation function module 12 for displaying recommended text. It is located below the input dialog box 11 in the interaction interface 1 so that the user can refer to or use the recommended text 13 in the recommendation function module 12 before or while inputting the content for interacting with the agent. The recommendation function module 12 usually includes one or more pieces of recommended text 13, such as the recommended text 130 presented in the form of a question; and one or more target multimedia resources 14, such as the multimedia resource 141 corresponding to the recommended text 131, the multimedia resource 142 corresponding to the recommended text 132, and / or the multimedia resource 143 corresponding to the recommended text 133, etc. Alternatively, the recommended text 13 can also be displayed as standby text in the input dialog box 11 and recommended to the user. It can be understood that the icons of the target multimedia resources 14 are only used to describe different forms of the multimedia resources recommended by the agent, and are not used to limit the visual presentation effect in the interaction interface 1. The interaction position and interaction form of the target multimedia resources 14 can be personalized according to the layout of the interaction interface 1 and / or the user's usage habits.

[0031] In a non-limiting embodiment, obtaining the target multimedia resource associated with the recommended text includes obtaining the associated multimedia resource 141 in the form of an image based on the recommended text 131. The recommended text 131 is the question "Why have small and niche tourist attractions suddenly become extremely popular?" The semantics "small and niche tourist attractions" contained therein are more conducive to user perception when using visual representation. Therefore, when providing the recommended text 131, a scenic or crowd picture of this tourist attraction will be provided to increase the user's likelihood of selecting the recommended text 131.

[0032] In another non-limiting embodiment, obtaining the target multimedia resource associated with the recommended text includes obtaining the associated multimedia resource 142 in the form of audio based on the recommended text 132. The recommended text 142 is the question "What is the timbre of the bili?" The semantics "bili" and "timbre" contained therein are conducive to user perception when using auditory representation. Therefore, when recommending the recommended text 132, a scale display of this instrument or an audio clip of a classic performance piece will be provided.

[0033] In another non-limiting embodiment, obtaining the target multimedia resource associated with the text to be recommended includes obtaining the associated multimedia resource 143 in the form of video based on the text to be recommended 133. The text to be recommended 133 is an instruction of "recommending a popular film or TV series", and the semantics "film or TV series" contained therein are suitable for visual expression, so when providing the recommendation text 133, a video clip of a film or TV series with high recent playback data will be provided; of course, similar to the aforementioned multimedia resources 141 or 142, other forms of multimedia resources 14 such as stills and audio of the film or TV series can also be displayed in the interactive interface 1 in a matching manner.

[0034] Please refer to the following Figure 3 , which shows a flowchart of obtaining target multimedia resources associated with the text to be recommended according to some embodiments of the present disclosure. In step S301, historical questions associated with multimedia resources are determined from historical conversation records; in step S302, historical questions are classified to determine the type keywords corresponding to each type of historical questions; in step S303, based on the type keywords of the historical questions corresponding to the text to be recommended, the type keywords corresponding to the text to be recommended are determined as target keywords; in step S304, historical conversations associated with the target keywords are retrieved from historical conversation records; and in step S305, based on the multimedia keywords contained in the historical conversations, multimedia resources corresponding to the multimedia keywords are obtained as target multimedia resources, wherein the multimedia keywords are subclasses of the target keywords.

[0035] In some embodiments, step S301 is mainly used to analyze the historical dialogue records between the user and the agent, wherein according to the machine learning model, interactive content involving multimedia resources is searched from the historical dialogue records, including historical dialogues directly involving multimedia resources, such as a user asking the agent "what kind of plant is this" for an image of a plant, and texts that evaluate or discuss multimedia resources, such as a user asking the agent "where was this plot filmed" for a certain film or television work, "what dialect does this character speak in", etc., and these interactive contents are determined as historical issues associated with multimedia resources. It is understandable that the "historical issues" in the present disclosure are not limited to interactive content in the form of questions, but are used to describe various forms of topics or themes involving multimedia resources in historical dialogue records.

[0036] In some embodiments, step S302 mainly includes: semantically identifying historical issues; based on the results of semantic identification, classifying historical issues, and determining the type keywords corresponding to each type of historical issue. Specifically, a machine learning model is used to perform semantic identification on the text of the historical issues determined in step S301 to understand and classify these historical issues. As mentioned above, for the historical issue "What plant is this", its semantics is judged to belong to "plant encyclopedia question", and the type keyword corresponding to this type of historical issue is determined to be "plant"; for the historical issue "Where is the filming location of this plot", its semantics is judged to belong to "scenery and scenic spot question", and the type keyword corresponding to this type of historical issue is determined to be "scenery"; for the historical issue "What dialect does this character's accent belong to", its semantics is judged to belong to "dialect and regional question", and the type keyword corresponding to this type of historical issue is determined to be "dialect", etc. It can be understood that the type keywords in the above examples are only examples, and the actual standards for classifying historical issues may be more generalized or more detailed, and different classification standards are set according to needs. As a result, historical issues related to multimedia resources in historical dialogue records have been divided into multiple different types represented by type keywords.

[0037] In some embodiments, step S303 is mainly used to correspond the determined text to be recommended to a specific type of historical question. Specifically, in the recommendation method of the present disclosure, the text to be recommended (i.e. Figure 1Step S101 shown includes: determining historical issues associated with multimedia resources from the historical dialogue records between the user and the agent; and determining the text to be recommended based on at least one of the historical issues and the timeliness issues. In a non-limiting embodiment, the user discussed the historical issue of "what plant is this" with the agent in the historical dialogue for a plant image, and the agent replied that the plant image is an image of Phalaenopsis in the interaction, then the historical issue can be directly used as the text to be recommended, and the text to be recommended can also be determined as content associated with "flower varieties" or "Phalaenopsis growth habits" based on the historical issue. In another non-limiting embodiment, since multiple news events with timeliness have occurred at the current time point when the user interacts with the agent, such as a tourist attraction receiving a record-breaking number of tourists, a rose variety being successfully planted in a desert area, and the ending of a TV series causing heated discussions, etc., then based on the historical issue of "what plant is this" and timeliness issues such as "desert rose", the text to be recommended can be determined as "why deserts are suitable for planting roses" or other related content. Alternatively, determining the text 13 to be recommended by the user in the interactive interface 1 may also include determining the text 13 to be recommended based on timeliness issues. Still taking the above-mentioned multiple news events with timeliness as an example, the text 13 to be recommended may be determined as the latest and most discussed "Guide to Popular Tourist Attractions during the Short Holidays" or other related content based on the occurrence time, online discussion degree and other criteria of these timeliness issues.

[0038] Further, in some embodiments, determining the text to be recommended according to at least one of the historical questions and the timeliness questions includes: determining the text to be recommended according to the frequency of occurrence of the historical questions in the historical dialogue records. In the historical dialogue records between the user and the agent, a historical question may appear multiple times in the same or different forms, such as the user may ask the agent "what plant is this" for an image of Phalaenopsis, or may directly ask the agent "how to grow Phalaenopsis" in text form, or may ask the agent "the variety and habits of Phalaenopsis" based on the timeliness question "when will the new variety of Phalaenopsis be on the market". These interactions may be collectively classified into historical questions related to Phalaenopsis due to the relevance of the content, that is, historical questions related to Phalaenopsis appear more frequently in the historical dialogue records, and the higher the frequency of occurrence, the higher the possibility that the user selects the relevant text to be recommended. For each text to be recommended, the frequency of occurrence of its associated historical questions in the historical dialogue records can be used to evaluate the priority, so as to improve the user's experience of using the recommended text.

[0039] Further, in some embodiments, determining the text to be recommended according to at least one of the historical question and the timeliness question includes: determining the context of the historical question in the historical conversation record; determining the text to be recommended based on the historical question according to the user's behavior in the context. In the historical conversation record between the user and the agent, the historical question generally does not appear alone, and the user and the agent will at least conduct a round of questions and answers on the historical question. Therefore, by looking up the context of the historical question, one or more of the historical conversation content before the start of the historical question (hereinafter referred to as the "previous topic"), the historical conversation content that starts after the end of the historical question (hereinafter referred to as the "next topic"), and the number of interactions of the historical question itself determined thereby can be determined. Then the user's behavior in the context includes but is not limited to the evaluation when entering the historical question from the previous topic and the evaluation when entering the next topic from the historical question. Here, the "evaluation" includes the tone of the user when changing the topic and the semantics of the interactive text, wherein "thank you for your answer", "this question is very interesting", etc. can represent the user's positive evaluation, and "change to another topic", "you are wrong", etc. can represent the user's negative evaluation. Alternatively, "evaluation" can also include actions such as "like", "collection" or "regeneration" in the interactive interface, where "like" and "collection" reflect that the user has a good experience with the historical question and has a relatively positive evaluation, while "regeneration" and other actions reflect that the user may have a poor interactive experience with the historical question and have a relatively negative evaluation. In addition, the number of interactions of the historical question itself also reflects that the content of the historical question meets the user's interactive needs and is a point of interest for the user.

[0040] Additionally, determining the text to be recommended based on the historical issues according to the user's behavior with respect to the context includes: determining the recommendation weight of the historical issue according to the user's behavior with respect to the context; and determining the text to be recommended according to the recommendation weight. As mentioned above, if the user's behavior with respect to the context of the historical issue indicates a positive evaluation of a certain historical issue, the recommendation weight of the historical issue can be appropriately increased, and accordingly, historical issues with high recommendation weights are preferentially selected to determine the text to be recommended. On the contrary, if the user's behavior with respect to the context indicates a negative evaluation of a certain historical issue, the recommendation weight of the historical issue can be appropriately reduced.

[0041] Continue to refer Figure 3 In some embodiments, step S304 is mainly used to retrieve historical conversations of the type corresponding to the target keyword from the historical conversation records. As mentioned above, historical conversations classified into the same type may have different forms, so when the target keyword is used as a search item, multiple groups of historical conversations may be involved, so as to provide a possible resource library for obtaining the target multimedia resource in the subsequent step S305.

[0042] In some embodiments, step S305 mainly obtains the target multimedia resource based on the aforementioned multiple groups of historical conversations retrieved, wherein the multimedia keywords contained are first determined based on these historical conversations, and then the target multimedia resource corresponding to the multimedia keyword is obtained. In particular, in the multimedia library corresponding to the target keyword, the multimedia resource corresponding to the multimedia keyword is retrieved as the target multimedia resource, and each of the corresponding type keywords in these multimedia libraries. Specifically, still taking the aforementioned historical question of "Phalaenopsis" as an example, its target keyword can be determined as "plant", then the retrieved multiple groups of historical conversations about "what plant is this", "how to grow Phalaenopsis", "new varieties of Phalaenopsis on the market time", "Phalaenopsis varieties and habits", etc., contain the multimedia keyword "Phalaenopsis", etc., and "Phalaenopsis" belongs to the subclass of "plant".

[0043] Additionally, the extensive multimedia resources required by the intelligent agent interaction software are prepared and classified according to the target keywords. When determining the type keyword, a multimedia library corresponding to the type keyword can be prepared and determined accordingly, and then more refined multimedia keywords can be retrieved from the multimedia library. That is, when the historical question corresponding to the recommended text involves the type keyword (target keyword) of "plant", the multimedia library of "plant" is first determined from the multimedia library; then, since the multimedia keyword determined based on the historical dialogue is "Phalaenopsis", multimedia resources associated with "Phalaenopsis" are further retrieved from the "plant" multimedia library.

[0044] Alternatively, the wide range of multimedia resources required by the intelligent agent interaction software is annotated so that each multimedia resource has a label of content description nature to provide retrieval. These labels may include type keywords corresponding to historical questions, and may also include multimedia keywords as subcategories contained in historical conversations; the same multimedia resource is annotated with type keyword labels and multimedia keyword labels at the same time, and due to the relevance of its content under different classification standards, different levels of labels may also have more than one. For the above-mentioned step S305, generally, by first dividing the targeted multimedia library and then searching for subcategory keywords, the accuracy of retrieval resources can be improved; but it will also cause the complexity of the operation steps accordingly. According to the aforementioned annotation form, in order to speed up the retrieval efficiency, the target keywords and multimedia keywords can also be used as input conditions for the retrieval at the same time, which optimizes the step-by-step retrieval operation, reduces the amount of calculation of the intelligent agent interaction software, and improves the interaction efficiency between the intelligent agent and the user.

[0045] Additionally or alternatively, in some embodiments, in response to the text to be recommended including content associated with generating multimedia resources based on the text, the target multimedia resources associated with the text to be recommended are obtained. Specifically, the text to be recommended may include instructions that require the agent to use machine learning to generate content, such as text-generated images, text-generated videos, and other content associated with generating multimedia resources based on the text. At this time, according to the form of the multimedia resources corresponding to these content generation instructions, the target multimedia resources associated with the text to be recommended in a specific form can be obtained, thereby enhancing the relevance between the text to be recommended and the target multimedia resources.

[0046] Additionally or alternatively, in some embodiments, in response to the text to be recommended including recommendation information that is assisted by multimedia resources, a target multimedia resource for assisting in understanding the recommendation information is obtained. Specifically, the text to be recommended may include recommendation information that needs to be assisted by multimedia resources. In other words, the semantics is relatively weak when presented only in text form, such as Figure 1 The recommended text 131 "Why are niche tourist attractions suddenly popular" in the interactive interface 1 may include "niche tourist attractions" and "popular", etc. It is not easy to attract users' attention and selection when recommending only from the perspective of text. For this reason, some tourist attractions with significantly increased tourist reception and not popular in the public's cognition can be referred to. At this time, displaying the scenery photos (images 141) of these tourist attractions in the interactive interface 1 will effectively help users understand the meaning of the recommended text 131 and provide a better interactive experience.

[0047] Additionally or alternatively, in some embodiments, obtaining the target multimedia resource associated with the text to be recommended may also include: determining the target keyword associated with the text to be recommended based on at least one of the historical conversation records and the timeliness problem; and using the machine learning model to generate the target multimedia resource based on the target keyword. Specifically, the source of the target multimedia resource is not limited to the existing information in the historical conversation records and the timeliness problem, but the target multimedia resource can also be directly generated from the text to be recommended according to the generative machine learning model, thereby enriching the display content of the recommendation function module where the text to be recommended is located and optimizing the user's interactive experience.

[0048] For further information please refer to Figure 4, which shows a schematic block diagram of a recommendation device for a user and an agent according to some embodiments of the present disclosure. Specifically, the recommendation method can be implemented by the recommendation device 4. The recommendation device 4 may include a processor and a memory (not shown), wherein the processor may refer to various implementations of a digital circuit system, an analog circuit system, or a mixed signal (a combination of analog and digital) circuit system that performs functions in a computing system. The processing circuit may include, for example, circuits such as an integrated circuit (IC), an application-specific integrated circuit (ASIC), a portion or circuit of a separate processor core, an entire processor core, a separate processor, a programmable hardware device such as a field programmable gate array (FPGA), and / or a system including multiple processors. Additionally, the memory of the recommendation device 4 may store information generated by the processor and programs and data for processor operations. The memory may be a volatile memory and / or a non-volatile memory. For example, the memory may include, but is not limited to, a random access memory (RAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), a read-only memory (ROM), and a flash memory. Generally, the processor may be configured to execute instructions stored on the memory to implement the recommendation method for interaction between a user and an agent in the present disclosure.

[0049] Specifically, if Figure 4 As shown, in some embodiments, the recommendation device 4 of the present disclosure may include a determination module 41, an acquisition module 42, and a display module 43. Specifically, the determination module 41 is configured to determine the user's text to be recommended; the acquisition module 42 is configured to acquire the target multimedia resource associated with the text to be recommended based on at least one of the historical dialogue record between the user and the agent and the timeliness problem, and the text to be recommended; and the display module 43 is configured to display the text to be recommended and the target multimedia resource on the interaction interface between the user and the agent.

[0050] The present disclosure also provides a recommendation device, which may include a processor and a processor coupled to a memory, wherein the processor is configured to execute the recommendation method for an interactive interface between a user and an agent according to any of the aforementioned embodiments of the present disclosure based on instructions stored in the memory. The recommendation device may refer to Figure 5 , which shows a block diagram of an electronic device according to some embodiments of the present disclosure.

[0051] Figure 5 A block diagram of an electronic device 5 according to some embodiments of the present disclosure is shown.

[0052] The memory 51 is used to store one or more computer-readable instructions. The memory 51 may include any combination of various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory, including but not limited to random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), read-only memory (ROM), and flash memory. The memory 51 may store, for example, an operating system, an application, a boot loader (BootLoader), a database, and other programs, and may also store various applications and various data.

[0053] The processor 52 is used to run computer-readable instructions to implement the song screening method described in any of the above embodiments or the method described in any of the above embodiments. The specific implementation of each step of the method can refer to the above embodiments, and the repeated parts are not repeated here.

[0054] The processor 52 may be configured to execute Figures 1 to 6 The processor 52 may be embodied as various processing devices, such as a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The central processing unit (CPU) may be an X86 or ARM architecture, etc.

[0055] The processor 52 and the memory 51 may communicate with each other directly or indirectly. For example, the processor 52 and the memory 51 may communicate with each other via a network. The network may include a wireless network, a wired network, and / or any combination of a wireless network and a wired network. The processor 52 and the memory 51 may also communicate with each other via a system bus, which is not limited in the present disclosure.

[0056] It should be noted that Figure 5 The components of the electronic device 5 shown are only exemplary and non-restrictive. The electronic device 5 may also have other components according to actual application requirements. The processor 52 may control other components in the electronic device 5 to perform desired functions.

[0057] The electronic device 5 may be implemented by software, firmware and / or hardware, and may be integrated into a device installed with relevant application programs.

[0058] Figure 6 A block diagram of an electronic device according to some other embodiments of the present disclosure is shown.

[0059] Figure 6The electronic device 6 shown may be a computer system with a dedicated hardware structure, which can execute corresponding functions when a relevant application program is installed.

[0060] Electronic devices include, but are not limited to, mobile terminals such as smart phones, laptops, personal digital assistants (PDA), tablet computers (Tablet Personal Computer, Tablet PC), PMP (portable multimedia player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), wearable devices, etc., as well as fixed terminals such as digital televisions, desktop computers, etc.

[0061] like Figure 6 As shown, the central processing unit (CPU) 61 performs various processes according to the program stored in the read-only memory (ROM) 62 or the program loaded from the storage part 68 to the random access memory (RAM) 63. In the RAM 63, data required when the CPU 61 performs various processes, etc. is stored as needed. The central processing unit is only exemplary, and it can also be other types of processors, such as the various processors described above. The ROM 62, RAM 63 and the storage part 68 can be various forms of computer-readable storage media. It should be noted that although Figure 6 ROM 62, RAM 63 and storage section 68 are shown separately in FIG. 1 , but one or more of them may be combined or located in the same or different memory or storage modules.

[0062] The CPU 61, the ROM 62, and the RAM 63 are connected to one another via a bus 64. To the bus 64, an input / output interface 65 is also connected.

[0063] The following components are connected to the input / output interface 65: an input section 66, such as a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output section 67, including a display, such as a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage section 68, including a hard disk, a magnetic tape, etc.; and a communication section 69, including a network interface card such as a LAN card, a modem, etc. The communication section 69 allows communication processing to be performed via a network such as the Internet. It is easy to understand that although Figure 6 Some of the electronic devices 6 are shown to communicate via a bus 64, but they may also communicate via a network or other means, wherein the network may include a wireless network, a wired network, and / or any combination of a wireless network and a wired network.

[0064] A drive 610 is also connected to the input / output interface 65 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like is mounted on the drive 610 as needed so that a computer program read therefrom is installed into the storage section 68 as needed.

[0065] When the above-described series of processing is implemented by software, a program constituting the software can be installed from a network such as the Internet or a storage medium such as the removable medium 611 .

[0066] According to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which, when the computer program product is run on a computer, enables the computer to implement the method described in any of the aforementioned embodiments. The computer program product includes a computer instruction carried on a computer-readable medium, containing a program code for executing the method shown in the flowchart. In such an embodiment, the computer instruction can be downloaded and installed from the network through the communication part 69, or installed from the storage part 68, or installed from the ROM 62. When the computer program is executed by the CPU 61, the method of the embodiment of the present disclosure is executed.

[0067] It should be noted that, in the context of the present disclosure, a computer-readable medium may be a tangible medium that may contain or store a program for use by an instruction execution system, apparatus, or device or for use in conjunction with an instruction execution system, apparatus, or device.

[0068] The computer readable medium may be a computer readable storage medium, or a computer readable signal medium, or any combination of the two.

[0069] Computer-readable storage media include, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device. Computer instructions are stored on a computer-readable storage medium, and when the instructions are executed by a processor, the method described in any of the foregoing embodiments is implemented.

[0070] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer readable program codes. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than a computer readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0071] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0072] In some embodiments, a computer program is further provided, comprising: instructions, which, when executed by a processor, cause the processor to execute the method described in any of the above embodiments. For example, the instructions may be embodied as computer program codes.

[0073] In embodiments of the present disclosure, computer program codes for performing the operations of the present disclosure may be written in one or more programming languages ​​or combinations thereof, including but not limited to object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In situations involving a remote computer, the remote computer may be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet).

[0074] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0075] The functions described above may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0076] Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A recommendation method, comprising: Determine the user's text to be recommended; Based on at least one of the historical conversation record between the user and the agent and the timeliness problem, and the text to be recommended, acquiring a target multimedia resource associated with the text to be recommended; as well as The text to be recommended and the target multimedia resource are displayed on the interactive interface between the user and the agent.

2. The recommendation method according to claim 1, wherein: Acquiring a target multimedia resource associated with the text to be recommended includes: Determining historical issues associated with multimedia resources from the historical conversation records; Classify the historical issues and determine the type keywords corresponding to each type of historical issues; Based on the type keyword of the historical question corresponding to the text to be recommended, determining the type keyword corresponding to the text to be recommended as a target keyword; Retrieving the historical conversations associated with the target keyword from the historical conversation records; and Based on the multimedia keywords included in the historical conversation, a multimedia resource corresponding to the multimedia keyword is acquired as the target multimedia resource, wherein the multimedia keyword is a subclass of the target keyword.

3. The recommendation method according to claim 2, wherein: Based on the multimedia keywords contained in the historical conversation, obtaining multimedia resources corresponding to the multimedia keywords as the target multimedia resources includes: In the multimedia libraries corresponding to the target keyword, multimedia resources corresponding to the multimedia keyword are retrieved as the target multimedia resources, wherein each multimedia library corresponds to one of the type keywords.

4. The recommendation method according to claim 3, wherein: The historical issues are classified and the type keywords corresponding to each type of historical issue are determined to include: Performing semantic identification on the historical issues; Based on the result of the semantic recognition, the historical issues are classified and the type keywords corresponding to each type of the historical issues are determined.

5. The recommendation method according to claim 1, wherein: Acquiring a target multimedia resource associated with the text to be recommended includes: In response to the text to be recommended including content associated with generating multimedia resources based on the text, the target multimedia resource associated with the text to be recommended is acquired.

6. The recommendation method according to claim 1, wherein: Acquiring a target multimedia resource associated with the text to be recommended includes: In response to the text to be recommended including recommendation information that is assisted in understanding by multimedia resources, the target multimedia resources that assist in understanding the recommendation information are acquired.

7. The recommendation method according to claim 1, wherein: Acquiring a target multimedia resource associated with the text to be recommended includes: Determining a target keyword associated with the text to be recommended based on at least one of the historical conversation record and the timeliness issue; and The target multimedia resource is generated based on the target keyword using a machine learning model.

8. The recommendation method according to claim 1, wherein: Determining the text to be recommended for the user includes: Determining historical issues associated with multimedia resources from the historical conversation records; and The text to be recommended is determined according to at least one of the historical issue and the timeliness issue.

9. The recommendation method according to claim 1, wherein: Determining the text to be recommended for the user includes: According to the timeliness issue, the text to be recommended is determined.

10. The recommendation method according to claim 8, wherein: Determining the text to be recommended according to at least one of the historical issue and the timeliness issue includes: The text to be recommended is determined according to the appearance frequency of the historical question in the historical conversation record.

11. The recommendation method according to claim 8, wherein: Determining the text to be recommended according to at least one of the historical issue and the timeliness issue includes: In the historical conversation record, determining the context of the historical question; The text to be recommended is determined according to the user's behavior with respect to the context and based on the historical question.

12. The recommendation method according to claim 11, wherein: Determining the text to be recommended according to the user's behavior in the context and based on the historical question includes: Determining a recommendation weight of the historical question based on the user's behavior with respect to the context; and The text to be recommended is determined according to the recommendation weight.

13. A recommendation device, comprising: A determination module, configured to determine the user's text to be recommended; An acquisition module, configured to acquire a target multimedia resource associated with the text to be recommended based on at least one of a historical conversation record between the user and the agent and a timeliness issue, and the text to be recommended; as well as The display module is configured to display the text to be recommended and the target multimedia resource on the interactive interface between the user and the agent.

14. An electronic device comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the recommendation method according to any one of claims 1 to 12 based on instructions stored in the memory.

15. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the recommendation method according to any one of claims 1 to 12 is performed. 16 . A computer program product comprising computer executable instructions, which, when executed by a processor, cause the processor to implement the recommendation method according to claim 1 .

Images