Recommendation method, recommendation device and computer readable storage medium

Abstract

The invention relates to a recommendation method, a recommendation device and a computer readable storage medium. The recommendation method comprises the steps that multimedia content to be recommended is determined according to at least one of the demand dimension of a user, a question concerned by the user in the dialogue process with a first agent and the expanded content of the question, the demand dimension is determined based on historical dialogue information of the user and the first agent, and the expanded content of the question is determined based on the historical dialogue information of the user and the first agent. The historical dialogue information is obtained through user authorization; and displaying a recommendation stream including the multimedia content in a dialogue interface of the user and the first 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, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of computer technology, more personalized recommendations can be provided to users in various fields of life. Especially in multimedia application fields such as short videos, providing users with personalized recommendations with their personal characteristics can greatly improve the user experience.

[0003] However, in the related art, it is impossible to accurately perform personalized recommendations for users during the chat between users and intelligent agents, resulting in a poor user experience. Summary of the invention

[0004] In view of this, the embodiments of the present disclosure provide a recommendation method, a recommendation device, and a computer-readable storage medium, which determine the multimedia content recommended to the user based on the user's needs and the dialogue between the user and the intelligent agent, so that the recommended multimedia content better meets the user's needs, improves the accuracy of personalized recommendations, and thus improves the user experience.

[0005] According to some embodiments of the present disclosure, a recommendation method is provided, comprising: determining multimedia content to be recommended based on a user's demand dimension, a question that the user is concerned about during a conversation with a first agent, and at least one of the expanded contents of the question, wherein the demand dimension is determined based on historical conversation information between the user and the first agent, and the historical conversation information is obtained with the authorization of the user; displaying a recommendation flow including the multimedia content in a conversation interface between the user and the first agent;

[0006] According to other embodiments of the present disclosure, a recommendation device is provided, comprising: a determination module, configured to determine multimedia content to be recommended based on at least one of a user's demand dimension, an issue that the user is concerned about during a conversation with a first intelligent agent, and expanded content of the issue, wherein the demand dimension is determined based on historical conversation information between the user and the first intelligent agent, and the historical conversation information is obtained with the authorization of the user; and a display module, configured to display a recommendation stream including the multimedia content in a conversation interface between the user and the first intelligent agent.

[0007] 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 the recommended method of any embodiment described in the present disclosure based on instructions stored in the memory.

[0008] 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 recommended method of any embodiment described in the present disclosure is performed.

[0009] 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

[0010] 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:

[0011] Figure 1 A flowchart showing a recommended method according to some embodiments of the present disclosure is shown;

[0012] Figure 2A is a schematic diagram showing a display recommendation flow according to some embodiments of the present disclosure;

[0013] Figure 2B is a schematic diagram showing a playback interface of multimedia content according to some embodiments of the present disclosure;

[0014] Figure 2C is a schematic diagram showing a list of displayed agents according to some embodiments of the present disclosure;

[0015] Figure 3 is a schematic diagram showing a display recommendation flow according to some other embodiments of the present disclosure;

[0016] Figure 4 is a flowchart illustrating determining multimedia content to be recommended according to some embodiments of the present disclosure;

[0017] Figure 5 is a flowchart showing determination of multimedia content to be recommended according to other embodiments of the present disclosure;

[0018] Figure 6 is a schematic diagram showing a display recommendation flow according to some further embodiments of the present disclosure;

[0019] Fig. 7A is a schematic diagram showing a display recommendation flow according to some other embodiments of the present disclosure;

[0020] Figure 7B is a schematic diagram showing a display recommendation flow according to still other embodiments of the present disclosure;

[0021] Figure 8 is a flowchart showing a recommended method according to other embodiments of the present disclosure;

[0022] Fig. 9is a schematic diagram showing displaying a historical conversation according to some embodiments of the present disclosure;

[0023] Fig.10 is a flowchart illustrating a recommended method according to some further embodiments of the present disclosure;

[0024] Fig.11A is a schematic diagram showing a second agent or application program for displaying a recommendation according to some embodiments of the present disclosure;

[0025] Fig. 11B is a schematic diagram showing a dialogue interface with a second agent according to some embodiments of the present disclosure;

[0026] Fig. 11C is a schematic diagram showing a second agent or application after display update according to some embodiments of the present disclosure;

[0027] Fig.12 is a schematic diagram showing a second agent or application program for displaying a recommendation according to other embodiments of the present disclosure;

[0028] Fig.13 is a schematic diagram showing a second agent or application program for displaying a recommendation according to still other embodiments of the present disclosure;

[0029] Fig.14 is a flow chart illustrating determining a recommended second agent or application according to some embodiments of the present disclosure;

[0030] Fig.15 A block diagram showing an information processing device according to some embodiments of the present disclosure;

[0031] Fig.16 A block diagram showing a recommendation device according to some other embodiments of the present disclosure;

[0032] Fig.17 A block diagram of an electronic device according to some embodiments of the present disclosure is shown.

[0033] 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

[0034] 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.

[0035] 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 and numerical values ​​of the parts and steps set forth in these embodiments should be interpreted as being merely exemplary and not limiting the scope of the present disclosure.

[0036] The term "including" and its variations used in the present disclosure are 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 "based at least in part on". The modifications of "one" and "plurality" mentioned in the present disclosure are illustrative and not restrictive. Those skilled in the art should understand that unless the context clearly indicates otherwise, they should be understood as "one or more".

[0037] It should be noted that the concepts of "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units. Unless otherwise specified, the concepts of "first", "second", etc. are not intended to imply that the objects described in this way must be in a given order in time, space, ranking, or any other manner.

[0038] The names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information. 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 the present 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 need to comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0039] 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.

[0040] The recommendation function in the related art cannot accurately provide personalized recommendations to users, resulting in poor user experience. In order to improve the user experience, the disclosed embodiment provides a new recommendation method, which determines the multimedia content recommended to the user based on the user's needs and the dialogue between the user and the intelligent agent, so that the recommended multimedia content is more in line with the user's focus, improves the accuracy of personalized recommendations, and thus improves the user experience.

[0041] Figure 1 A flowchart of a recommended method according to some embodiments of the present disclosure is shown.

[0042] like Figure 1 As shown, the recommendation method includes: step S1, determining the multimedia content to be recommended based on the user's demand dimension, the issues that the user is concerned about during the dialogue with the first intelligent agent, and at least one of the extended contents of the issues, wherein the demand dimension is determined based on the historical dialogue information between the user and the first intelligent agent, and the historical dialogue information is obtained with the authorization of the user; step S2, displaying the recommendation flow including the multimedia content in the dialogue interface between the user and the first intelligent agent.

[0043] The above-mentioned intelligent agent can be an intelligent agent based on a large language model (LLM) or other natural language processing (NLP) models. For example, the intelligent agent can be constructed using machine learning models such as variational autoencoder (VAE), convolutional neural network (CNN), transformer, etc.

[0044] Through the analysis of data by the intelligent agent, responses to the user's questions can be generated and displayed, thereby completing the processing of the user's questions.

[0045] The user's demand dimensions are, for example, different aspects of the user's demand. For example, in the case where the user instructs the first agent to recommend food to the user during the historical conversation with the first agent, it can be determined that the user's demand dimensions include food. In the case where the user often asks the first agent about weather conditions during the historical conversation with the first agent, it can be determined that the user's demand dimensions include weather.

[0046] The questions that the user is concerned about during the conversation with the first agent are, for example, questions that the user has discussed in depth during the conversation with the first agent. For example, after the user first asks a question, he or she asks a follow-up question regarding the agent's reply, or the user carefully browses the agent's reply to the question raised by the user. In the above case, it can be determined that the question raised by the user is a key question that the user is concerned about, and is thus used to make recommendations for the user.

[0047] The extended content of a question is, for example, the extended content related to the question raised by the user. For example, after a user raises a question about a lipstick, other lipsticks of the same brand as the lipstick, usage examples of the lipstick, the color number of the lipstick, and other related content can be determined as the extended content of the user's question for recommendation to the user, thereby further realizing personalized recommendations for the user.

[0048] Based on the above factors, multimedia content suitable for recommendation to the user can be determined. By displaying the recommendation stream including the above multimedia content in the conversation page between the user and the first agent, personalized recommendation can be achieved during the conversation between the user and the agent.

[0049] The recommendation method provided by the present disclosure determines the multimedia content recommended to the user based on the user's needs and the dialogue between the user and the intelligent agent, so that the recommended multimedia content is more in line with the user's focus, improves the accuracy of personalized recommendations, and thus improves the user experience.

[0050] Combined with the above Figure 1 The basic process of recommending to users in the recommendation method disclosed in the present invention is introduced. FIG. 2A to FIG. 2C This section describes how to display the recommendation stream and the multimedia content in the recommendation stream.

[0051] Figure 2A is a schematic diagram showing a display recommendation flow according to some embodiments of the present disclosure. Figure 2A As shown, in the dialogue interface with the first agent, a recommendation stream 21 can be displayed, and the recommendation stream 21 includes multimedia content 211 to multimedia content 214. In addition, in the dialogue interface, a prompt statement 22 "The following is the content recommended for you" can also be displayed to remind the user and improve the user experience.

[0052] After the user clicks on the multimedia content displayed in a double-column manner in the recommendation stream 21 , the user may enter an interface that displays the multimedia content in an immersive manner. Figure 2B is a schematic diagram showing a playback interface of multimedia content according to some embodiments of the present disclosure. Figure 2B As shown, when the user clicks Figure 2A After the multimedia content 211 is displayed, the multimedia content 211 can be displayed in an immersive manner.

[0053] In some embodiments, in the playback interface of multimedia content, such as Figure 2B As shown, a dialog box 23 for conversing with the first agent can also be displayed, and in response to the user's conversation with the first agent through the input box on the playback interface of the multimedia content, multimedia content further recommended to the user in the recommendation flow is determined and displayed.

[0054] In other words, through the above-mentioned dialog box 23, the user can communicate with the first agent during the process of browsing multimedia content, especially during immersive browsing of multimedia content, and the content of the dialogue can also be used to make further multimedia content recommendations to the user.

[0055] For example, refer to Figure 2B In the example shown, the multimedia content that the user is browsing is, for example, multimedia content 211. After the user performs a specified operation such as swiping up, the next multimedia content, for example, multimedia content 212, will be recommended to the user.

[0056] In the above process, if the user and the agent have a dialogue through the input box, the subsequent recommendations can be updated according to the content of the dialogue. For example, if the multimedia content 211 is about food, and the user has a dialogue with the first agent about the food in the multimedia content 211, it can be considered that the user is more interested in food at this time. After the user performs a specified operation, the multimedia content 213 related to food can be recommended to the user instead of multimedia content 212 related to music, thereby further realizing personalized recommendations for the user.

[0057] In some embodiments, in the playback interface of multimedia content, such as Figure 2B As shown, a first control 24 may also be displayed, and in response to the user triggering an operation on the first control in the playback interface of the multimedia content, the dialogue interface between the user and the first agent is returned.

[0058] Through the first control, a convenient return Figure 2A In the manner of the dialogue interface shown, for example, after the user clicks the first control, the user can directly return to the dialogue interface, thereby further improving the user experience. After returning to the dialogue interface, the historical dialogue between the user and the first agent and the recommendation stream including multimedia content recommended for the user can also be displayed.

[0059] In some embodiments, in the playback interface of multimedia content, such as Figure 2B As shown, a second control 25 may also be displayed to display a list of agents that have interacted with the user in response to the user triggering the second control in the playback interface of the multimedia content.

[0060] Figure 2C is a schematic diagram showing a list of displayed agents according to some embodiments of the present disclosure. Figure 2C As shown, after the user performs a trigger operation on the second control 25, such as clicking the second control, a list 26 of agents that have interacted with the user can be displayed, and the list includes, for example, agents 261 to 264. In addition, the most recent conversation content or an overview of the conversation content between the user and the agent can also be displayed in the list interface.

[0061] Through the second control, a list of historical agents with which the user has had conversations can be provided to the user, so that the user can browse the history of conversations with agents or have further conversations with a certain agent regarding current multimedia content, thereby improving the user experience.

[0062] Combined with the above FIG. 2A to FIG. 2C The specific method and operation flow of displaying the recommendation stream and multimedia content in the present disclosure are introduced. Next, how to determine the multimedia content to be recommended will be further introduced.

[0063] In some embodiments, determining the multimedia content to be recommended based on the user's demand dimension, the issues that the user is concerned about during the conversation with the first intelligent agent, and at least one of the extended contents of the issues may include: determining the multimedia content based on the user's demand dimension, the issues that the user is concerned about, at least one of the extended contents of the user's issues, and the interaction scenario between the user and the first intelligent agent.

[0064] In other words, in the process of making recommendations for users, the interaction scenarios between users and agents can also be considered. Different recommendation strategies can be adopted in different interaction scenarios, thereby further improving the accuracy of personalized recommendations for users.

[0065] Different processes of determining multimedia content to be recommended in the present disclosure will be described below in conjunction with different interaction scenarios.

[0066] In the first case, the interaction scenario is that the user returns to the dialogue interface. The above-mentioned determining the multimedia content according to at least one of the user's demand dimension, the user's concerned questions, and the extended content of the user's questions, and the interaction scenario between the user and the first agent includes: in response to the interaction scenario that the user returns to the dialogue interface, determining the multimedia content according to the user's demand dimension. In other words, in the scenario where the user returns to the dialogue interface, the multimedia content recommended to the user can be determined mainly based on the user's demand dimension.

[0067] Regarding the user returning to the dialogue interface, for example, when the time interval between the user currently entering the dialogue interface and the user's last entry into the dialogue interface exceeds a preset threshold, it can be determined that this is an interaction scenario in which the user returns to the dialogue interface.

[0068] Figure 3 is a schematic diagram showing a display recommendation flow according to some other embodiments of the present disclosure. Figure 3 As shown, in an interactive scenario where the user returns to the dialogue interface, in response to the user returning to the dialogue interface, a recommendation stream 32 can be displayed in a double-column manner after a designated symbol 31, and the recommendation stream 32 can include multimedia content 321 to 324 recommended for the user.

[0069] Figure 3 The separator in is only an example of the designated symbol 31, and the designated symbol 31 may also be other symbols, such as a dash, a wavy line, etc.

[0070] The following will be combined Figure 4 The specific steps of determining the multimedia content to be recommended according to the user's demand dimension are introduced. Figure 4 is a flowchart illustrating determining multimedia content to be recommended according to some embodiments of the present disclosure.

[0071] like Figure 4 As shown, determining the multimedia content according to the demand dimension of the user may include: step S11, clustering the historical conversation information between the user and the first agent to determine multiple demand dimensions; step S12, determining a target demand dimension from the multiple demand dimensions as the demand dimension of the user according to the type of the user's behavior with respect to the historically browsed multimedia content, the user status data authorized by the user, and at least one of the external data; step S13, determining the multimedia content according to the demand dimension of the user.

[0072] In step S11, cluster analysis can be performed on the historical conversation information between the user and the first agent. For example, the historical conversation information can be clustered into multiple topics, such as "food", "music", "movies", "travel", etc., to determine the user's demand dimensions. Topics including a large number of conversations, a number of conversations exceeding a threshold, or a large number of conversation words can be determined as the user's demand dimensions.

[0073] After clustering the historical conversation information, the topics "food" and "music" include 10 conversations respectively, while "movies" include 5 conversations and "travel" includes 3 conversations. In the above case, "food" and "music" can be determined as the user's demand dimensions.

[0074] The above method is merely exemplary and not restrictive, and other methods may also be used to perform cluster analysis on the historical conversation information between the user and the first agent.

[0075] In step S12, after determining the user's multiple demand dimensions, these demand dimensions may be used as candidate demand dimensions, and based on other reference data, a target demand dimension that the user is more likely to be interested in is determined from the candidate demand dimensions, thereby making recommendations to the user.

[0076] Specifically, the types of the above-mentioned user's behavior with respect to the multimedia content browsed historically include, for example, active behavior and passive behavior. Active behavior refers to the behavior initiated by the user, which may include the user's search behavior, comment behavior, like behavior, collection behavior, etc. Passive behavior refers to the behavior initiated by a subject other than the user, and the user is the receiving party, which may include the user's consumption behavior after the multimedia platform recommends the multimedia content to the user, etc.

[0077] By using the types of past user behaviors on historical multimedia content corresponding to multiple demand dimensions, the demand dimensions with higher user demands can be determined among the demand dimensions. For example, the demand dimensions corresponding to multimedia content on which the user has actively performed behaviors can be determined as target demand dimensions.

[0078] For example, in the past, the user may have specifically searched for videos related to "food", while browsing videos related to "music" in response to recommendations from multimedia platforms. After determining that the user's demand dimensions include "food" and "music" based on the conversation between the user and the agent, it can be further determined based on the user's behavior that the user has a higher demand for "food" videos, and therefore "food" is used as the target demand dimension for recommendations for the user.

[0079] By analyzing the types of user behaviors, we can determine the demand dimensions that users have higher demand for, use them as target demand dimensions, and make recommendations to users based on them, thereby improving the accuracy of personalized recommendations.

[0080] The user status data authorized by the user may include, for example, the user's geographical location, the user's schedule, etc. After the user authorizes, the acquisition of the above data may be realized according to the hardware or other software of the user's mobile phone.

[0081] In some embodiments, in response to the user status data including the user's geographic location, a demand dimension corresponding to the type of the geographic location is determined from the multiple demand dimensions as the target demand dimension.

[0082] After obtaining the user's geographic location, the type of the user's current location can be determined, such as commercial areas, agricultural areas, tourist attractions, transportation hubs, etc., and the target demand dimensions for recommendations for the user are determined based on this.

[0083] For example, the user's multiple demand dimensions include "food", "music", and "photography". If the user is located in a commercial district in the city center, "food" can be determined as the target demand dimension. If the user is located in a natural scenic spot in the suburbs, "photography" can be determined as the target demand dimension.

[0084] In some embodiments, in response to the user status data including the user's schedule, a demand dimension related to an item in the schedule is determined from the multiple demand dimensions as the target demand dimension.

[0085] For example, the user's multiple demand dimensions include "food", "sports", and "photography". If the user's subsequent schedule includes a business dinner, "food" can be determined as the target demand dimension. If the user's subsequent schedule includes golf, "sports" can be determined as the target demand dimension.

[0086] In the above process, the user's own status data is used to determine the target demand dimension that is more closely related to the user's current status from the user's multiple demand dimensions, thereby improving the accuracy of personalized recommendations for the user.

[0087] The above-mentioned external data refers to data other than the user himself, such as time-sensitive information such as weather conditions, traffic conditions, market conditions, and hot topics such as hot news and technological development.

[0088] In some embodiments, in response to the external data including timeliness information, a demand dimension related to changes in the timeliness information is determined from the multiple demand dimensions as the target demand dimension, and the timeliness information includes at least one of weather conditions, traffic conditions, and market conditions.

[0089] For example, multiple demand dimensions of a user include "food", "business", and "weather". According to the timeliness information, when the external weather conditions change, such as temperature rises, temperature drops, or rains, "weather" can be determined as the user's demand dimension. When market conditions change, such as the price of a certain commodity rises, "business" can be determined as the user's demand dimension.

[0090] In some embodiments, in response to the external data including a hot topic, a demand dimension corresponding to the subject of the hot topic is determined from the multiple demand dimensions as the target demand dimension.

[0091] For example, multiple demand dimensions of users include "food", "military", and "music". Hot topics in external data include, for example, discussion topics about a singer's new album. By extracting the theme of the topic, "music" can be determined as the target demand dimension. For another example, hot topics include discussion topics about special snacks in a certain place. In this case, "food" can be determined as the target demand dimension.

[0092] In the above process, external data is used to determine the demand dimensions related to external changes or external hot spots from multiple demand dimensions of the user, and recommendations are made to the user based on this, thereby improving the accuracy of personalized recommendations.

[0093] In step S13, multimedia content recommended for the user may be determined based on the determined target demand dimension.

[0094] Furthermore, the determined multimedia content may be related to the data used to determine the target demand dimension. For example, when the target demand dimension is determined to be "food" by the user's geographical location being in a certain commercial district, the content related to the food in and around the commercial district may be determined as the multimedia content to be recommended. When the target demand dimension is determined to be "sports" by the user's schedule including golf, the content related to golf may be determined as the multimedia content to be recommended.

[0095] Determining the multimedia content recommended to the user by the target demand dimension can make the determined multimedia content more in line with the user's needs, thereby improving the accuracy of personalized recommendations.

[0096] Combined with the above Figure 3 to Figure 4 The recommendation process in which the interaction scenario is for the user to return to the dialogue interface is introduced. The recommendation method proposed in this disclosure can recommend more accurate personalized content to users. Figures 5 to 8 The recommended process for the interactive scenario is introduced as the user's dialogue with the intelligent agent.

[0097] Figure 5 FIG. 1 is a flowchart showing how to determine multimedia content to be recommended according to other embodiments of the present disclosure. Figure 5As shown, the multimedia content is determined based on the user's demand dimension, the issues that the user is concerned about during the conversation with the first agent, at least one of the expanded contents of the issues, and the interaction scenario between the user and the first agent, including: step S11', in response to the interaction scenario being a conversation between the user and the first agent, the issues that the user is concerned about during the conversation with the first agent are determined based on at least one of the number of conversations between the user and the first agent regarding the user's issues and the length of time the user browses the replies of the user to the first agent regarding the user's issues; step S12', determining the multimedia content based on the issues that the user is concerned about and at least one of the expanded contents of the issues.

[0098] exist Figure 5 In the illustrated embodiment, relevant recommendations may be made to the user, for example, in response to the case where the number of conversations between the user and the first agent regarding a certain issue exceeds a threshold, or relevant recommendations may be made to the user, for example, in response to the case where the time spent by the user browsing the first agent's reply to a certain issue exceeds a threshold.

[0099] By setting the above conditions, excessively frequent recommendations during the conversation between the user and the first agent are avoided, which ensures that recommendations meet user needs and improves user experience.

[0100] In step S11', the issues that the user is concerned about during the conversation with the first agent, such as the issues that the user needs in-depth answers, can be determined based on the number of conversations between the user and the first agent and the time the user takes to browse replies.

[0101] For example, if the number of conversations between the user and the first agent on a certain question exceeds a threshold, the question can be determined as a question that the user is concerned about. For example, the user may ask the first agent "What is A?", and the first agent replies "A is a material including B and C". The user may then ask the first agent "What is B?", "What effect does the change of C have on A?", etc. In response to the user further asking questions three times, the user's question "What is A?" can be determined as a question that the user is concerned about, and personalized recommendations can be made to the user based on this.

[0102] For another example, when the time a user spends browsing an agent's response to a question exceeds a threshold, such as more than 30 seconds, the question can be determined as a question that the user is concerned about, and personalized recommendations can be made to the user based on this.

[0103] In step S12', multimedia content recommended to the user may be determined based on the questions that the user is concerned about and the extended contents of the questions.

[0104] In some embodiments, step S12', determining the multimedia content according to at least one of the issues that the user is concerned about and the extended content of the issues includes: determining the extended content of the issues according to the issues that the user is concerned about, the extended content including at least one of background information, principle mechanism, application scenario, impact and significance, operation method, and comparative analysis; determining the multimedia content according to at least one of the issues that the user is concerned about and the extended content of the issues.

[0105] In other words, we can first determine the expanded content of the issue based on the user's concerns, including background information, principle mechanism, application scenario, impact and significance, operation method, comparative analysis and other information, and determine the multimedia content recommended to the user based on at least one of the above contents.

[0106] For example, when the user's concern is "Is a certain lipstick good?", the extended content may include "the price of this lipstick", "the use of this lipstick", "other lipsticks of this brand and other lipsticks of this color", "comparison of this lipstick with other lipsticks", etc. Based on the user's above questions and extended content, the user may be recommended a trial video of the lipstick, a comparison video of the lipstick with other lipsticks, and the product interface of the lipstick, etc.

[0107] Through the above process, it can be ensured that the content of personalized recommendation for users is complete, and it is avoided that the multimedia content recommended for users is too single.

[0108] Next, we will combine Figure 6 The above process of making recommendations to users in the interactive scenario where the user is talking to an intelligent agent is specifically introduced.

[0109] Figure 6 is a schematic diagram showing a display recommendation flow according to some further embodiments of the present disclosure. Figure 6 As shown, in an interactive scenario in which a user is conversing with a first agent, the recommendation flow may be displayed after the first agent's reply.

[0110] Figure 6 In response to the number of conversations between the user and the agent exceeding a certain number or the time for the user to browse the reply of the agent exceeding a threshold, a prompt statement 51 and a recommendation stream 52 can be displayed after the reply of the first agent, wherein the prompt statement 51 is, for example, "It is detected that you may be interested in the following content, recommended for you", and the multimedia content 521 to multimedia content 523 in the recommendation stream 52 are generated based on the questions that the user is concerned about in the conversation between the user and the first agent and the expanded content of the questions, and are questions related to material A.

[0111] In the above recommendation process, by determining the issues that have been discussed in depth between the user and the intelligent agent and the extended content of the issue, and recommending multimedia content to the user based on this, the accuracy of personalized recommendations can be improved, thereby improving the user experience.

[0112] Combined with the above Figure 5 and Figure 6 The interactive scenario is a dialogue between a user and a first agent. In the above process, it can be automatically determined whether a recommendation is needed based on the dialogue between the user and the first agent. FIG. 7A to FIG. 7B The interaction scenario is introduced as another situation in which the user dialogues with the first agent. In this case, recommendations can be made to the user in response to the user's instructions or the user performing a specified trigger operation.

[0113] In some embodiments, determining the multimedia content based on the user's demand dimension, the issues that the user is concerned about during the conversation with the first intelligent agent, at least one of the extended contents of the issues, and the interaction scenario between the user and the first intelligent agent includes: in response to the interaction scenario being a conversation between the user and the first intelligent agent, and in response to the user's instructions or the user performing a specified trigger operation, determining the multimedia content based on the issues that the user is concerned about during the conversation with the first intelligent agent, at least one of the extended contents of the issues, the extended content including at least one of background information, principle mechanism, application scenario, impact and significance, operation method, and comparative analysis.

[0114] For example, when a user sends an instruction such as "recommend me some shooting teaching videos", the question related to the user's instruction can be determined as the question that the user is concerned about, and based on the question and the extended content of the question, the multimedia content recommended to the user is determined. The extended content of the question has been described above and will not be repeated here.

[0115] For another example, when the user performs a designated trigger operation such as swiping up, the user's question in the most recent conversation between the user and the first agent can be determined as the user's concern, and based on the question and the extended content of the question, multimedia content recommended to the user can be determined.

[0116] Fig. 7A is a schematic diagram showing a display recommendation flow according to some other embodiments of the present disclosure. Fig. 7A As shown, in an interactive scenario in which a user is conversing with a first agent, in response to the user performing a designated trigger operation or in response to the user's instructions, the recommendation flow can be displayed in a double-column manner at a fixed position in the conversation interface.

[0117] Fig. 7AIn the example, after the user performs an upward swipe operation or in response to the user's instructions, a recommendation stream 71 can be displayed in a fixed position at the bottom of the dialogue interface in a double-column manner. The recommendation stream 71 may include multimedia content 711 to 712 generated based on questions in the user's most recent dialogue with the agent.

[0118] Figure 7B is a schematic diagram showing a display recommendation flow according to yet other embodiments of the present disclosure. Figure 7B As shown, in an interactive scenario in which a user is conversing with a first agent, in response to the user performing a specified trigger operation, the recommendation flow can be displayed in an immersive manner.

[0119] Figure 7B In the example, after the user performs an upward sliding operation, the recommendation stream 71' may be displayed in an immersive manner. For example, the multimedia content 711' in the recommendation stream 71' may be displayed in an immersive manner.

[0120] In the above recommendation process, in response to the user's instructions or the user's behavior, by determining the issues that have been discussed in depth between the user and the intelligent agent and the extended content of the issue, and based on this, recommending multimedia content to the user, the accuracy of personalized recommendations can be improved, thereby improving the user experience.

[0121] In addition to the recommendation process described in the above embodiments, the method proposed in the present disclosure may also include displaying historical conversations related to the recommended multimedia content and generating subsequent replies based on the historical conversations.

[0122] Figure 8 is a flow chart showing a recommendation method according to other embodiments of the present disclosure. Figure 8 As shown, the recommended method is Figure 1 On the basis of step S1 and step S2, the method may also include: step S3, displaying historical conversations related to the multimedia content; step S4, in response to the user input indication information, generating and displaying a reply to the indication information according to the historical conversations and the multimedia content.

[0123] Step S1 and step S2 have been introduced above and will not be repeated here. The following focuses on describing step S3 and step S4.

[0124] In step S3, historical conversations related to the recommended multimedia content can also be displayed. For example, in the case of recommending a food video, past conversations between the user and the first agent about the food can be displayed, so that the user can obtain more information and improve the user experience.

[0125] Fig. 9 is a schematic diagram showing displaying historical conversations according to some embodiments of the present disclosure. Fig. 9As shown, in the conversation interface, a recommendation stream 91 and a historical conversation 92 may be displayed, the recommendation stream 91 includes multimedia content 911 to 914, and the historical conversation 92 may be displayed in the form of a floating window.

[0126] The form of the above floating window is only exemplary and non-restrictive, and for example, it can also be displayed at the bottom of the multimedia content. In addition, in the case of immersive display of the recommendation stream, historical conversations can also be displayed, and the present disclosure does not limit this.

[0127] In step S4, after the user inputs the indication information, a reply to the user's indication information may be generated based on the displayed historical conversation and multimedia content, thereby improving the accuracy of the reply.

[0128] For example, the multimedia content is a video about a certain scenic spot in a certain place, the historical conversation is a conversation between the user and the agent about a travel plan to travel to the place, and the user's further instruction information is, for example, "to make a travel plan, you need to first go to the scenic spot in the video." The first agent can regenerate a travel plan including the scenic spot in the multimedia content based on the travel plan in the historical conversation and the multimedia content as a reply to the user's instruction information.

[0129] In the above process, the user experience can be improved by displaying the historical conversation to the user and generating a subsequent reply to the user's instruction information based on the historical conversation.

[0130] Combined with the above Figures 1 to 9 The recommendation process of displaying recommendation flows in various scenarios is introduced. The recommendation method proposed in this disclosure can improve the accuracy of personalized recommendations for users, thereby improving user experience. Fig.10 To introduce how to perform intelligent agent recommendation in the recommendation method disclosed in the present invention.

[0131] Fig.10 is a flow chart showing a recommendation method according to some further embodiments of the present disclosure. Fig.10 As shown, the recommended method can be Figure 1 Based on step S1 and step S2 in the above, the method further includes: step S3', according to the conversation between the user and the first agent, displaying at least one of the recommended second agents and applications, wherein the second agents include service agents and skill agents. Steps S1 and S2 have been introduced above and will not be repeated here. The following focuses on step S3'.

[0132] In step S3', a second agent or application may be further recommended to the user, such as an enterprise service agent, an artificial intelligence creation agent, a game application, etc., so as to further meet the needs of the user and improve the user experience.

[0133] Specifically, the needs of the user can be determined based on the conversation between the user and the first agent, so as to recommend an agent or application to the user.

[0134] In some embodiments, in response to the user's needs including needs related to life services, a recommended service-type agent is displayed. For example, when a conversation between a user and a first agent mentions a restaurant, it can be determined that the user has needs related to the restaurant, i.e., needs related to life services, and a service-type agent related to the restaurant is recommended to the user, such as a business agent of the restaurant.

[0135] In other embodiments, in response to the user's needs including needs related to creation, work, or learning, a recommended skill-based agent is displayed. For example, in the case where the conversation between the user and the first agent mentions image generation, it can be determined that the user has a need for image generation, i.e., a need related to creation, and a skill-based agent related to image generation is recommended to the user.

[0136] In some other embodiments, in response to the user's needs including needs related to entertainment, recommended applications are displayed. For example, in the case where the conversation between the user and the first agent mentions games, it can be determined that the user has a need to play games, that is, a need related to entertainment, and applications such as mini-games are recommended to the user.

[0137] The above steps can improve the accuracy of recommending the second agent or application to the user by classifying the user's needs, thereby improving the user experience. The following describes how to combine the user's behavior towards the recommended multimedia content and the user's conversation with the first agent to recommend other agents or applications.

[0138] Based on the conversation between the user and the first agent, at least one of the recommended second agents and applications is displayed. The second agents include service agents and skill agents and may also include: based on the user's behavior towards at least one of the multimedia contents and the conversation between the user and the first agent, at least one of the recommended second agents and applications is determined and displayed.

[0139] The user's behavior on the multimedia content includes, for example, the user clicking on the multimedia content displayed in a double-column manner in the recommendation flow, or the user browsing the multimedia content.

[0140] For example, after a user has a conversation with a first agent about food, the recommendation stream displayed to the user may include multimedia content related to a restaurant, such as the restaurant's signature dish. When the user clicks on the multimedia content, it can be determined that the user has a demand related to life services, and the enterprise agent of the restaurant is recommended to the user as the second agent, thereby achieving personalized recommendations for the user and improving the user experience.

[0141] The specific method of displaying the recommended second agent or application in step S3' will be described below.

[0142] In some embodiments, displaying at least one of the recommended second agents and applications based on the conversation between the user and the first agent may include: displaying at least one of the recommended second agents and applications at a fixed position on the conversation interface.

[0143] Fig.11A is a schematic diagram showing a second agent or application program that displays a recommendation according to some embodiments of the present disclosure. Fig.11A As shown, in addition to the displayed recommendation stream 111 including multimedia contents 1111 to 1114, a recommended second agent 112 and application 113 may also be displayed at a fixed position above the dialogue interface.

[0144] The above display method is only exemplary and not restrictive. The agent or application can also be displayed at the bottom or side of the dialogue interface.

[0145] Fig. 11B is a schematic diagram showing a dialogue interface with a second agent according to some embodiments of the present disclosure. Fig.11A After the second agent 112 is displayed in Fig. 11B As shown in the dialogue interface with the intelligent agent. Fig. 11B As shown, the dialogue interface may include a dialog box 114 and agent introduction information 115. By displaying the introduction information, the user can understand the function of the agent, thereby meeting the user's needs and improving the user experience.

[0146] As the user continues to interact with the first agent, the recommended second agent or application can be updated if specified conditions are met.

[0147] The above-mentioned specified conditions refer to, for example, that the number of times a user clicks on multimedia content of a certain topic exceeds a threshold, the length of time a user browses multimedia content of a certain topic exceeds a threshold, the number of conversations between the user and the first agent on a certain topic exceeds a threshold, or the time since the last recommendation exceeds a threshold. In the above cases, it can be determined that after the last recommendation, the user's new needs are on the topic, and a new second agent or application is recommended to the user based on the topic.

[0148] Fig. 11C is a schematic diagram showing a second agent or application after display update according to some embodiments of the present disclosure. Fig. 11C As shown, the updated recommendation includes the second agent 116 and the application 113. In other words, after the user talked about painting with the first agent many times, the second agent 116 for artificial intelligence painting was recommended to the user, replacing the second agent 112 for restaurant ordering.

[0149] In the above-mentioned updating process, the number of new second agents or applications in each recommendation can be further limited to be less than a threshold, thereby avoiding user interface jumps caused by too many agents recommended each time, and further improving user experience.

[0150] In some embodiments, displaying a recommended second agent or at least one application based on the conversation between the user and the first agent includes: displaying a recommended second agent or at least one application and a message sent by the second agent or at least one application, wherein the message is generated based on the multimedia content and the conversation between the user and the first agent.

[0151] In other words, in addition to displaying the recommended second agent or application, the information actively sent by it can also be displayed. This information can be generated based on the dialogue between the user and the first agent, thereby further meeting the user's needs and improving the user experience.

[0152] Fig.12 is a schematic diagram showing a second agent or application program for displaying a recommendation according to some other embodiments of the present disclosure. Fig.12 As shown, when the conversation between the user and the first agent involves cartoon characters many times, a second agent 121 for artificial intelligence painting recommended to the user can be displayed, and a greeting message 122 sent by the second agent 121 can also be displayed, such as "I can create cartoon characters."

[0153] In the above process, messages are sent through the second agent or application, so that the user can further understand whether the function of the second agent or application meets the user's needs, thereby improving the user experience.

[0154] In the above embodiment, the second agent or application recommended to the user displayed in the dialogue interface is merely exemplary and not restrictive, and the second agent or application recommended to the user may also be displayed in other interfaces.

[0155] Fig.13 is a schematic diagram showing a second agent or application program for displaying a recommendation according to some further embodiments of the present disclosure. Fig.13 As shown, the first agent 131, the recommendation stream 132, the second agent 133 and the second agent 134 can be displayed in an interface different from the dialogue interface, such as a home page interface, and the recommendation stream 132 includes multimedia content 1321 to multimedia content 1324.

[0156] The user can enter the dialogue interface with the corresponding agent by clicking the first agent 131, the second agent 133 or the second agent 134 displayed in the interface.

[0157] Next, we will combine Fig.14 The complete process of determining the recommended second agent or application in the embodiment of the present disclosure is introduced. Fig.14 is a flowchart illustrating determining a recommended second agent or application according to some embodiments of the present disclosure.

[0158] like Fig.14 As shown, determining the second agent or application may include steps S141 to S147.

[0159] In step S141, user data is obtained, which may include user behavior data regarding recommended multimedia content and conversations between the user and the agent.

[0160] In step S142, it can be determined whether the user's data meets the recommendation conditions, which may include, for example, at least one or more of the following conditions: whether the number of times the user clicks on multimedia content exceeds a threshold, whether the number of conversations between the user and the first agent on a certain topic exceeds a threshold, etc.

[0161] If the user data meets the above conditions, step S143 is executed. If not, the process returns to step S141 and the user data is acquired again after a certain period of time.

[0162] In step S143, the user's needs are determined by analyzing the above data of the user, including the creative needs or life service needs as described above.

[0163] In step S144, keyword matching is performed among the agents and applications corresponding to this type of demand to determine the agent or application that matches the keyword in the user's data.

[0164] In step S145, it is determined whether there is an agent or application related to the user's concern in the above keyword matching results. If there is, step S146 is executed. If not, the recommendation is terminated.

[0165] In step S146, the quality of the agent or application related to the user's concern determined in step S145 is analyzed, and whether the quality of the agent or application is qualified is determined based on the number of users of the agent or application, the length of time used, the evaluation score, etc. If qualified, step S147 is executed. If not qualified, the recommendation is terminated.

[0166] In step S147, the qualified intelligent agents are used to make recommendations to the user.

[0167] The above-mentioned recommendation method disclosed in the present invention makes recommendations to users based on the needs of users and the dialogue between users and intelligent agents, so that the recommendations are more in line with the needs of users, and the accuracy of personalized recommendations is improved, thereby improving the user experience.

[0168] The above is a recommended method for the embodiment of the present disclosure. Fig.15 and Fig.16 An information processing device according to an embodiment of the present disclosure is described, which is used to execute any embodiment of the above-mentioned information processing method. Fig.15 A block diagram of an information processing device according to some embodiments of the present disclosure is shown.

[0169] like Fig.15 As shown, the recommendation device 15 includes: a determination module 151, configured to determine the multimedia content to be recommended based on the user's demand dimension, the issues that the user is concerned about during the dialogue with the first intelligent agent, and at least one of the extended contents of the issues, wherein the demand dimension is determined based on the historical dialogue information between the user and the first intelligent agent, and the historical dialogue information is obtained with the authorization of the user; a display module 152, configured to display a recommendation flow including the multimedia content in the dialogue interface between the user and the first intelligent agent.

[0170] The determination module 151 of the recommendation device 15 can be used to perform, for example Figure 1 The display module 152 of the recommendation device 15 can be used to perform Figure 1 Step S2.

[0171] Fig.16 A block diagram showing a recommendation device according to some other embodiments of the present disclosure.

[0172] The memory 161 is used to store one or more computer-readable instructions. The memory 161 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), flash memory. The memory 161 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.

[0173] The processor 162 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 be referred to the above embodiments, and the repeated parts are not repeated here.

[0174] The processor 162 may be configured to execute Figures 1 to 14 The processor 162 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.

[0175] The processor 162 and the memory 161 may communicate with each other directly or indirectly. For example, the processor 162 and the memory 161 may communicate with each other through 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 162 and the memory 161 may also communicate with each other through a system bus, which is not limited in the present disclosure.

[0176] It should be noted that Fig.16 The components of the recommendation device 16 shown are only exemplary and non-limiting. The recommendation device 16 may also have other components according to actual application requirements. The processor 162 may control other components in the recommendation device 16 to perform desired functions.

[0177] The recommendation device 16 may be implemented in software, firmware and / or hardware, and may be integrated into a device installed with a related application program.

[0178] Fig.17 A block diagram of an electronic device according to some embodiments of the present disclosure is shown.

[0179] Fig.17The electronic device 17 shown may be a computer system with a dedicated hardware structure, which can execute corresponding functions when a relevant application program is installed.

[0180] 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.

[0181] like Fig.17 As shown, the central processing unit (CPU) 171 performs various processes according to the program stored in the read-only memory (ROM) 172 or the program loaded from the storage part 178 to the random access memory (RAM) 173. In the RAM 173, data required when the CPU 171 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 172, the RAM 173 and the storage part 178 can be various forms of computer-readable storage media. It should be noted that although Fig.17 ROM 172, RAM 173 and storage section 178 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.

[0182] The CPU 171, the ROM 172, and the RAM 173 are connected to one another via a bus 174. To the bus 174, an input / output interface 175 is also connected.

[0183] The following components are connected to the input / output interface 175: an input portion 176, such as a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output portion 177, including a display, such as a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage portion 178, including a hard disk, a magnetic tape, etc.; and a communication portion 179, including a network interface card such as a LAN card, a modem, etc. The communication portion 179 allows communication processing to be performed via a network such as the Internet. It is easy to understand that although Fig.17 Some of the electronic devices 17 are shown to communicate via bus 174, 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 wireless networks and wired networks.

[0184] A drive 1710 is also connected to the input / output interface 175 as needed. A removable medium 1711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is mounted on the drive 1710 as needed so that a computer program read therefrom is installed into the storage section 178 as needed.

[0185] When the series of processing described above is realized 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 1711 .

[0186] 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 179, or installed from the storage part 178, or installed from the ROM 172. When the computer program is executed by the CPU 171, the method of the embodiment of the present disclosure is executed.

[0187] 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.

[0188] The computer readable medium may be a computer readable storage medium, or a computer readable signal medium, or any combination of the two.

[0189] 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.

[0190] 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.

[0191] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0192] 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.

[0193] 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).

[0194] 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.

[0195] 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.

[0196] 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 multimedia content to be recommended according to at least one of the user's demand dimension, the question that the user is concerned about during the conversation with the first agent, and the extended content of the question, wherein the demand dimension is determined based on the historical conversation information between the user and the first agent, and the historical conversation information is obtained with the authorization of the user; In the dialogue interface between the user and the first agent, a recommendation stream including the multimedia content is displayed.

2. The recommendation method according to claim 1, wherein: Determining the multimedia content to be recommended according to at least one of the user's demand dimension, the user's concern during the conversation with the first agent, and the extended content of the question includes: The multimedia content is determined based on the user's demand dimensions, the issues that the user is concerned about during the conversation with the first intelligent agent, at least one of the extended contents of the issues, and the interaction scenario between the user and the first intelligent agent.

3. The recommendation method according to claim 2, wherein: The determining of the multimedia content according to the user's demand dimension, the user's concern during the conversation with the first agent, at least one of the extended contents of the question, and the interaction scene between the user and the first agent includes: The dialogue interface is returned to the user in response to the interaction scenario, and the multimedia content is determined according to the user's demand dimension.

4. The recommendation method according to claim 3, wherein: The determining the multimedia content according to the user's demand dimension includes: Clustering historical conversation information between the user and the first agent to determine multiple demand dimensions; Determining a target demand dimension from the multiple demand dimensions as the demand dimension of the user according to the type of the user's behavior with respect to the historically browsed multimedia content, the user state data authorized by the user, and at least one of the external data; The multimedia content is determined according to the user's demand dimension.

5. The recommendation method according to claim 4, wherein: The target demand dimension is determined from the multiple demand dimensions according to the type of the user's behavior on the historically browsed multimedia content, the user state data authorized by the user, and at least one of the external data, and the demand dimension of the user includes at least one of the following: In response to the user status data including the user's geographical location, determining, from the multiple demand dimensions, a demand dimension corresponding to the type of the geographical location as the target demand dimension; In response to the user status data including the user's schedule, determining, from the plurality of demand dimensions, a demand dimension associated with an item in the schedule as the target demand dimension; In response to the external data including timeliness information, a demand dimension related to changes in the timeliness information is determined from the multiple demand dimensions as the target demand dimension, and the timeliness information includes at least one of weather conditions, traffic conditions, and market conditions. In response to the external data including a hot topic, a demand dimension corresponding to a subject of the hot topic is determined from the multiple demand dimensions as the target demand dimension.

6. The recommendation method according to claim 2, wherein: The determining of the multimedia content according to the user's demand dimension, the user's concern during the conversation with the first agent, at least one of the extended contents of the question, and the interaction scene between the user and the first agent includes: In response to the interaction scenario being a conversation between the user and the first agent, determining the issues that the user is concerned about during the conversation with the first agent according to at least one of the number of conversations between the user and the first agent regarding the user's issues and the browsing time of the user's reply to the first agent regarding the user's issues; The multimedia content is determined according to at least one of the question that the user is concerned about and the extended content of the question.

7. The recommendation method according to claim 6, wherein: The determining of the multimedia content according to at least one of the question concerned by the user and the extended content of the question includes: Determine, according to the question concerned by the user, the expanded content of the question, wherein the expanded content includes at least one of background information, principle mechanism, application scenario, impact and significance, operation method, and comparative analysis; The multimedia content is determined according to at least one of the question that the user is concerned about and the extended content of the question.

8. The recommendation method according to claim 2, wherein: The determining of the multimedia content according to the user's demand dimension, the user's concern during the conversation with the first agent, at least one of the extended contents of the question, and the interaction scene between the user and the first agent includes: In response to the interaction scenario being a conversation between the user and the first intelligent agent, and in response to the user's instructions or the user performing a designated trigger operation, the multimedia content is determined based on at least one of the issues that the user is concerned about during the conversation with the first intelligent agent and the extended content of the issues, wherein the extended content includes at least one of background information, principle mechanism, application scenario, impact and significance, operation method, and comparative analysis.

9. The recommendation method according to claim 1, further comprising: Based on the conversation between the user and the first agent, at least one of the recommended second agents and applications is displayed, and the second agents include service agents and skill agents.

10. The recommendation method according to claim 9, wherein: The displaying of at least one of the recommended second agent and application according to the conversation between the user and the first agent comprises: Based on the user's behavior towards at least one of the multimedia contents and the conversation between the user and the first agent, at least one of the recommended second agents and applications is determined and displayed.

11. The recommendation method according to claim 9, wherein: The displaying of at least one of the recommended second agent and application according to the conversation between the user and the first agent comprises: Determining the needs of the user based on the conversation between the user and the first agent; In response to the user's needs including needs related to life services, displaying recommended service-type intelligent agents; In response to the user's needs including needs related to creation, work, or learning, displaying recommended skill-based intelligent agents; In response to the user's needs including needs related to entertainment, recommended applications are displayed.

12. The recommendation method according to claim 9, wherein: The displaying of at least one of the recommended second agent and application according to the conversation between the user and the first agent comprises: At least one of the recommended second agents and applications is displayed at a fixed position of the dialogue interface.

13. The recommendation method according to claim 9, wherein: The displaying of at least one of the recommended second agent and application according to the conversation between the user and the first agent comprises: Displaying at least one of a recommended second agent and an application and a message sent by at least one of the second agent and the application, wherein the message is generated based on the multimedia content and the conversation between the user and the first agent.

14. The recommendation method according to claim 1, further comprising: Displaying historical conversations related to the multimedia content; In response to the user inputting indication information, a reply to the indication information is generated and displayed according to the historical conversation and the multimedia content.

15. The recommendation method according to claim 1, wherein: The displaying of the recommendation stream including the multimedia content in the dialogue interface between the user and the first agent includes at least one of the following: In response to the user returning to the dialog interface or the user performing a specified trigger operation, displaying the recommendation flow; Displaying the recommendation flow after the reply of the first agent, after a designated symbol, or at a fixed position in the dialogue interface; The recommendation flow is displayed in a double-column or immersive manner.

16. The recommendation method according to claim 1, further comprising at least one of the following: In response to the user's dialogue with the first agent through the input box on the multimedia content playback interface, determining and displaying multimedia content further recommended for the user in the recommendation stream; In response to the user triggering an operation on a first control in the playback interface of the multimedia content, returning to a dialogue interface between the user and the first agent; In response to the user's triggering operation on a second control in the playback interface of the multimedia content, a list of agents that have interacted with the user is displayed.

17. A recommendation device, comprising: a determination module configured to determine the multimedia content to be recommended according to at least one of a user's demand dimension, a question that the user is concerned about during a conversation with a first agent, and an extended content of the question, wherein the demand dimension is determined based on historical conversation information between the user and the first agent, and the historical conversation information is obtained with the authorization of the user; A display module is configured to display a recommendation stream including the multimedia content in a dialogue interface between the user and the first agent.

18. A recommendation 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 16 based on instructions stored in the memory.

19. 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 16 is implemented.

20. A computer program product, when the computer program product is run on a computer, enables the computer to implement the recommendation method according to any one of claims 1 to 16.

Images