Intelligent agent-based man-machine interaction method and device, electronic equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an agent-based man-machine interaction method and device, electronic equipment and a medium. According to the specific implementation scheme, at least one corresponding service recommendation component is recalled according to interaction information input by a user in an intelligent agent; and according to the interaction information and the at least one service recommendation component, generating a corresponding reply result in the intelligent agent. According to the method and the device, the corresponding answer text can be generated for the user interaction information through the intelligent agent, and the service recommendation component displayed to the user can be dynamically adjusted through the personalized recall service recommendation component, so that the user experience can be improved, the user stickiness can be increased, and the business conversion capability of the intelligent agent can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to an agent-based human-computer interaction method, device, electronic device and medium. Background Art

[0002] With the increasing maturity of artificial intelligence technology, AI agents are increasingly widely used in all walks of life. However, in the process of improving user experience and promoting the prosperity of the platform ecosystem, how to effectively integrate commercial elements and achieve a win-win situation of user experience and commercial value has become an urgent problem to be solved. The existing intelligent agent human-computer interaction method has a single application scenario and it is difficult to embed a variety of service recommendation components without affecting the user interaction experience. It is unable to identify user needs and recommend relevant services that match user needs in a targeted manner. The existing service recommendation lacks flexibility and intelligence and is difficult to adapt to the dynamic changes and diversity of user needs. Summary of the invention

[0003] The present invention provides a human-computer interaction method, device, electronic device and storage medium based on an intelligent agent.

[0004] According to a first aspect of the present disclosure, there is provided an agent-based human-computer interaction method, comprising:

[0005] Recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent;

[0006] Based on the interaction information and at least one of the service recommendation components, a corresponding reply result is generated in the agent.

[0007] According to a second aspect of the present disclosure, there is provided an agent-based human-computer interaction device, comprising:

[0008] A component recall module, configured to recall at least one corresponding service recommendation component according to the interaction information input by the user in the agent;

[0009] The reply result generation module is configured to generate a corresponding reply result in the agent based on the interaction information and at least one of the service recommendation components.

[0010] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0011] at least one processor; and

[0012] a memory communicatively connected to the at least one processor; wherein,

[0013] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any method in any of the above technical solutions.

[0014] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any one of the methods described in the above technical solutions.

[0015] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program, wherein the computer program implements any one of the methods described in the above technical solutions when executed by a processor.

[0016] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0018] Figure 1 is a schematic diagram of the steps of the agent-based human-computer interaction method in an embodiment of the present disclosure;

[0019] Figure 2 is a schematic diagram of the steps of a second agent-based human-computer interaction method in an embodiment of the present disclosure;

[0020] Figure 3 is a display interface diagram of a service recommendation component in an embodiment of the present disclosure;

[0021] Figure 4 is a principle block diagram of an agent-based human-computer interaction device in an embodiment of the present disclosure;

[0022] Figure 5 is a principle block diagram of a second agent-based human-computer interaction device in an embodiment of the present disclosure;

[0023] Figure 6 It is a block diagram of an electronic device used to implement the agent-based human-computer interaction method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0025] Glossary:

[0026] Agent: An agent is an entity that can perceive the environment and take autonomous actions to achieve specific goals. It can be software, hardware or a system with autonomy, adaptability and interaction capabilities. It can be understood as a quick application based on the underlying AI (Artificial Intelligence) large model capabilities, mainly based on conversational interaction, generally created on the agent platform and provided with operation and distribution capabilities by the platform.

[0027] In view of the single application scenario of the intelligent agent in the prior art, the prior art usually adopts a static service recommendation component display. For example, a fixed service recommendation component is inserted into a video, such as a product link, a service link, etc. The same service recommendation component is displayed when all users access the video, and it is impossible to match the corresponding service recommendation component according to the needs of different users. The service recommendation components associated with the intelligent agent in each application scenario are also differentiated. The existing solutions cannot customize specific service recommendation components according to the personalized needs of users, and users cannot get a better experience. Accordingly, the intelligent agent cannot exert better business conversion capabilities. Therefore, the present disclosure provides a human-computer interaction method based on an intelligent agent, such as Figure 1 As shown, including:

[0028] Step S101, recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent.

[0029] After the intelligent agent related to the user's search text is recalled, the intelligent agent is presented to the user in the form of a dialog box. The user can enter interactive information in the dialog box of the intelligent agent, such as "how to make Peking duck", and one or more related service recommendation components can be further recalled based on the user's interactive information. Among them, the service recommendation component can be a product link, a mini program link, a web page link, a lead collection form, etc. Figure 3 As shown, in the car purchase consultation agent, the corresponding service recommendation component 301, i.e., the car mini program link, can be recalled based on the user's interactive information "What are the alternative vehicles within 100,000 yuan?"

[0030] Step S102, generating a corresponding reply result in the intelligent agent according to the interaction information and at least one service recommendation component.

[0031] Specifically, the service recommendation component is one or more commercialization modules in the intelligent agent, which generates personalized recommendation content for users by analyzing user behavior and applying recommendation algorithms. These recommended contents can include various types such as goods, service appointments, pictures, information, videos, etc., such as product links, service appointments, course purchases, lead conversions, etc. Users can click on the service recommendation component to view the detailed information of related goods and services, or jump to the purchase link of related goods and services, or leave relevant clues to facilitate contact with related service providers. Not only can the intelligent agent generate corresponding answer texts for user questions, but the present disclosure can also dynamically adjust the service recommendation components displayed to users through personalized recall of service recommendation components, which can improve user experience, increase user stickiness, and enhance the commercial conversion capabilities of the intelligent agent.

[0032] Based on the user's interaction information, the intelligent agent specifically recalls the service recommendation components that match the user's interaction information, which makes it easier for the user to choose whether to visit the relevant service page, thereby improving the commercial value of the intelligent agent, and recalls the service recommendation components with the highest matching degree based on the interaction information between the user and the intelligent agent. The type of service recommendation component is closely related to the user's interaction information, realizing personalized recommendation components, avoiding recommending service recommendation components that the user is not interested in or have low relevance, which affects the user experience.

[0033] As an optional implementation, Figure 2 As shown, step S201, before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, further includes:

[0034] Step S201, based on the search text input by the user, recall the agent corresponding to the search text from multiple pre-configured agents.

[0035] Specifically, users can enter search text in the search engine, and the search engine can match agents related to the search text according to the search text entered by the user. For example, when a user searches for "scrambled eggs with tomatoes", agents related to food can be recalled; when a user searches for "what are the commuter vehicles under 100,000 yuan", agents related to buying a car can be recalled. The present disclosure provides a highly integrated agent platform, where developers can create a large number of agents with different capabilities, and distribute agents that meet user demands to users based on the agent platform or search distribution portal. At the same time, it can also meet the promotion demands of merchants and enhance the commercial conversion capabilities of agents.

[0036] As an optional implementation, step S201, based on the search text input by the user, recalling the agent corresponding to the search text from the pre-configured multiple agents includes:

[0037] Step S201a, determining the user's search intention based on the search text input by the user;

[0038] Step S201b, recalling the agent corresponding to the search intent from multiple pre-configured agents based on the search intent.

[0039] Specifically, developers can pre-configure a large number of agents for different application scenarios on the agent platform, such as food, cars, travel, financial management, course training, etc., and define a name for each agent, such as food assistant, car purchase consultation assistant, etc. In this embodiment, the user's search intent is determined by semantic recognition, or the user's search intent is determined based on keyword matching, or the user's search intent is determined by combining semantic matching and keyword matching. For example, the user searches for "scrambled eggs with tomatoes", determines that the search intent is "food", and recalls the agent corresponding to food; the user searches for "how to operate cross-border e-commerce", determines that the search intent is "cross-border e-commerce training", and recalls the corresponding agent cross-border e-commerce training assistant.

[0040] As an optional implementation, step S101, before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, further includes:

[0041] The corresponding first keyword information is configured for each service recommendation component in advance.

[0042] Specifically, in this embodiment, one or more corresponding keywords are set for each service recommendation component, which are used for keyword matching with the user's interaction information in the future, so as to determine the service recommendation component related to the user's interaction information. This step is configured by the developer on the intelligent agent platform during the development stage or update stage.

[0043] As an optional implementation, pre-configuring the corresponding first keyword information for each service recommendation component includes:

[0044] The name of each service recommendation component is pre-segmented to obtain the corresponding first keyword as the first keyword information.

[0045] The first keyword information refers to the keyword configuration information of the service recommendation component, which is usually presented in the form of a keyword list. In this embodiment, one or more first keywords are obtained by segmenting the name of the service recommendation component. Each service recommendation component is configured with a first keyword list A consisting of multiple first keywords, and the first keyword list A is bound to the corresponding component. For example, component ID (xxxx), component keywords: {{keyword 1, keyword 2, keyword 3}}.

[0046] As an optional implementation, pre-configuring the corresponding first keyword information for each service recommendation component includes:

[0047] The name of each service recommendation component is segmented in advance to obtain the corresponding first keyword.

[0048] The user-defined keyword is combined with the first keyword to obtain the first keyword information.

[0049] Specifically, since the name of each service recommendation component contains few keywords, it is easy to fail to match the keywords of the user interaction information, resulting in the situation that no service recommendation component is recalled. Therefore, in this embodiment, based on the word segmentation of the name of the service recommendation component, the developer can add custom keywords to make the information used for keyword matching richer, and improve the recall rate of the component while ensuring the recall accuracy of the service recommendation component.

[0050] As an optional implementation, step S101, based on the interaction information input by the user in the agent, recalling at least one corresponding service recommendation component includes:

[0051] Step S101a, performing word segmentation processing on the interactive information to obtain corresponding second keyword information.

[0052] Specifically, the second keyword information in this embodiment may be a second keyword list B consisting of one or more second keywords obtained by performing word segmentation processing on the user's interactive information during the user's use of the agent.

[0053] Step S101b: Match the second keyword information with the first keyword information.

[0054] Step S101c: in response to the second keyword information matching the first keyword information, recall at least one service recommendation component matching the second keyword information.

[0055] In this embodiment, the second keyword list B composed of the second keywords is matched with the first keyword list A corresponding to each service recommendation component. If the second keyword in the second keyword list B hits any first keyword in a first keyword list A, the service recommendation component corresponding to the first keyword list A is recalled. By matching the user's interaction information with the keyword information of the service recommendation component, the precise matching of the service recommendation component is achieved.

[0056] As an optional implementation, step S101, before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, includes:

[0057] The corresponding first keyword information is configured for each service recommendation component in advance, and the corresponding component description information is configured for each service recommendation component.

[0058] In this embodiment, not only the first keyword information is configured for each service recommendation component, but also component description information is configured, such as function description, product information, pictures, component type, etc. Information associated with the component is configured. A prompt word template is constructed based on the first keyword information and the component description information, and keyword matching and semantic matching are combined to realize the recall of service recommendation components.

[0059] As an optional implementation, step S101, based on the interaction information input by the user in the agent, recalling at least one corresponding service recommendation component includes:

[0060] Step S101a, performing word segmentation processing on the interactive information to obtain corresponding second keyword information.

[0061] In this embodiment, when the user uses the intelligent agent, the user's interactive information is firstly segmented to obtain a second keyword list B consisting of one or more second keywords as the second keyword information.

[0062] Step S101b: Match the second keyword information with the first keyword information.

[0063] In this embodiment, the keyword matching method is preferably used for precise matching, which requires that the query keyword is completely consistent with the target data. This matching method can ensure the accuracy of the query result, but has limitations when dealing with typos, synonyms and grammatical variants.

[0064] Step S101c: in response to the second keyword information matching the first keyword information, recall at least one service recommendation component matching the second keyword information.

[0065] If the keyword in the user interaction information can be matched to the keyword of any service recommendation component, it is considered that the keyword is matched successfully, and the corresponding service recommendation component is recalled, and the component matching process ends.

[0066] Step S101d, in response to the second keyword information not matching the first keyword information, semantically identifying the interaction information, determining the user's interaction intent, semantically matching the interaction intent with the component description information, and recalling at least one service recommendation component that matches the interaction intent.

[0067] If the keywords in the user interaction information cannot be matched to the keywords of the service recommendation component, it is considered that the keyword match is recognized, and the process jumps to the semantic matching process. Semantic matching can handle synonyms, near-synonyms and grammatical variants, search for results that are similar or related to the interaction intent, and improve the accuracy and recall rate of the query. In this embodiment, semantic matching is used as a supplement. When the keywords cannot be recalled, semantic matching is used to recall the corresponding service recommendation component. In this embodiment, the combination of keyword matching and semantic matching for service recommendation component recall can improve the accuracy and comprehensiveness of component recall, enhance user experience, and enhance the adaptability and flexibility of the system.

[0068] As an optional implementation, after step S101, based on the interaction information input by the user in the agent, recalling at least one corresponding service recommendation component, the method further includes:

[0069] At least one service recommendation component is sorted to determine at least one service recommendation component displayed in the intelligent agent.

[0070] Specifically, a large number of service recommendation components are recalled based on the user's interactive information. However, due to the limited display interface of the intelligent body, the matching degrees of the recalled service recommendation components can be sorted, and the one with the highest matching degree or the top few service recommendation components can be determined as the final displayed service recommendation components to improve the matching accuracy.

[0071] As an optional implementation, step S102, generating a corresponding reply result in the agent according to the interaction information and at least one service recommendation component includes:

[0072] Step S102a, performing semantic recognition based on the interaction information to determine the user's interaction intention.

[0073] Interaction intent refers to the user's interactive information, the user's real purpose or need behind it. For example, when the user enters "a means of transportation within 100,000 yuan", the interaction intention is to obtain the information about and purchase a means of transportation within 100,000 yuan.

[0074] Step S102b, generating a corresponding reply text according to the interaction intention.

[0075] After the intelligent agent recognizes the user's interaction intention through natural language processing technology, it can generate relevant information such as the brand name, price, performance, etc. of vehicles under 100,000 yuan as a reply text.

[0076] Step S102c, displaying the reply text and service recommendation components as reply results in the intelligent agent.

[0077] Specifically, in addition to displaying the reply text to the user, the relevant service recommendation components will also be displayed to the user, providing the user with access channels to related services and products, or providing the user with a clue form. If the user hopes that the relevant merchant will contact him or her, the user can leave contact information, purchase intention, etc. through the clue form, so that the merchant providing services or products can contact the user.

[0078] As an optional implementation, step S101, before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, further includes:

[0079] Configure multiple different types of service recommendation components according to different application scenarios, and associate each agent with at least one service recommendation component.

[0080] Specifically, the service recommendation component can be applied to agents in different application scenarios, such as food, cars, weather, test training, services, and a large number of other application scenarios. Developers can configure the service recommendation component corresponding to each agent through a visual interface. The service recommendation component includes but is not limited to product recommendations, service reservations, lead conversion, purchase consultation, course reservations, etc. You can also configure the basic information of each agent, such as the name of the agent, function description, character settings (for example, when a user asks for food recommendations, recommend the most suitable restaurant, dishes, etc. based on the user's preferences, geographic location, budget, etc.), opening remarks, automatic follow-up questions, etc.

[0081] As an optional implementation, the type of service recommendation component includes at least one of the following: a product link component; a link mounting component; a lead conversion component.

[0082] Specifically, the product link component can be configured with product information, product images, product access links, etc.; the link mount component can be configured with relevant service-related information, service consultation links, mini-program links, etc. Developers can also configure the lead conversion component to display guidance words in the dialog box of the intelligent body to guide users to leave leads, such as "Leave a contact number and arrange for the relevant person in charge to contact you", and provide corresponding phone / text input boxes. Configuring different types of service recommendation components can match the demands of different users and improve the user experience.

[0083] The present disclosure also provides a human-computer interaction device 400 based on an intelligent agent, such as Figure 4 As shown, including:

[0084] The component recall module 401 is configured to recall at least one corresponding service recommendation component according to the interaction information input by the user in the agent.

[0085] After the intelligent agent related to the user's search text is recalled, the intelligent agent is presented to the user in the form of a dialog box. The user can enter interactive information in the dialog box of the intelligent agent, such as "how to make Peking duck", and one or more related service recommendation components can be further recalled based on the user's interactive information. Among them, the service recommendation component can be a product link, a mini program link, a web page link, a lead collection form, etc. Figure 3 As shown, in the car buying consultation agent, the corresponding service recommendation component 301, i.e., the car mini program link, can be recalled based on the user's interactive information "What are the alternative vehicles within 100,000 yuan?"

[0086] The reply result generation module 402 is configured to generate a corresponding reply result in the intelligent agent based on the interaction information and at least one service recommendation component.

[0087] Specifically, the service recommendation component is one or more commercialization modules in the intelligent agent, which generates personalized recommendation content for users by analyzing user behavior and applying recommendation algorithms. These recommended contents can include various types such as goods, service appointments, pictures, information, videos, etc., such as product links, service appointments, course purchases, lead conversions, etc. Users can click on the service recommendation component to view the detailed information of related goods and services, or jump to the purchase link of related goods and services, or leave relevant clues to facilitate contact with related service providers. The present disclosure can not only generate corresponding answer texts for user interaction information through the intelligent agent, but also dynamically adjust the service recommendation components displayed to users through personalized recall of service recommendation components, which can improve user experience, increase user stickiness, and enhance the commercial conversion capabilities of the intelligent agent.

[0088] Based on the user's interaction information, the intelligent agent specifically recalls the service recommendation components that match the user's interaction information, which makes it easier for the user to choose whether to visit the relevant service page, thereby improving the commercial value of the intelligent agent, and recalls the service recommendation components with the highest matching degree based on the user's interaction information. The type of service recommendation component is closely related to the user's interaction information, realizing personalized recommendation components, avoiding recommending service recommendation components that the user is not interested in or have low relevance, which affects the user experience.

[0089] As an optional implementation, Figure 5 As shown, the agent-based human-computer interaction device 400 further includes:

[0090] The agent recall module 501 is configured to recall the agent corresponding to the search text from a plurality of pre-configured agents based on the search text input by the user.

[0091] Specifically, users can enter search text in the search engine, and the search engine can match agents related to the search text according to the search text entered by the user. For example, when a user searches for "scrambled eggs with tomatoes", agents related to food can be recalled; when a user searches for "what are the commuter vehicles under 100,000 yuan", agents related to buying a car can be recalled. The present disclosure provides a highly integrated agent platform, where developers can create a large number of agents with different capabilities, and distribute agents that meet user demands to users based on the agent platform or search distribution portal. At the same time, it can also meet the promotion demands of merchants and enhance the commercial conversion capabilities of agents.

[0092] As an optional implementation, the agent recall module 501 includes:

[0093] The determination unit is configured to determine the user's search intention based on the search text input by the user.

[0094] The agent recall unit is configured to recall the agent corresponding to the search intent from a plurality of pre-configured agents according to the search intent.

[0095] Specifically, developers can pre-configure a large number of agents for different application scenarios on the agent platform, such as food, cars, travel, financial management, course training, etc., and define a name for each agent, such as food assistant, car purchase consultation assistant, etc. In this embodiment, the user's search intent is determined by semantic recognition, or the user's search intent is determined based on keyword matching, or the user's search intent is determined by combining semantic matching and keyword matching. For example, the user searches for "scrambled eggs with tomatoes", determines that the search intent is "food", and recalls the agent corresponding to food; the user searches for "how to operate cross-border e-commerce", determines that the search intent is "cross-border e-commerce training", and recalls the corresponding agent cross-border e-commerce training assistant.

[0096] As an optional implementation, the agent-based human-computer interaction device 400 further includes:

[0097] The component information configuration module is configured as a component recall module 401, which pre-configures the corresponding first keyword information for each service recommendation component before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the intelligent agent.

[0098] Specifically, in this embodiment, one or more corresponding keywords are set for each service recommendation component, which are used for keyword matching with the user's interaction information in the future, so as to determine the service recommendation component related to the user's interaction information. This step is configured by the developer on the intelligent agent platform during the development stage or update stage.

[0099] As an optional implementation manner, the component information configuration module pre-configures the corresponding first keyword information for each service recommendation component, including:

[0100] The name of each service recommendation component is pre-segmented to obtain the corresponding first keyword as the first keyword information.

[0101] The first keyword information refers to the keyword configuration information of the service recommendation component, which is usually presented in the form of a keyword list. In this embodiment, one or more first keywords are obtained by segmenting the name of the service recommendation component. Each service recommendation component is configured with a first keyword list A consisting of multiple first keywords, and the first keyword list A is bound to the corresponding component. For example, component ID (xxxx), component keywords: {{keyword 1, keyword 2, keyword 3}}.

[0102] As an optional implementation manner, the component information configuration module pre-configures the corresponding first keyword information for each service recommendation component, including:

[0103] The name of each service recommendation component is segmented in advance to obtain the corresponding first keyword.

[0104] The user-defined keyword is combined with the first keyword to obtain the first keyword information.

[0105] Specifically, since the name of each service recommendation component contains few keywords, it is easy to fail to match the keywords of the user interaction information, resulting in the situation that no service recommendation component is recalled. Therefore, in this embodiment, based on the word segmentation of the name of the service recommendation component, the developer can add custom keywords to make the information used for keyword matching richer, and improve the recall rate of the component while ensuring the recall accuracy of the service recommendation component.

[0106] As an optional implementation, the component recall module 401 includes:

[0107] The keyword extraction unit is configured to perform word segmentation processing on the interactive information to obtain corresponding second keyword information.

[0108] Specifically, the second keyword information in this embodiment may be a second keyword list B consisting of one or more second keywords obtained by performing word segmentation processing on the user's interactive information during the user's use of the agent.

[0109] The keyword matching unit is configured to match the second keyword information with the first keyword information, and in response to the second keyword information matching with the first keyword information, recall at least one service recommendation component matching the second keyword information.

[0110] In this embodiment, the second keyword list B composed of the second keywords is matched with the first keyword list A corresponding to each service recommendation component. If the second keyword in the second keyword list B hits any first keyword in the first keyword list A, the service recommendation component corresponding to the first keyword list A is recalled.

[0111] As an optional implementation, the agent-based human-computer interaction device 400 includes:

[0112] The component information configuration module is configured as a component recall module 401. Before recalling at least one corresponding service recommendation component based on the interaction information input by the user in the intelligent agent, it pre-configures the corresponding first keyword information for each service recommendation component and configures the corresponding component description information for each service recommendation component.

[0113] In this embodiment, the component information configuration module not only configures the first keyword information for each service recommendation component, but also configures the component description information, such as function description, product information, pictures, component type, etc. Information associated with the component. A prompt word template is constructed based on the first keyword information and the component description information, and keyword matching and semantic matching are combined to realize the recall of the service recommendation component.

[0114] As an optional implementation, the component recall module 401 includes:

[0115] The keyword extraction unit is configured to perform word segmentation processing on the interactive information to obtain corresponding second keyword information.

[0116] In this embodiment, when the user uses the intelligent agent, the user's interactive information is firstly segmented to obtain a second keyword list B consisting of one or more second keywords as the second keyword information.

[0117] The keyword matching unit is configured to match the second keyword information with the first keyword information, and in response to the second keyword information matching with the first keyword information, recall at least one service recommendation component matching the second keyword information.

[0118] In this embodiment, the keyword matching method is preferred for precise matching, which requires that the query keyword is completely consistent with the target data. If the keyword in the user interaction information can match the keyword of any service recommendation component, it is considered that the keyword match is successful, and the corresponding service recommendation component is recalled, and the component matching process ends. This matching method can ensure the accuracy of the query results, but it has limitations when dealing with typos, synonyms, and grammatical variations.

[0119] The semantic matching unit is configured to perform semantic recognition on the interaction information in response to the second keyword information not matching the first keyword information, determine the user's interaction intention, semantically match the interaction intention with the component description information, and recall at least one service recommendation component that matches the interaction intention.

[0120] If the keywords in the user interaction information cannot be matched to the keywords of the service recommendation component, it is considered that the keyword match is recognized, and the process jumps to the semantic matching process. Semantic matching can handle synonyms, near-synonyms and grammatical variants, search for results that are similar or related to the interaction intent, and improve the accuracy and recall rate of the query. In this embodiment, semantic matching is used as a supplement. When the keywords cannot be recalled, semantic matching is used to recall the corresponding service recommendation component. In this embodiment, the combination of keyword matching and semantic matching for service recommendation component recall can improve the accuracy and comprehensiveness of component recall, enhance user experience, and enhance the adaptability and flexibility of the system.

[0121] As an optional implementation, the agent-based human-computer interaction device 400 further includes:

[0122] The sorting module is configured as a component recall module 401. After recalling at least one corresponding service recommendation component according to the interaction information input by the user in the intelligent agent, the at least one service recommendation component is sorted to determine at least one service recommendation component displayed in the intelligent agent.

[0123] Specifically, a large number of service recommendation components are recalled based on the user's interactive information. However, due to the limited display interface of the intelligent agent, the matching degrees of the recalled service recommendation components can be sorted to determine the one with the highest matching degree or the top few service recommendation components as the final displayed service recommendation components.

[0124] As an optional implementation, the reply result generating module 402 includes:

[0125] The intention determination unit is configured to perform semantic recognition based on the interaction information to determine the user's interaction intention.

[0126] The intention determination unit determines the real purpose or demand of the user contained in the user interaction information. For example, when the user inputs "a vehicle within 100,000 yuan", the interaction intention is to obtain information about and purchase a vehicle within 100,000 yuan.

[0127] The text generation unit is configured to generate corresponding reply text according to the interaction intention.

[0128] After the intelligent agent recognizes the user's interaction intention through natural language processing technology, it can generate relevant information such as the brand name, price, performance, etc. of vehicles under 100,000 yuan as a reply text.

[0129] The display unit is configured to display the reply text and the service recommendation component as the reply result in the intelligent agent.

[0130] Specifically, in addition to displaying the reply text to the user, the display unit will also display related service recommendation components to the user, provide the user with access channels to related services and products, or provide the user with a lead form. If the user hopes that the relevant merchant will contact him or her, the user can leave contact information, purchase intention, etc. through the lead form, so that the merchant providing the service or product can contact the user.

[0131] As an optional implementation, the agent-based human-computer interaction device 400 further includes:

[0132] The component configuration module is configured to configure multiple different types of service recommendation components according to different application scenarios before recalling at least one corresponding service recommendation component based on the interaction information input by the user in the intelligent agent, and associate at least one service recommendation component with each intelligent agent.

[0133] Specifically, the service recommendation component can be applied to agents in different application scenarios, such as food, cars, weather, test training, services, and a large number of other application scenarios. Developers can configure the service recommendation component corresponding to each agent through a visual interface. The service recommendation component includes but is not limited to product recommendations, service reservations, lead conversion, purchase consultation, course reservations, etc. You can also configure the basic information of each agent, such as the name of the agent, function description, character settings (for example, when a user asks for food recommendations, recommend the most suitable restaurant, dishes, etc. based on the user's preferences, geographic location, budget, etc.), opening remarks, automatic follow-up questions, etc.

[0134] As an optional implementation, the type of service recommendation component includes at least one of the following: a product link component; a link mounting component; a lead conversion component.

[0135] Specifically, the product link component can be configured with product information, product images, product access links, etc.; the link mount component can be configured with relevant service-related information, service consultation links, mini-program links, etc. Developers can also configure the lead conversion component to display guidance words in the dialog box of the intelligent body to guide users to leave leads, such as "Leave a contact method and arrange for the relevant person in charge to contact you", and provide corresponding phone / text input boxes.

[0136] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0137] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0138] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0139] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0140] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0141] The computing unit 601 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 601 performs the various methods and processes described above, such as an agent-based human-computer interaction method. For example, in some embodiments, the agent-based human-computer interaction method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the agent-based human-computer interaction method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to execute the agent-based human-computer interaction method in any other appropriate manner (eg, by means of firmware).

[0142] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0143] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0144] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0145] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0146] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0147] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0148] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of this disclosure can be achieved, and this document is not limited here.

[0149] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A human-computer interaction method based on an intelligent agent, comprising: Recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent; Based on the interaction information and at least one of the service recommendation components, a corresponding reply result is generated in the agent.

2. The method according to claim 1, wherein: Before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, the method further includes: According to the search text input by the user, the agent corresponding to the search text is recalled from a plurality of pre-configured agents.

3. The method according to claim 2, wherein: The step of recalling the agent corresponding to the search text from a plurality of pre-configured agents according to the search text input by the user comprises: Determining the user's search intent based on the search text input by the user; The agent corresponding to the search intention is recalled from a plurality of pre-configured agents according to the search intention.

4. The method according to claim 1, wherein: Before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, the method further includes: The corresponding first keyword information is pre-configured for each of the service recommendation components.

5. The method according to claim 4, wherein: The first keyword information corresponding to each of the service recommendation components is configured in advance, including: The name of each of the service recommendation components is pre-segmented to obtain a corresponding first keyword as the first keyword information.

6. The method according to claim 4, wherein: The first keyword information corresponding to each of the service recommendation components is configured in advance, including: Pre-segmenting the name of each of the service recommendation components to obtain a corresponding first keyword; The user-defined keyword is combined with the first keyword to obtain the first keyword information.

7. The method according to any one of claims 4 to 6, wherein: The step of recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent comprises: Performing word segmentation processing on the interactive information to obtain corresponding second keyword information; matching the second keyword information with the first keyword information; In response to the second keyword information matching the first keyword information, at least one of the service recommendation components matching the second keyword information is recalled.

8. The method according to claim 1, wherein: Before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, the method includes: The corresponding first keyword information is configured for each of the service recommendation components in advance, and the corresponding component description information is configured for each of the service recommendation components.

9. The method according to claim 8, wherein: The step of recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent comprises: Performing word segmentation processing on the interactive information to obtain corresponding second keyword information; matching the second keyword information with the first keyword information; In response to the second keyword information matching the first keyword information, recalling at least one of the service recommendation components matching the second keyword information; In response to the second keyword information not matching the first keyword information, semantic recognition is performed on the interaction information to determine the user's interaction intent, and the interaction intent is semantically matched with the component description information to recall at least one of the service recommendation components that matches the interaction intent.

10. The method according to claim 1, wherein: After recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, the method further includes: At least one of the service recommendation components is sorted to determine at least one of the service recommendation components displayed in the agent.

11. The method according to claim 1, wherein: Generating a corresponding reply result in the agent according to the interaction information and at least one of the service recommendation components comprises: Performing semantic recognition based on the interaction information to determine the user's interaction intention; Generate a corresponding reply text according to the interaction intention; The reply text and the service recommendation component are displayed in the agent as the reply result.

12. The method according to any one of claims 1 to 11, wherein: Before recalling at least one corresponding service recommendation component according to the interaction information input by the user in the agent, the method further includes: A plurality of different types of the service recommendation components are configured according to different application scenarios, and each of the intelligent agents is associated with at least one of the service recommendation components.

13. The method according to claim 12, wherein: The type of the service recommendation component includes at least one of the following: a product link component; a link mounting component; a lead conversion component.

14. A human-computer interaction device based on an intelligent agent, comprising: A component recall module, configured to recall at least one corresponding service recommendation component according to the interaction information input by the user in the agent; The reply result generation module is configured to generate a corresponding reply result in the agent based on the interaction information and at least one of the service recommendation components.

15. The device according to claim 14, wherein: Also includes: The agent recall module is configured to recall the agent corresponding to the search text input by the user from multiple pre-configured agents before the component recall module recalls at least one corresponding service recommendation component based on the interaction information input by the user in the agent.

16. The device according to claim 15, wherein: The agent recall module includes: a determination unit, configured to determine the user's search intention according to the search text input by the user; The agent recall unit is configured to recall the agent corresponding to the search intention from a plurality of pre-configured agents according to the search intention.

17. The device according to claim 14, wherein: Also includes: The component information configuration module is configured to pre-configure the corresponding first keyword information for each service recommendation component before the component recall module recalls the corresponding at least one service recommendation component based on the interaction information input by the user in the intelligent agent.

18. The device according to claim 17, wherein: The component information configuration module pre-configures the corresponding first keyword information for each of the service recommendation components, including: The name of each of the service recommendation components is pre-segmented to obtain a corresponding first keyword as the first keyword information.

19. The device according to claim 17, wherein: The component information configuration module pre-configures the corresponding first keyword information for each of the service recommendation components, including: Pre-segmenting the name of each of the service recommendation components to obtain a corresponding first keyword; The user-defined keyword is combined with the first keyword to obtain the first keyword information.

20. The device according to any one of claims 17 to 19, wherein: The component recall module includes: A keyword extraction unit is configured to perform word segmentation processing on the interaction information to obtain corresponding second keyword information; The keyword matching unit is configured to match the second keyword information with the first keyword information, and in response to the second keyword information matching with the first keyword information, recall at least one of the service recommendation components matching the second keyword information.

21. The device according to claim 14, wherein: include: The component information configuration module is configured to pre-configure the corresponding first keyword information for each service recommendation component and configure the corresponding component description information for each service recommendation component before the component recall module recalls at least one corresponding service recommendation component based on the interaction information input by the user in the intelligent agent.

22. The device according to claim 21, wherein The component recall module includes: A keyword extraction unit is configured to perform word segmentation processing on the interaction information to obtain corresponding second keyword information; a keyword matching unit, configured to match the second keyword information with the first keyword information, and in response to the second keyword information matching the first keyword information, recall at least one of the service recommendation components matching the second keyword information; A semantic matching unit is configured to perform semantic recognition on the interaction information in response to the second keyword information not matching the first keyword information, determine the user's interaction intention, semantically match the interaction intention with the component description information, and recall at least one of the service recommendation components that matches the interaction intention.

23. The device according to claim 14, wherein: Also includes: The sorting module is configured to recall at least one corresponding service recommendation component according to the interaction information input by the user in the intelligent agent, and then sort at least one of the service recommendation components to determine at least one of the service recommendation components displayed in the intelligent agent.

24. The device according to claim 14, wherein: The reply result generating module comprises: an intention determination unit, configured to perform semantic recognition according to the interaction information to determine the user's interaction intention; A text generation unit, configured to generate a corresponding reply text according to the interaction intention; A display unit is configured to display the reply text and the service recommendation component as the reply result in the agent.

25. The device according to any one of claims 15 to 24, wherein: Also includes: The component configuration module is configured to configure multiple different types of service recommendation components according to different application scenarios before recalling at least one corresponding service recommendation component based on the interaction information input by the user in the intelligent agent, and associate at least one of the service recommendation components with each of the intelligent agents.

26. The device according to claim 25, wherein The type of the service recommendation component includes at least one of the following: a product link component; a link mounting component; a lead conversion component.

27. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 13.

28. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-13.

29. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 13.

Images