Interaction method and apparatus, and computer-readable storage medium

Abstract

An interaction method and apparatus, and a computer-readable storage medium, relating to the technical field of computers. The interaction method comprises: receiving a natural language instruction of a user, so as to determine demand information of the user, the demand information being obtained by understanding the natural language instruction on the basis of a machine learning model; feeding back to the user a candidate object corresponding to the demand information; in response to the user determining a target object on the basis of the candidate object, prompting the user to provide a resource used to acquire the target object.

Description

Interaction methods, devices and computer-readable storage media

[0001] Cross-reference of related applications

[0002] This application is based on and claims priority to Chinese application No. 202410663549.2, filed on May 27, 2024, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] This disclosure relates to the field of computer technology, and in particular to an interaction method, an interaction device, a computer-readable storage medium, and a computer program product. Background Technology

[0004] In related technologies, if a user wants to obtain a target object through resource exchange, they need to browse the relevant information of each object on the resource exchange page to understand which objects meet their needs; select various resource provision methods; and click "Confirm Provision" to complete the resource exchange. Summary of the Invention

[0005] According to some embodiments of this disclosure, an interaction method is provided, including: receiving a user's natural language instruction to determine the user's demand information, the demand information being obtained based on a machine learning model understanding the natural language instruction; providing the user with candidate objects corresponding to the demand information; and prompting the user to provide resources for obtaining the target object in response to the user determining the target object based on the candidate objects.

[0006] In some embodiments, the requirement information includes object type information and object attribute information. Receiving a user's natural language instruction to determine the user's requirement information includes: receiving a first natural language instruction from the user to determine the object type information required by the user, wherein the object type information is obtained based on the understanding of the first natural language instruction by a machine learning model; providing feedback to the user with first query information based on the object type information, wherein the first query information is used to inquire about the object attribute information corresponding to the object type information; and receiving a second natural language instruction from the user based on the first query information to determine the object attribute information, wherein the object attribute information is obtained based on the understanding of the second natural language instruction by a machine learning model.

[0007] In some embodiments, in response to the user determining a target object based on candidate objects, prompting the user to provide resources for obtaining the target object includes: providing the user with second question information, the second question information being used to inquire about the user's method of providing resources; receiving a third natural language instruction from the user based on the second question information, to determine the user's method of providing resources, the method of providing resources being obtained based on a machine learning model understanding the third natural language instruction.

[0008] In some embodiments, providing the user with second question information includes: providing the user with a selection of candidate resource provision methods, wherein the candidate resource provision method is determined based on the user's identity information.

[0009] In some embodiments, the candidate resource provision method is determined from multiple resource provision methods based on the resource provision success rate and / or historical usage count of each of the multiple resource provision methods associated with the identity information.

[0010] In some embodiments, in response to the user determining a target object based on candidate objects, prompting the user to provide resources for obtaining the target object includes: authenticating the user based on the user's image information and / or voice information; and in response to the user passing the authentication, obtaining the resources provided by the user so that the user can obtain the target object.

[0011] In some embodiments, providing feedback to the user on the candidate objects corresponding to the request information includes: providing the user with object attribute information of the candidate objects; receiving the user's fourth natural language instruction to determine the candidate objects that the user is interested in, wherein the candidate objects that the user is interested in are obtained based on the understanding of the fourth natural language instruction by a machine learning model; and providing the user with detailed information on the candidate objects that the user is interested in.

[0012] In some embodiments, the object attribute information includes at least one of image information, resource information for obtaining candidate objects, usage information, and usage condition information, and the detailed information includes resource information provided by the candidate object to the user.

[0013] According to some other embodiments of this disclosure, an interactive device is provided, including: a receiving unit for receiving a user's natural language instruction to determine the user's demand information, wherein the demand information is obtained based on the understanding of the natural language instruction by a machine learning model; a feedback unit for providing feedback to the user on candidate objects corresponding to the demand information; and a prompting unit for prompting the user to provide resources for obtaining the target object in response to the user determining the target object based on the candidate objects.

[0014] In some embodiments, the requirement information includes object type information and object attribute information. The receiving unit receives a first natural language instruction from the user to determine the object type information required by the user. The object type information is obtained by understanding the first natural language instruction based on a machine learning model. The feedback unit provides feedback to the user with first question information based on the object type information. The first question information is used to inquire about the object attribute information corresponding to the object type information. The receiving unit receives a second natural language instruction from the user based on the first question information to determine the object attribute information. The object attribute information is obtained by understanding the second natural language instruction based on a machine learning model.

[0015] In some embodiments, the feedback unit provides the user with a second question, which inquires about the user's resource provision method; the receiving unit receives a third natural language instruction from the user based on the second question to determine the user's resource provision method, which is obtained based on a machine learning model's understanding of the third natural language instruction.

[0016] In some embodiments, the feedback unit provides the user with candidate resource provision methods for the user to choose from, as a second question, and the candidate resource provision method is determined based on the user's identity information.

[0017] In some embodiments, the candidate resource provision method is determined from multiple resource provision methods based on the resource provision success rate and / or historical usage count of each of the multiple resource provision methods associated with the identity information.

[0018] In some embodiments, the prompting unit authenticates the user based on the user's image information and / or voice information, and in response to the user passing the authentication, obtains the resources provided by the user so that the user can obtain the target object.

[0019] In some embodiments, the feedback unit provides the user with object attribute information of the candidate object; the receiving unit receives the user's fourth natural language instruction to determine the candidate object of interest to the user, the candidate object of interest to the user is obtained based on the understanding of the fourth natural language instruction by a machine learning model; the feedback unit provides the user with detailed information about the candidate object of interest.

[0020] In some embodiments, the object attribute information includes at least one of image information, resource information for obtaining candidate objects, usage information, and usage condition information, and the detailed information includes resource information provided by the candidate object to the user.

[0021] According to further embodiments of this disclosure, an interactive device is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the interactive method of any of the above embodiments based on instructions stored in the memory device.

[0022] According to further embodiments of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the interaction method of any of the above embodiments.

[0023] According to further embodiments of this disclosure, a computer program product is also provided, including instructions that, when executed by a processor, cause the processor to perform the interaction method according to any of the foregoing embodiments. Attached Figure Description

[0024] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the specification, serve to explain the principles of this disclosure.

[0025] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0026] Figure 1 shows flowcharts of some embodiments of the interaction method of this disclosure;

[0027] Figures 2a-2e illustrate schematic diagrams of some embodiments of the interaction method of this disclosure;

[0028] Figures 3a-3c illustrate schematic diagrams of some embodiments of the resource exchange method of this disclosure;

[0029] Figure 3d shows a flowchart of some embodiments of the resource exchange method of this disclosure;

[0030] Figure 4 shows flowcharts of some embodiments of the interactive device of this disclosure;

[0031] Figure 5 shows a block diagram of some other embodiments of the interactive device of this disclosure;

[0032] Figure 6 shows a block diagram of some further embodiments of the interactive device of this disclosure. Detailed Implementation

[0033] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0034] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0035] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0036] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0037] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0038] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0039] The inventors of this disclosure have discovered the following problem in the aforementioned related technologies: users need to browse a large amount of information in order to select the target object they want to obtain, resulting in low efficiency of user interaction.

[0040] In view of this, this disclosure proposes an interactive technology solution that can improve the efficiency of user interaction.

[0041] As mentioned earlier, users need to browse a large amount of information to obtain the target object that meets their needs through resource exchange. This increases the time and effort costs of interaction, resulting in low efficiency in user interaction.

[0042] To address or partially address the aforementioned technical problems, the technical solution disclosed herein utilizes machine learning technology to interact with users based on natural language, thereby understanding user needs and facilitating resource exchange. In this way, users can obtain target objects that meet their needs without browsing large amounts of information, thus improving the efficiency of user interaction.

[0043] In addition, users not only need to browse a large amount of information, but also often need to manually complete the resource exchange process, resulting in low efficiency of user interaction.

[0044] Taking a user wanting to obtain group-buying coupons on an app as an example, resource exchange can be completed through the following process: The user first needs to manually open the relevant group-buying app; the user browses available group-buying projects within the app and views the project details; the user selects the desired group-buying coupon by manually clicking or other means according to their needs; the user clicks the purchase button by manually clicking or other means, and may need to log in or register; the user clicks or other means to manually select a suitable resource provision method to complete the resource exchange.

[0045] This manual method requires users to open the resource exchange page and manually select the resource provision method, making the operation cumbersome and requiring a long time and effort, resulting in low efficiency in user interaction.

[0046] Furthermore, this manual method relies too heavily on the interactive interface. For certain groups, such as the elderly and visually impaired, this can present operational difficulties, leading to inefficient user interaction.

[0047] Moreover, this manual method has limitations in application. In certain scenarios where manual operation is inconvenient, such as driving, this manual method may not be suitable, leading to technical problems such as low efficiency in user interaction.

[0048] To address or partially address the aforementioned technical problems, the technical solution disclosed herein utilizes machine learning technology to interact with users based on natural language, thereby determining user needs and simplifying the interactive operations during resource exchange. In this way, users can complete resource exchanges without cumbersome manual operations, thus improving the efficiency of user interaction.

[0049] For example, the technical solution of this disclosure can be implemented through the following embodiments.

[0050] Figure 1 shows a flowchart of some embodiments of the interaction method of this disclosure.

[0051] As shown in Figure 1, in step 110, the user's natural language instruction is received to determine the user's demand information. The demand information is obtained based on the understanding of the natural language instruction by the machine learning model.

[0052] In some embodiments, the demand information includes object type information and object attribute information. For example, object type information includes item type, service type, etc., and service type may include non-physical objects such as group-buying coupons and discount coupons. For example, object attribute information may include at least one of the following: object image information, resource information used to obtain the object, object usage method information, and object usage condition information.

[0053] In this way, users do not need to browse a lot of information; they can directly express their needs through natural language during the interaction, thereby improving the efficiency of user interaction.

[0054] In step 120, candidate objects corresponding to the user's request information are provided. For example, the object attribute information of the candidate objects is provided to the user so that the user can select the candidate object of interest.

[0055] In step 130, in response to the user identifying the target object based on the candidate objects, the user is prompted to provide resources to acquire the target object. For example, the interaction with the user can be based on natural language to determine how the user will provide resources and complete the resource exchange.

[0056] In the above embodiments, machine learning technology is used to interact with users based on natural language to understand user needs and then complete resource exchanges. In this way, users can obtain target objects that meet their needs without browsing large amounts of information, thereby improving the efficiency of user interaction.

[0057] The following examples illustrate how to interact with users based on natural language to determine user needs.

[0058] In some embodiments, a first natural language instruction from a user is received to determine the object type information requested by the user. The object type information is obtained based on the understanding of the first natural language instruction by a machine learning model. For example, the user can express the requested object type information by issuing a first natural language instruction to the agent; the agent understands the first natural language instruction based on a machine learning model to determine the object type information requested by the user.

[0059] For example, users can issue first natural language commands to the agent through voice or other non-manual means in the interaction interface; the agent understands the first natural language commands based on a machine learning model to determine the type of object information required by the user, and then provides feedback to the user in natural language in the interaction interface.

[0060] In this way, users do not need to browse a lot of information; they can directly express their needs through natural language during the interaction, thereby improving the efficiency of user interaction.

[0061] For example, user requirements can be determined through the embodiments shown in Figures 2a and 2b.

[0062] Figures 2a and 2b show schematic diagrams of some embodiments of the interaction method of this disclosure.

[0063] As shown in Figure 2a, the agent's interactive interface includes a user dialog box 21a for displaying the user's natural language commands and an agent dialog box 22a for displaying the agent's natural language responses. For example, in the interactive interface with the agent, the user can issue a first natural language command to the agent through manual methods such as keyboard input or non-manual methods such as voice input to express the object type information of the user's needs.

[0064] For example, if a user wants to obtain information about dining services, they can issue a first natural language command via voice: "Recommend a good restaurant nearby." In response to the user's first natural language command, the command "Recommend a good restaurant nearby" can be displayed in the user dialog box 21a.

[0065] In some embodiments, the agent provides the user with candidate objects corresponding to their request information. For example, the agent can use a machine learning model to process the first natural language instruction in the user dialog box 21a to determine the object type information requested by the user. For instance, the agent uses a machine learning model to process "Recommend a good restaurant nearby" and determines that the object type information requested by the user is dining service; the agent searches for candidate objects that match the user's request and provides the object attribute information of the candidate objects in the agent dialog box 22a.

[0066] For example, object attribute information includes at least one of the following: image information, resource information for obtaining candidate objects, usage information, usage condition information, and applicable object information.

[0067] As shown in Figure 2a, the agent can provide object attribute information of at least one candidate object that meets the user's needs in the agent dialog box 22a. This information may include the candidate object's name, location, and resource information used to obtain the candidate object, such as "Restaurant A, located at location A, average cost per person: xx yuan," or "Restaurant B, located at location B, average cost per person: xx yuan," for the user to select.

[0068] As shown in Figure 2b, the agent's interactive interface includes a user dialog box 21b for displaying the user's natural language commands and an agent dialog box 22b for displaying the agent's natural language responses.

[0069] For example, if a user wants to obtain a group-buying service for dining, they can issue a first natural language command via voice: "What suitable group-buying coupons are available?" In response to the user's first natural language command, the command "What suitable group-buying coupons are available?" can be displayed in the user dialog box 21b.

[0070] In some embodiments, the agent provides the user with candidate objects corresponding to their desired information. For example, the agent can use a machine learning model to process the first natural language instruction in the user dialog box 21b to determine the object type information required by the user. For instance, the agent uses a machine learning model to process "What are some suitable group-buying coupons?" and determines that the object type information required by the user is a group-buying service for dining; the agent searches for candidate objects that match the user's needs and provides the object attribute information of the candidate objects in the agent dialog box 22b.

[0071] For example, the agent can provide object attribute information of at least one candidate object that meets the user's requirements in the agent dialog box 22b. This information may include the candidate object's image information, name information, location information, resource information used to obtain the candidate object, and applicable object information.

[0072] For example, Restaurant A offers group-buying services to meet user needs, including two types of group-buying vouchers: Group-buying A1 and Group-buying A2. The object attribute information for Group-buying A1 includes "an image of Restaurant A, the name of Restaurant A, the location of Restaurant A (location A), the price, and it is applicable to two people"; the object attribute information for Group-buying A2 includes "an image of Restaurant A, the name of Restaurant A, the location of Restaurant A (location A), the price, and it is applicable to four people".

[0073] For example, Restaurant B also offers group-buying services to meet user needs, including two types of group-buying vouchers: Group-buying voucher B1 and Group-buying voucher B2. The object attribute information for Group-buying voucher B1 includes "an image of Restaurant B, the name of Restaurant B, the location of Restaurant B (location B), price, and is applicable to two people"; the object attribute information for Group-buying voucher B2 includes "an image of Restaurant B, the name of Restaurant B, the location of Restaurant B (location B), price, and is applicable to four people".

[0074] In the above embodiments, machine learning techniques are used to understand the natural language initiated by the user to determine the type of object the user needs, and then provide the user with candidate objects that meet their needs. In this way, users can obtain the target object that meets their needs without having to browse through a large amount of information, thereby improving the efficiency of user interaction.

[0075] The following examples illustrate how, after providing the user with information about the object type of their needs, natural language interaction can be used to determine the object attribute information of the user's needs, so as to more accurately provide the user with candidate objects that meet their needs.

[0076] In some embodiments, based on the object type information required by the user, a first question is fed back to the user. The first question is used to inquire about the object attribute information corresponding to the object type information. A second natural language instruction fed back by the user based on the first question is received to determine the object attribute information. The object attribute information is obtained by understanding the second natural language instruction based on a machine learning model.

[0077] For example, by understanding the user's first natural language instruction, "What suitable group-buying coupons are available?", the system determines that the user wants to obtain information about dining group-buying services. It then provides the user with the first query, "How many people are dining?", to inquire about the corresponding object attribute information (i.e., applicable object information). Next, by understanding the user's second natural language instruction, "Dining for two", the system determines that the user wants to obtain information about dining group-buying services suitable for two people. Based on the user's desired object type information, "dining group-buying services," and the object attribute information, "suitable for two people," the system provides the user with candidate options that meet their needs. For example, Group-buying A1 includes "an image of Restaurant A, the name of Restaurant A, the location of Restaurant A (location A), the price, and is suitable for two people," while Group-buying B1 includes "an image of Restaurant B, the name of Restaurant B, the location of Restaurant B (location B), the price, and is suitable for two people" for the user to choose from.

[0078] In some embodiments, based on providing the user with candidate objects whose object type information can meet their needs, it is also possible to directly receive the second natural language instruction initiated by the user without initiating the first question to the user.

[0079] In the above embodiments, machine learning techniques are used to understand the natural language initiated by the user to determine the type and attribute information of the object requested by the user, and then provide the user with candidate objects that meet their needs. In this way, users can obtain the target object that meets their needs without having to browse through a large amount of information, thereby improving the efficiency of user interaction.

[0080] The following examples illustrate how, after providing users with candidate objects that meet their needs, the target object for those needs can be determined through natural language interaction.

[0081] In some embodiments, a user's fourth natural language instruction is received to determine candidate objects of interest to the user. The candidate objects of interest are obtained based on the understanding of the fourth natural language instruction by a machine learning model. For example, the user can use the fourth natural language instruction (such as specifying the name of the object) to specify the candidate objects of interest from the candidate objects; the user can also use the fourth natural language instruction to express filtering conditions (such as specifying location information, resource information for obtaining candidate objects, applicable object information, etc.) to filter out the candidate objects of interest from the candidate objects.

[0082] In some embodiments, detailed information about candidate objects of interest is provided to the user. For example, the detailed information may include resource information that the candidate object offers to the user. Resource information may include services, items, etc., that can be offered to the user.

[0083] For example, the candidate objects of interest to the user can be determined through the embodiments shown in Figures 2c-2d.

[0084] Figures 2c-2e illustrate schematic diagrams of some embodiments of the interaction methods of this disclosure.

[0085] As shown in Figure 2c, the user inputs natural language through the interactive interface to identify the candidate object of interest as a group-buying coupon for restaurant A. The user dialog box 21c can then display "Take a look at restaurant A". For example, the agent can display candidate objects that meet the user's needs in the interactive interface. This includes displaying object attribute information such as usage conditions (e.g., "Applicable Monday to Sunday", "Applicable Monday to Friday"), applicable object information (e.g., "Two-person meal", "Four-person meal"), and resource information used to obtain the candidate object (e.g., "Price A1", "Price A2").

[0086] For example, the agent can also provide feedback in natural language to the user to ask for filtering criteria (such as applicable object information) for candidates that the user is interested in, such as displaying "Restaurant A has the following group deals, how many people are you dining with?" in the agent dialog box 22c to determine the candidates that the user is interested in.

[0087] As shown in Figure 2d, the user responds to the agent's queries by issuing fourth natural language commands based on the object attribute information of the candidate objects provided by the agent. For example, the fourth natural language command "We are two people eating" can be displayed in the user dialog box 21d. The agent uses a machine learning model to understand that the candidate object filtering condition provided by the fourth natural language command is the object attribute information "suitable for two people". The agent dialog box 22d can display detailed information of the candidate objects that meet the filtering conditions, such as "Group purchase details, dish A11, dish A12...", etc., providing resource information to the user.

[0088] As shown in Figure 2e, the agent can also display detailed information of candidate objects that meet the filtering criteria in the agent dialog box 22e, such as the image information of dish A11 and dish A12.

[0089] In the above embodiments, the target object that the user wants to obtain is determined through natural language interaction. Based on this, resource exchange can be completed through natural language interaction, enabling the user to obtain the target object.

[0090] The following examples illustrate how to interact with users based on natural language to complete resource exchange.

[0091] In some embodiments, a second question is asked to the user, which inquires about the method of resource provision; a third natural language instruction is received from the user based on the second question to determine the method of resource provision, which is obtained by understanding the third natural language instruction through a machine learning model.

[0092] For example, the intelligent agent can ask the user a second question in the interactive interface, "How do I pay?"; the user can then respond with a third natural language instruction, "Pay with XXX card".

[0093] In some embodiments, a candidate resource provision method may be provided to the user for selection as a second question. The candidate resource provision method is determined based on the user's identity information.

[0094] For example, the intelligent agent can ask the user a second question in the interactive interface: "Would you like to pay with card A or card B?"; the user can then respond with a third natural language instruction: "Pay with card A".

[0095] In some embodiments, the candidate resource provision method is determined from multiple resource provision methods based on the resource provision success rate and / or historical usage count of each of the multiple resource provision methods associated with the identity information.

[0096] For example, an intelligent agent can determine the resource provision methods under a user's name based on the user's identity information, including three types: card A, card B, and card C; and recommend card A and card B, which have the highest success rate in providing resources, as candidate resource provision methods to the user.

[0097] In some embodiments, the user is authenticated based on the user's image information and / or voice information; in response to the user's successful authentication, the resources provided by the user are obtained so that the user can obtain the target object.

[0098] For example, resource exchange can be accomplished through the embodiments shown in Figures 3a-3c.

[0099] Figures 3a-3c show schematic diagrams of some embodiments of the resource exchange method of this disclosure.

[0100] As shown in Figure 3a, the user identifies the target object through natural language commands in the user dialog box 31 of the interactive interface. The intelligent agent can determine the resource provision methods under the user's name based on their identity information, including card A, card B, and card C; and recommend card A, which has the highest success rate, as the candidate resource provision method to the user. For example, the intelligent agent can display the natural language message "Okay, shall I pay with card A?" in the intelligent agent dialog box 32.

[0101] In response to the user specifying a resource provision method via natural language instructions in user dialog box 33, the agent can prompt the user in agent dialog box 34 to verify their identity to adopt a candidate resource provision method. For example, the agent dialog box 34 can prompt "Please verify your identity by scanning your face next".

[0102] As shown in Figure 3b, the user can be redirected to the face recognition authentication interface, where they will be prompted to complete the face recognition authentication.

[0103] As shown in Figure 3c, in response to a user passing facial recognition authentication, information indicating a successful resource exchange can be displayed to the user, such as "I have purchased a group-buying voucher A1 for restaurant A." For example, the system can also display object attribute information such as how to use the acquired target object, such as "You can redeem this voucher code at the restaurant."

[0104] In the above embodiments, machine learning technology is used to interact with users based on natural language, thereby determining user needs and simplifying the interaction process during resource exchange. This eliminates the need for cumbersome manual operations, improving the efficiency of user interaction.

[0105] For example, resource exchange can be accomplished through the embodiment shown in Figure 3d.

[0106] Figure 3d shows a flowchart of some embodiments of the resource exchange method of this disclosure.

[0107] As shown in Figure 3d, in step 310, the user confirms the target object to be obtained. For example, the user initiates an object retrieval request to the agent via voice command, such as expressing the desire to obtain a group-buying coupon; the agent interacts with the user through natural language to inquire about the object attribute information of the group-buying coupon needed, such as the group-buying item and quantity; the agent provides corresponding group-buying coupon options and detailed information based on the user's feedback; the user confirms the group-buying coupon they want to obtain through voice feedback.

[0108] In step 320, the resource provision method associated with the user's identity information is determined, such as provision via a certain card or other mobile provision methods.

[0109] In step 330, a candidate resource provisioning method is determined from multiple resource provisioning methods based on the resource provisioning success rate and / or historical usage count of multiple resource provisioning methods associated with the user's identity information.

[0110] In step 340, the user is provided with recommended candidate resource provision methods for selection.

[0111] In step 350, the user is authenticated based on the user's image information and / or voice information.

[0112] In step 360, after the intelligent agent performs the necessary identity verification on the user, it guides the user to complete the resource exchange process. For example, it can display a message indicating a successful resource exchange, such as "I have purchased a group-buying voucher A1 for restaurant A"; the user can then use the voucher at the merchant that provided it via voice.

[0113] In the above embodiments, by interacting with users based on natural language, cumbersome interface operations are avoided, and the target object can be obtained directly through natural language. This enables more convenient resource exchange and provides a better user experience. In addition, it overcomes the limitations of manual methods in various application scenarios, providing a more convenient, efficient, and flexible way to exchange resources.

[0114] Figure 4 shows flowcharts of some embodiments of the interactive device of this disclosure.

[0115] As shown in Figure 4, the interactive device 4 includes: a receiving unit 41, used to receive the user's natural language instructions to determine the user's demand information, the demand information being obtained by understanding the natural language instructions based on a machine learning model; a feedback unit 42, used to provide feedback to the user on the candidate objects corresponding to the demand information; and a prompting unit 43, used to prompt the user to provide resources for obtaining the target object in response to the user determining the target object based on the candidate objects.

[0116] In some embodiments, the requirement information includes object type information and object attribute information. The receiving unit 41 receives a first natural language instruction from the user to determine the object type information required by the user. The object type information is obtained by understanding the first natural language instruction based on a machine learning model. The feedback unit 42 provides feedback to the user with first question information based on the object type information. The first question information is used to inquire about the object attribute information corresponding to the object type information. The receiving unit 41 receives a second natural language instruction from the user based on the first question information to determine the object attribute information. The object attribute information is obtained by understanding the second natural language instruction based on a machine learning model.

[0117] In some embodiments, the feedback unit 42 provides the user with a second question, which is used to inquire about the user's resource provision method; the receiving unit 41 receives the user's third natural language instruction based on the second question to determine the user's resource provision method, which is obtained based on the understanding of the third natural language instruction by a machine learning model.

[0118] In some embodiments, the feedback unit 42 provides the user with a selection of candidate resource provision methods as a second question, and the candidate resource provision method is determined based on the user's identity information.

[0119] In some embodiments, the candidate resource provision method is determined from multiple resource provision methods based on the resource provision success rate and / or historical usage count of each of the multiple resource provision methods associated with the identity information.

[0120] In some embodiments, the prompting unit 43 authenticates the user based on the user's image information and / or voice information, and in response to the user passing the authentication, obtains the resources provided by the user so that the user can obtain the target object.

[0121] In some embodiments, the feedback unit 42 provides the user with object attribute information of the candidate object; the receiving unit 41 receives the user's fourth natural language instruction to determine the candidate object of interest to the user, and the candidate object of interest to the user is obtained based on the understanding of the fourth natural language instruction by a machine learning model; the feedback unit 42 provides the user with detailed information about the candidate object of interest.

[0122] In some embodiments, the object attribute information includes at least one of image information, resource information for obtaining candidate objects, usage information, and usage condition information, and the detailed information includes resource information provided by the candidate object to the user.

[0123] Figure 5 shows a block diagram of some other embodiments of the interactive device of this disclosure.

[0124] As shown in FIG5, the interactive device 5 of this embodiment includes a memory 51 and a processor 52 coupled to the memory 51. The processor 52 is configured to execute the interactive method in any embodiment of this disclosure based on instructions stored in the memory 51.

[0125] The memory 51 may include, for example, system memory, fixed non-volatile storage media, etc. The system memory stores, for example, the operating system, application programs, boot loader, database, and other programs.

[0126] Figure 6 shows a block diagram of some further embodiments of the interactive device of this disclosure.

[0127] As shown in FIG6, the interactive device 6 of this embodiment includes a memory 610 and a processor 620 coupled to the memory 610. The processor 620 is configured to execute the interactive method in any of the foregoing embodiments based on instructions stored in the memory 610.

[0128] The memory 610 may include, for example, system memory, fixed non-volatile storage media, etc. The system memory may store, for example, the operating system, application programs, boot loader, and other programs.

[0129] The interactive device 6 may also include an input / output interface 630, a network interface 640, and a storage interface 650. These interfaces 630, 640, and 650, as well as the memory 610 and processor 620, can be connected, for example, via a bus 660. The input / output interface 630 provides a connection interface for input / output devices such as a monitor, mouse, keyboard, touchscreen, microphone, and speakers. The network interface 640 provides a connection interface for various networked devices. The storage interface 650 provides a connection interface for external storage devices such as SD cards and USB flash drives.

[0130] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] The interaction methods, interaction devices, computer-readable storage media, and computer program products according to this disclosure have been described in detail. To avoid obscuring the concept of this disclosure, some details known in the art have not been described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein based on the above description.

[0132] The methods and systems of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the specific order described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0133] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. An interaction method, comprising: receiving a natural language instruction of a user to determine requirement information of the user, the requirement information being obtained based on a machine learning model understanding the natural language instruction; feeding back a candidate object corresponding to the requirement information to the user; in response to the user determining a target object according to the candidate object, prompting the user to provide a resource for obtaining the target object.

2. The interaction method of claim 1, wherein, The requirement information includes object type information and object attribute information, The receiving a natural language instruction of a user to determine requirement information of the user comprises: receiving a first natural language instruction of the user to determine object type information required by the user, the object type information being obtained based on the machine learning model understanding the first natural language instruction; feeding back first question information to the user according to the object type information, the first question information being used to inquire object attribute information corresponding to the object type information; receiving a second natural language instruction fed back by the user according to the first question information to determine the object attribute information, the object attribute information being obtained based on the machine learning model understanding the second natural language instruction.

3. The interaction method according to claim 1 or 2, wherein, The response to the user determining a target object according to the candidate object, prompting the user to provide a resource for obtaining the target object comprises: feeding back second question information to the user, the second question information being used to inquire the resource providing manner of the user; receiving a third natural language instruction fed back by the user according to the second question information to determine the resource providing manner of the user, the resource providing manner being obtained based on the machine learning model understanding the third natural language instruction.

4. The interaction method of claim 3, wherein, The feeding back second question information to the user comprises: feeding back candidate resource providing manners for the user to select as the second question information, the candidate resource providing manners being determined according to identity information of the user.

5. The interaction method of claim 4, wherein, The candidate resource providing manners are determined from a plurality of resource providing manners associated with the identity information according to a resource providing success rate and / or a historical use frequency of each of the plurality of resource providing manners.

6. The interaction method according to any of claims 1-5, wherein, The response to the user determining a target object according to the candidate object, prompting the user to provide a resource for obtaining the target object comprises: performing identity authentication on the user according to image information and / or sound information of the user; in response to the user passing the identity authentication, obtaining the resource provided by the user so that the user obtains the target object.

7. The interaction method according to any of claims 1-6, wherein, The feeding back a candidate object corresponding to the requirement information to the user comprises: feeding back object attribute information of the candidate object to the user; receiving a fourth natural language instruction of the user to determine a candidate object of interest of the user, the candidate object of interest of the user being obtained based on the machine learning model understanding the fourth natural language instruction; feeding back detailed information of the candidate object of interest to the user.

8. The interaction method of claim 7, wherein, The object attribute information includes at least one of image information, resource information for obtaining the candidate object, usage mode information, and usage condition information, and the detailed information includes resource information provided by the candidate object to the user. 9.An interactive apparatus, comprising: a receiving unit configured to receive a natural language instruction of a user to determine requirement information of the user, the requirement information being obtained based on a machine learning model to understand the natural language instruction; a feedback unit configured to feed back a candidate object corresponding to the requirement information to the user; a prompting unit configured to prompt the user to provide a resource for obtaining a target object in response to the user determining the target object based on the candidate object. 10.An interactive apparatus, comprising: a memory; and a processor coupled to the memory, the processor configured to execute an instruction stored in the memory to perform the interactive method of any one of claims 1-8. 11.A computer readable storage medium having stored thereon a computer program, the program being executed by a processor to implement the interactive method of any one of claims 1-8. 12.A computer program product comprising instructions which, when executed by a processor, cause the processor to carry out the interactive method of any one of claims 1-8.

Images