An information processing method, apparatus, electronic device, and storage medium

Abstract

The present disclosure relates to an information processing method, apparatus, electronic device, and storage medium. The method includes: receiving an information acquisition request of a target account; determining a current conversation scenario corresponding to the information acquisition request; obtaining key information in a historical conversation scenario of the target account; sending a target prompt message to the target account; and responding to the information acquisition request based on selection information sent by the target account in response to the target prompt message. When the server reuses key information in the context information of the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, so that the information acquisition request can be accurately responded to, and the accuracy rate of the information provided to the user is relatively high.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular, to an information processing method, apparatus, electronic device, and storage medium. Background Art

[0002] With the development of natural language technology, multi-turn dialogue technology is increasingly applied in scenarios such as intelligent customer service and intelligent outbound calls. While improving the efficiency of user services and user experience, it also greatly reduces the costs of enterprises.

[0003] In related technologies, most multi-turn dialogue systems basically complete the interaction of the dialogue according to a preset process. The multi-turn dialogue system may give an irrelevant answer. For example, in the first round of dialogue, the user queries the weather in Beijing. In the second round of dialogue, the user wants to rent a car. At this time, the multi-turn dialogue system recommends car rental information in Beijing to the user. However, in actual applications, the user may want to rent a car in Shanghai, thus resulting in an irrelevant answer.

[0004] As can be seen from the above description, in related technologies, the information provided by the multi-turn dialogue system to the user may not be the information that the user really wants to obtain, that is, the accuracy rate of the information provided to the user is relatively low. Summary of the Invention

[0005] In order to solve the above technical problems existing in related technologies, the present disclosure provides an information processing method, apparatus, electronic device, and storage medium. The technical solutions of the present disclosure are as follows:

[0006] According to the first aspect of the embodiments of the present disclosure, an information processing method is provided, including:

[0007] Receiving an information acquisition request of a target account;

[0008] Determining a current dialogue scenario corresponding to the information acquisition request;

[0009] Obtaining key information in the historical dialogue scenario of the target account; the historical dialogue scenario is different from the current dialogue scenario; the key information is information of a target type, and the target type is the type of information required to be acquired by the information acquisition request;

[0010] Sending a target prompt message to the target account; the target prompt message is used to prompt the target account to confirm whether to reuse the key information in the current dialogue scenario;

[0011] Responding to the information acquisition request based on the selection information sent by the target account for the target prompt message.

[0012] Optionally, the obtaining the key information in the historical dialogue scenario of the target account includes:

[0013] Determine the information type to which the information required for the current conversation scenario belongs;

[0014] Determine the target information belonging to the information type in the context information of the historical conversation scenario;

[0015] Determine the target information as the key information in the context information of the historical conversation scenario.

[0016] Optionally, after obtaining the key information in the historical conversation scenario of the target account, it further includes:

[0017] Store the key information as the context information of the current conversation scenario and set a historical scenario label for the context information; the historical scenario label is used to indicate that the context information belongs to the information obtained in the historical conversation scenario;

[0018] Correspondingly, the sending of the target prompt information to the target account includes:

[0019] Obtain the context information of the current conversation scenario;

[0020] When the obtained context information has a historical scenario label, generate a target prompt information based on the obtained context information and send the target prompt information to the target account.

[0021] Optionally, the selected information includes affirmative information; the responding to the information acquisition request based on the selected information sent by the target account for the target prompt information includes:

[0022] When the selected information is affirmative information, obtain the target information corresponding to the information acquisition request according to the key information and return the target information to the target account.

[0023] Optionally, the selected information includes negative information; the responding to the information acquisition request based on the selected information sent by the target account for the target prompt information includes:

[0024] Generate interaction information and send the interaction information to the target account; the interaction information is used to prompt the target account to input the information required for the current conversation scenario;

[0025] Obtain the target information corresponding to the information acquisition request based on the reply information of the target account for the interaction information and return the target information to the target account.

[0026] Optionally, before obtaining the key information in the historical conversation scenario of the target account, it further includes:

[0027] Determine the moment when the information acquisition request is received;

[0028] When there is a historical conversation record corresponding to the target account within a preset time period before the moment, execute acquiring the key information in the historical conversation scenario of the target account.

[0029] According to a second aspect of the embodiments of the present disclosure, there is provided an information processing apparatus, including:

[0030] A request acquisition module, configured to execute receiving an information acquisition request of a target account;

[0031] A conversation scenario determination module, configured to execute determining the current conversation scenario corresponding to the information acquisition request;

[0032] A key information acquisition module, configured to execute acquiring the key information in the historical conversation scenario of the target account; the historical conversation scenario is different from the current conversation scenario; the key information is information of a target type, and the target type is the type of information required to be acquired by the information acquisition request;

[0033] A prompt information sending module, configured to execute sending target prompt information to the target account; the target prompt information is used to prompt the target account to confirm whether to reuse the key information into the current conversation scenario;

[0034] A request response module, configured to execute responding to the information acquisition request based on the selection information sent by the target account for the target prompt information.

[0035] Optionally, the key information acquisition module is specifically configured to execute:

[0036] Determine the information type to which the information required by the current conversation scenario belongs;

[0037] Determine the target information belonging to the information type in the context information of the historical conversation scenario;

[0038] Determine the target information as the key information in the context information of the historical conversation scenario.

[0039] Optionally, it further includes:

[0040] A scenario label setting module, configured to, after acquiring the key information in the historical conversation scenario of the target account, store the key information as the context information of the current conversation scenario, and set a historical scenario label for the context information; the historical scenario label is used to characterize that the context information belongs to the information acquired in the historical conversation scenario;

[0041] Correspondingly, the prompt information sending module is specifically configured to execute:

[0042] Obtain the context information of the current conversation scenario;

[0043] When the obtained context information has a historical scenario label, generate target prompt information based on the obtained context information, and send the target prompt information to the target account.

[0044] Optionally, the selection information includes affirmative information; the request response module is specifically configured to execute:

[0045] When the selection information is affirmative information, obtain the target information corresponding to the information acquisition request according to the key information, and return the target information to the target account.

[0046] Optionally, the selection information includes negative information; the request response module is specifically configured to execute:

[0047] Generate interaction information, and send the interaction information to the target account; the interaction information is used to prompt the target account to input the information required for the current conversation scenario;

[0048] Based on the reply information of the target account to the interaction information, obtain the target information corresponding to the information acquisition request, and return the target information to the target account.

[0049] Optionally, it further includes:

[0050] A moment determination module, configured to execute to determine the moment when the information acquisition request is received before obtaining the key information in the historical conversation scenario of the target account;

[0051] When there is a historical conversation record corresponding to the target account within a preset time period before the moment, trigger the key information acquisition module to execute the acquisition of the key information in the historical conversation scenario of the target account.

[0052] According to the third aspect of the embodiments of the present disclosure, an electronic device is provided, including:

[0053] A processor;

[0054] A memory for storing executable instructions of the processor;

[0055] Wherein, the processor is configured to execute the instructions to implement the method as described in the first aspect.

[0056] According to a fourth aspect of the embodiments of the present disclosure, a storage medium is provided. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the method described in the first aspect.

[0057] According to a fifth aspect of the embodiments of the present disclosure, a computer program product is provided, characterized in that when it runs on a computer, the computer implements the method described in the first aspect.

[0058] The technical solution provided by the embodiments of the present disclosure, after receiving an information acquisition request of a target account, determines the current conversation scenario corresponding to the information acquisition request, and acquires key information in the historical conversation scenario of the target account; and the historical conversation scenario is different from the current conversation scenario, that is to say, the target account has carried out cross-scenario conversation interaction with the server. By acquiring the key information in the historical conversation scenario and sending a target prompt message to the target account, the target account is prompted to confirm whether to reuse the key information into the current conversation scenario through the target prompt message. Finally, the selection information sent by the target account in response to the target prompt message is received, and the information acquisition request is responded based on the selection information.

[0059] As can be seen from the above description, when the target account and the server perform cross-scenario conversation interaction, when the server reuses the key information in the historical conversation scenario into the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection information of the target account for the key information to be reused. Therefore, the server can accurately respond to the information acquisition request, and the accuracy rate of the information provided to the user is relatively high. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a flowchart of an information processing method shown according to an exemplary embodiment;

[0061] Figure 2 is a flowchart of another information processing method shown according to an exemplary embodiment;

[0062] FIG. 3(a) and FIG. 3(b) are flowcharts of the complete technical solution of the embodiments of the present disclosure;

[0063] Figure 4 is a block diagram of an information processing device shown according to an exemplary embodiment;

[0064] Figure 5 is a block diagram of another information processing device shown according to an exemplary embodiment. DETAILED DESCRIPTION

[0065] To enable those of ordinary skill in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0066] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0067] With the development of natural language technology, multi-turn dialogue technology is increasingly applied in scenarios such as intelligent customer service and intelligent outbound calls. While improving the efficiency of user services and user experience, it also greatly reduces the costs of enterprises.

[0068] In the related art, most multi-turn dialogue systems basically complete the interaction of the dialogue according to a preset process. There may be situations where the multi-turn dialogue system answers irrelevantly. For example, in the first round of dialogue, the user queries the weather in Beijing. In the second round of dialogue, the user wants to rent a car. At this time, the multi-turn dialogue system recommends car rental information in Beijing for the user. However, in actual applications, the user may want to rent a car in Shanghai, thus resulting in an irrelevant answer.

[0069] As can be seen from the above description, in the related art, the information provided by the multi-turn dialogue system for the user may not be the information that the user really wants to obtain, that is, the accuracy rate of the information provided for the user is relatively low.

[0070] To solve the above technical problems existing in the related art, the embodiments of the present disclosure provide an information processing method, apparatus, electronic device, and storage medium.

[0071] In a first aspect, first, an information processing method provided by the embodiments of the present disclosure will be elaborated in detail.

[0072] It should be noted that the execution subject of an information processing method provided by the embodiments of the present disclosure can be a server capable of realizing multi-turn dialogue with a target account.

[0073] As Figure 1 shown, an information processing method provided by the embodiments of the present disclosure may include the following steps:

[0074] S110, receiving an information acquisition request from a target account.

[0075] Specifically, the target account can be any account that interacts with the server acting as the execution entity. When the target account wants to request information from the server acting as the execution entity, it can send an information acquisition request to the server acting as the execution entity. For example, when the target account wants to query the weather, it can send an information acquisition request for querying the weather to the server.

[0076] S120. Determine the current conversation scenario corresponding to the information acquisition request.

[0077] Specifically, after receiving the information acquisition request sent by the target account, the server can perform entity and selection recognition on the information acquisition request. Among them, the selection represents a business operation that the account wants to complete, such as querying the weather, renting a car, etc.; the entity is the parameter required to complete the business operation, such as time, location, etc. The conversation scenario is composed of a selection and all the conversation interactions under that selection. Only when the account expresses a new selection or ends the conversation, the conversation scenario will switch or end.

[0078] After the server performs entity and selection recognition on the information acquisition request, it can determine the current conversation scenario corresponding to the information acquisition request. For example, the information carried by the information acquisition request can be to query the weather. Then, the selection is to check the weather, and the entity can include time and location, etc. The time is which time period of the weather needs to be queried, and the location is which city's weather needs to be queried. The current conversation scenario is the scenario for querying the weather.

[0079] S130. Obtain the key information in the historical conversation scenario of the target account.

[0080] Among them, the historical conversation scenario is different from the current conversation scenario.

[0081] It should be noted that the account information (including but not limited to the device information of the account, the personal information of the account, etc.), relevant data, etc. involved in the present disclosure are all information authorized by the account or authorized by all parties.

[0082] Specifically, before the target account sends an information acquisition request to the server, it may have had other historical conversations with the server. For the sake of clear description of the solution, the conversation scenario corresponding to the historical conversation between the target account and the server can be called the historical conversation scenario. And in practical applications, the historical conversation scenario may be different from the current conversation scenario, that is, the target account and the server have had a cross-scenario conversation. For example, the historical conversation scenario is the scenario for querying the weather, and the current conversation scenario is the scenario for renting a car.

[0083] In order to reuse certain key information in the context information of the historical conversation scenario of the target account in subsequent steps in the current conversation scenario, after determining the current conversation scenario corresponding to the information acquisition request, it is necessary to obtain the key information in the context information of the historical conversation scenario of the target account. To achieve information reuse across conversation scenarios and reduce the target account from repeatedly entering the same information.

[0084] Among them, the key information is information of the target type, and the target type is the type of information required to be acquired by the information acquisition request. Suppose the historical conversation scenario is asking about the weather in Beijing, and the current conversation scenario is wanting to take a taxi. Since the user needs a location to take a taxi, at this time, the target type is the location, and "Beijing" in the historical conversation scenario belongs to the location, that is, "Beijing" belongs to the target type. Therefore, "Beijing" in the historical conversation scenario can be determined as the key information.

[0085] In order to accurately obtain the key information in the historical conversation scenario of the target account, in one implementation, before obtaining the key information in the historical conversation scenario of the target account, the following two steps can be included, namely step a1 and step a2:

[0086] Step a1, determine the moment when the information acquisition request is received.

[0087] Step a2, when there is a historical conversation record corresponding to the target account within a preset time period before the moment, execute obtaining the key information in the historical conversation scenario of the target account.

[0088] Specifically, when obtaining the key information in the historical conversation scenario of the target account, it is first necessary to determine whether there is a historical conversation record for the target account. The way to determine whether there is a historical conversation record for the target account can be: determine the moment when the information acquisition request is received, and determine whether the target account has a conversation with the server within a preset time period before this moment, that is, whether there is a historical conversation record within the effective conversation period. Among them, the preset time period can be set according to the actual situation. For example, the preset time period can be 15 minutes before the moment when the information acquisition request is received, etc. The embodiments of the present disclosure do not make specific limitations on this.

[0089] And it can be understood that in practical applications, the preset time period should not be set too long. Otherwise, the relevance between the context information of the historical conversation scenario and the context information of the current conversation scenario is relatively small. If some information in the context information of the historical conversation scenario is reused in the context information of the current conversation scenario, there may be a situation of low accuracy.

[0090] It can be seen that through the technical solution provided by the embodiments of the present disclosure, by determining the moment when the information acquisition request is received, and when there is a historical conversation record corresponding to the target account within a preset time period before this moment, the key information in the historical conversation record of the target account is acquired. Since the context information of the historical conversation scenario and the context information of the current conversation scenario usually have a high degree of relevance, therefore, it helps to reuse the key information in the context information of the historical conversation scenario into the context information of the current conversation scenario in subsequent steps, thereby reducing the target account from repeatedly inputting the same information and simplifying the conversation process.

[0091] For the sake of clear description of the solution, the specific implementation manner of S130, acquiring the key information in the historical conversation scenario of the target account, will be elaborated in detail in the following embodiments.

[0092] S140, send a target prompt message to the target account.

[0093] Among them, the target prompt message is used to prompt the target account to confirm whether to reuse the key information into the current conversation scenario.

[0094] After determining the key information to be reused into the current conversation scenario, the server sends a target prompt message to the target account to prompt the target account whether to reuse the key information into the current conversation scenario, that is, to prompt the target account to clarify and confirm whether to reuse the key information.

[0095] For example, the historical conversation scenario is a scenario for querying the weather, the current conversation scenario is a scenario for renting a car, and the key information is Beijing. At this time, the generated target prompt message can be: Do you want to rent a car in Beijing. And send the generated target prompt message to the target account. In this way, the target account can clarify and confirm for the target prompt message. Specifically, the target account can clarify and confirm the entity to be reused into the context information of the current conversation scenario by answering two selection messages of yes or no.

[0096] S150, respond to the information acquisition request based on the selection message sent by the target account for the target prompt message.

[0097] Specifically, the target prompt message is used to prompt the target account to confirm whether to reuse the key information into the current conversation scenario. Therefore, after receiving the target prompt message, the target account can reply to the target prompt message. The reply message can be a selection message for the target prompt message. If the target account confirms to reuse the key information into the current conversation scenario, then, this selection message can be an affirmative message. For example, this selection message can be yes; if the target account confirms not to reuse the key information into the current conversation scenario, then, this selection message can be a negative message, and this selection message can be no or not, etc.

[0098] After the server receives the selection information sent by the target account, it can further respond to the information acquisition request according to the selection information. Specifically, when the selection information is affirmative, it indicates that the target account confirms to reuse the key information in the current conversation scenario. Then, the information acquisition request is responded to based on the key information. For example, the target prompt information is "Do you want to rent a car in Beijing?" The selection information of the target account is "Yes". Then, the server knows that the target account wants to rent a car in Beijing. Next, the server will execute other conversation logic processing for renting a car in Beijing, such as obtaining the specific location, time, vehicle model, etc. in Beijing.

[0099] When the selection information is negative, it indicates that the target account confirms not to reuse the key information in the current conversation scenario. At this time, the way to respond to the information acquisition request can be: generate interaction information, which is used to prompt the target account to input the information it needs to obtain, and send the interaction information to the target account. For example, the target prompt information is "Do you want to rent a car in Beijing?" The selection information of the target account is "No". Then, the server knows that the target account does not want to rent a car in Beijing. Next, the server will generate interaction information, which can be "Where do you want to rent a car?" and send this interaction information to the target account. After the target account receives this interaction information, it continues to input: "I want to rent a car in Shanghai". Then, the server will execute the conversation logic processing for renting a car in Shanghai, such as obtaining the specific location, time, vehicle model, etc. in Shanghai.

[0100] For the sake of clear description of the solution, the specific implementation manner of S150 will be elaborated in detail in the following embodiments.

[0101] The technical solution provided by the embodiments of the present disclosure, after receiving the information acquisition request of the target account, determines the current conversation scenario corresponding to the information acquisition request, and obtains the key information in the historical conversation scenario of the target account; and the historical conversation scenario is different from the current conversation scenario, that is to say, the target account and the server have carried out cross-scenario conversation interactions. By obtaining the key information in the historical conversation scenario and sending the target prompt information to the target account, the target account is prompted to confirm whether to reuse the key information in the current conversation scenario through the target prompt information. Finally, the selection information sent by the target account for the target prompt information is received, and the information acquisition request is responded to based on the selection information.

[0102] As can be seen from the above description, when the target account and the server carry out cross-scenario conversation interactions, when the server reuses the key information in the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection information of the target account for the key information to be reused. Thus, the server can accurately respond to the information acquisition request, and the accuracy rate of the information provided to the user is relatively high.

[0103] Based on Figure 1 the embodiment shown, in one implementation, S130, obtaining key information in the historical conversation scenario of the target account may include the following steps, namely steps b1 to b3:

[0104] Step b1, determining the information type to which the information required for the current conversation scenario belongs.

[0105] Specifically, after determining the current conversation scenario, the information type to which the information required for the current conversation scenario belongs can be determined. For example, when the current conversation scenario is a car rental scenario, since car rental requires location, time, vehicle type, etc., it can be determined that the information type to which the information required for the current conversation scenario belongs is location, time, vehicle type, etc.

[0106] Step b2, determining the target information belonging to the information type in the context information of the historical conversation scenario.

[0107] Specifically, after determining the information type to which the information required for the current conversation scenario belongs, in order to accurately reuse the information belonging to this information type in the context information of the historical conversation scenario in the current conversation scenario, it is necessary to determine the target information belonging to this information type in the context information of the historical conversation scenario.

[0108] For example, when the current conversation scenario is a car rental scenario, the information type to which the information required for the current conversation scenario belongs includes location, and the historical conversation scenario is a weather query scenario. Beijing in the context information of the historical conversation scenario belongs to the location. Therefore, Beijing can be determined as the target information, that is, the probability of the target account renting a car in Beijing is relatively high.

[0109] Step b3, determining the target information as the key information in the context information of the historical conversation scenario.

[0110] Since the information type of the target information in the context information of the historical conversation scenario is the information type of the information required in the current conversation scenario, the target information is very likely to be the information to be collected in the current conversation scenario. Therefore, the target information is determined as the key information in the context information of the historical conversation scenario. In this way, the accuracy rate of the determined key information is relatively high.

[0111] It can be seen that, through the technical solution provided by the embodiments of the present application, since the information type of the target information is the same as the information type of the information required to be collected in the current conversation scenario, therefore, by determining the target information as the key information in the historical conversation scenario, the accuracy of the determined key information is relatively high, so that in subsequent steps, a target prompt message including the key information is sent to the target account, and the target account confirms whether to reuse the key information in the current conversation scenario. When the target account receives the target prompt message, it only needs to reply with a selection message. If the selection message is an affirmative message, the server directly responds to the information acquisition request based on the key information, without the target account repeating the input of the target information.

[0112] Based on the Figure 1 embodiment shown, in one implementation manner, after obtaining the key information in the historical conversation scenario of the target account in S130, as Figure 2 shown, the information processing method may further include the following steps:

[0113] S140a, store the key information as the context information of the current conversation scenario, and set a historical scenario label for the context information.

[0114] Wherein, the historical scenario label is used to represent that the context information belongs to the information obtained in the historical conversation scenario.

[0115] Specifically, since the key information belongs to the information in the historical conversation scenario of the target account, and subsequently it is necessary for the target account to confirm whether to reuse the key information belonging to the historical conversation scenario in the current conversation scenario. Therefore, after obtaining the key information in the historical conversation scenario of the target account, the key information can be stored in the context information of the current conversation scenario, and a historical scenario label is set for the context information. In this way, in subsequent steps, when sending a target prompt message to the target account, the historical scenario label can be used to determine which key information needs to be confirmed by the target account whether to be reused in the current conversation scenario.

[0116] It can be understood that, in addition to setting a historical scenario label for the key information, a current scenario label can also be set for other context information of the current conversation scenario, and the current scenario label is used to represent that the context information belongs to the information obtained in the current conversation scenario.

[0117] At this time, S140, sending a target prompt message to the target account, as Figure 2 shown, may include the following steps:

[0118] S141, obtain the context information of the current conversation scenario.

[0119] S142. When the obtained context information corresponds to a historical scenario label, generate target prompt information based on the obtained context information, and send the target prompt information to the target account.

[0120] After storing the key information as the context information of the current conversation scenario and setting a historical scenario label for the key information, when sending the target prompt information to the target account, first determine whether the obtained context information of the current conversation scenario is set with a historical scenario label. If the obtained context information is set with a historical scenario label, it means that the obtained context information is not collected in the current conversation scenario but in a historical conversation scenario, and the target account needs to clarify and confirm this context information, that is, the target account needs to confirm whether to reuse this context information in the current conversation scenario.

[0121] Therefore, generate target prompt information to be confirmed by the target account based on the obtained context information. After receiving the target prompt information, the target account can send selection information to the server for the target prompt information, thus realizing the clarification and confirmation of the key information in the context information to be reused in the current conversation scenario by the target account, which is beneficial for the server to accurately respond to the information acquisition request and avoid the situation of answering irrelevantly.

[0122] It can be seen that through the technical solution provided by the embodiments of the present disclosure, by storing the key information into the context information of the current conversation scenario and setting a historical scenario label for the context information, when the target account and the server have a cross-scenario conversation, it can be distinguished whether the context information in the current conversation scenario belongs to the information collected in the current conversation scenario or the information collected in a historical conversation scenario. Furthermore, when a certain context information does not belong to the information collected in the current conversation scenario, generate target prompt information based on the context information, so that the target account can confirm whether to reuse the context information in the current conversation scenario, which is beneficial for the server to accurately respond to the information acquisition request and provide relatively high-accuracy information for users.

[0123] In Figure 1 Based on the shown embodiments, in one implementation manner, the information processing method may further include the following steps:

[0124] When generating the target prompt information based on the key information, generate a confirmation label for reusing the key information and store the confirmation label for reusing the key information.

[0125] Specifically, when generating the target prompt information to be confirmed by the target account based on the key information, a confirmation label for reusing the key information may also be generated and the confirmation label for reusing the key information may be stored, that is, store the clarification label for cross-scenario reuse of information.

[0126] For example, in a historical conversation scenario, the target account queries the weather in Beijing. In the current conversation scenario, when the target account rents a car, the key information in the historical conversation scenario, which is Beijing, is reused. At this time, the server generates a target prompt message "Do you want to rent a car in Beijing?" and sets an information cross-scenario reuse tag in the current round of conversation. When the server receives an affirmative message such as "yes" or a negative message from the target account, it can be determined that the affirmative or negative message is a clarification response to the target message.

[0127] Moreover, since the target account has clarified whether to reuse the key information, the reuse key information confirmation tag can be reset, that is, the reuse key information confirmation tag can be cancelled. When key information needs to be reused in the next round of conversation, the reuse key information confirmation tag is set again to improve the accuracy of responding to the target account.

[0128] It can be seen that through the technical solution provided by this embodiment, if there is an information cross-scenario reuse tag and the selection information of the target account is an affirmative message or a negative message, it can be determined that the probability of the affirmative or negative message being a clarification of the reused key information is relatively high, and the information acquisition request can be accurately responded to based on the affirmative or negative message.

[0129] Based on the above embodiment, in one implementation, the selection information includes an affirmative message; responding to the information acquisition request based on the selection information sent by the target account for the target prompt message may include the following step c1:

[0130] Step c1, when the selection information is an affirmative message, obtain the target information corresponding to the information acquisition request according to the key information, and return the target information to the target account.

[0131] Specifically, if the selection information is an affirmative message, it means that the target account determines to reuse the key information in the current conversation scenario. Therefore, the target information corresponding to the information acquisition request is obtained based on the key information. For example, the target prompt message is "Do you need to rent a car in Beijing?" The selection information sent by the target account is "yes". Then, the server performs other logical conversation processing for renting a car in Beijing. For example, continue to collect vehicle models, rental prices, etc.

[0132] It can be seen that through the technical solution provided by this embodiment, the selection information being an affirmative message indicates that the target account determines to reuse the key information in the current conversation scenario. Therefore, the target information corresponding to the information acquisition request is obtained based on the key information, and thus the information acquisition request can be accurately responded to, and the accuracy rate of the information provided to the user is relatively high.

[0133] In another embodiment, the selection information includes negative information; based on the selection information sent by the target account for the target prompt information, in response to the information acquisition request, the following steps may be included, namely step d1 and step d2:

[0134] Step d1, generate interaction information and send the interaction information to the target account.

[0135] Among them, the interaction information is used to prompt the target account to input the information required for the current conversation scenario.

[0136] Step d2, based on the reply information of the target account for the interaction information, obtain the target information corresponding to the information acquisition request, and return the target information to the target account.

[0137] Specifically, if the selection information is negative information, it means that the target account determines not to reuse the key information in the current conversation scenario. Therefore, generate interaction information and send the interaction information to the target account, and respond to the information acquisition request based on the reply information of the target account for the interaction information, thereby obtaining the target information corresponding to the information acquisition request, and returning the target information to the target account.

[0138] For example, the target prompt information is "Do you need to rent a car in Beijing?", and the negative information is "No". Then, the server generates the interaction information: "Where do you need to rent a car?"; the target account replies: "Rent a car in Shanghai"; the server executes the logical dialogue processing of renting a car in Shanghai. For example, continue to collect information such as the vehicle model and rental price.

[0139] It can be seen that through the technical solution provided by this embodiment, the selection information is negative information, indicating that the target account determines not to reuse the key information in the current conversation scenario. Therefore, generate interaction information and send the interaction information to the target account, and respond to the information acquisition request based on the reply information of the target account for the interaction information, and thus can accurately respond to the information acquisition request, and the accuracy rate of the information provided for the user is relatively high.

[0140] For the sake of clear description of the solution, the complete technical solution of the present disclosure will be elaborated in detail below with specific examples. As shown in FIGS. 3(a) and 3(b).

[0141] For the sake of clear description of the solution, before elaborating the embodiments shown in FIGS. 3(a) and 3(b), several concepts involved in FIGS. 3(a) and 3(b) are explained.

[0142] Selection: Represents a business operation to be completed by the user, such as checking the weather, checking the balance, transferring money, etc. Among them, the user is the target account described in the above embodiments.

[0143] Entities: The parameters required to complete a business operation, such as time, location, card number, etc. One selection plus several entities can complete a business operation.

[0144] Dialogue node: Defines a round of interaction between the user and the server during the dialogue process. The user's request will trigger a certain dialogue node to process the request and obtain information to answer the user.

[0145] Scenario: One selection and all the dialogue interactions under this selection, such as collecting entities and clarifying and reusing information, constitute a scenario. Only when the user expresses a new selection or ends the dialogue, will the scenario switch or end.

[0146] Session: A series of conversations between the same user and the server within a certain time period is a session, such as a phone call in telephone customer service.

[0147] Dialogue: A question-and-answer between the user and the server is a round of dialogue.

[0148] As shown in FIGS. 3(a) and 3(b), the complete technical solution of the embodiments of the present disclosure may include the following steps:

[0149] S310, Receive an information acquisition request input by a target account.

[0150] S320, Determine whether there is context information of a historical scenario in the session. If so, execute S330; otherwise, execute S340.

[0151] S330, Merge the historical context information of the target account with the context information of the current scenario corresponding to the information acquisition request, and execute S350.

[0152] S340, Construct the context information of the current scenario corresponding to the information acquisition request, and execute S350.

[0153] S350, According to the information acquisition request input by the target account, perform selection and entity recognition; determine whether the target account has expressed a selection. If so, execute S360; otherwise, execute S370.

[0154] S360, According to the selection and entity recognition results, determine whether the selection is an affirmative selection or a negative selection. If it is an affirmative selection or a negative selection, execute S3100; otherwise, execute S380.

[0155] S370, Determine whether the current dialogue is within a certain scenario. If so, execute S380; otherwise, execute S390;

[0156] S380, Enter the corresponding dialogue node within the corresponding scenario, and continue S3160.

[0157] S390, generating a server reply script to guide the user to express his or her own choice, and the conversation round ends.

[0158] S3100, determine whether there is a cross-scenario information reuse clarification mark in the session; if yes, execute S3110; otherwise, execute S380.

[0159] S3110, determine whether the selection is a positive selection; if yes, execute S3120; otherwise, execute S3130;

[0160] S3120, continue to use the information currently confirmed by the user, and execute S3140.

[0161] S3130, clear the information denied by the user in the current context, and execute S3150.

[0162] S3140, reset the mark for cross-scenario reuse of information in the conversation, perform other conversation logic processing based on the collected conversation information, and execute S3230.

[0163] S3150, guide the user to input new information and execute S3180.

[0164] S3160, determining whether the information required to be collected by the dialog node already exists in the context information, if yes, executing S3170; otherwise, executing S3180;

[0165] S3170, judging whether the information to be collected by the current dialogue node belongs to the current scene according to the label of the information in the context, if yes, executing S3180; otherwise, executing S3190.

[0166] S3180, use the context information in the current dialogue scene normally, or collect the information of the current dialogue scene and store it in the context information of the current dialogue scene, and execute S3240.

[0167] S3190, assemble the clarification information according to the information in the context that needs to be reused, and execute S3210.

[0168] S3200, insert a marker for cross-scenario information reuse in the conversation to support the next round of clarification and confirmation dialogue interaction between the server and the user, and continue with S3210.

[0169] S3210, then session management, that is, storing the current scene context information and some clarification flags in the session, and continuing to S3220;

[0170] S3220, generate robot response words according to the result of dialogue logic processing.

[0171] S3230, tag the context information collected at the dialogue node with the current scenario label, and continue to execute S3140.

[0172] Since in Figures 1 to 2 the embodiments shown, each step of the embodiments shown in FIGS. 3(a) and 3(b) has been elaborated in detail and will not be repeated here.

[0173] According to a second aspect of the embodiments of the present disclosure, there is provided an information processing apparatus, as Figure 4 shown, including:

[0174] A request acquisition module 410, configured to execute receiving an information acquisition request of a target account;

[0175] A dialogue scenario determination module 420, configured to execute determining a current dialogue scenario corresponding to the information acquisition request;

[0176] A key information acquisition module 430, configured to execute acquiring key information in a historical dialogue scenario of the target account; the historical dialogue scenario is different from the current dialogue scenario; the key information is information of a target type, and the target type is the type of information required to be acquired by the information acquisition request;

[0177] A prompt information sending module 440, configured to execute sending target prompt information to the target account; the target prompt information is used to prompt the target account to confirm whether to reuse the key information in the current dialogue scenario;

[0178] A request response module 450, configured to execute responding to the information acquisition request based on the selection information sent by the target account in response to the target prompt information.

[0179] The technical solution provided by the embodiments of the present disclosure, after receiving an information acquisition request of a target account, determines a current dialogue scenario corresponding to the information acquisition request, and acquires key information in a historical dialogue scenario of the target account; and the historical dialogue scenario is different from the current dialogue scenario, that is to say, the target account has performed cross-scenario dialogue interaction with the server. By acquiring key information in the historical dialogue scenario and sending target prompt information to the target account, the target account is prompted by the target prompt information to confirm whether to reuse the key information in the current dialogue scenario. Finally, the selection information sent by the target account in response to the target prompt information is received, and the information acquisition request is responded based on the selection information.

[0180] As described above, when the target account interacts with the server in a cross-scenario conversation, when the server reuses the key information in the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection information of the target account for the key information to be reused. Thus, the server can accurately respond to the information acquisition request and provide relatively accurate information for the user.

[0181] Optionally, the key information acquisition module is specifically configured to execute:

[0182] Determine the information type to which the information required for the current conversation scenario belongs;

[0183] Determine the target information belonging to the information type in the context information of the historical conversation scenario;

[0184] Determine the target information as the key information in the context information of the historical conversation scenario.

[0185] Optionally, it further includes:

[0186] A scenario label setting module, configured to, after acquiring the key information in the historical conversation scenario of the target account, store the key information as the context information of the current conversation scenario and set a historical scenario label for the context information; the historical scenario label is used to indicate that the context information belongs to the information acquired in the historical conversation scenario;

[0187] Correspondingly, the prompt information sending module is specifically configured to execute:

[0188] Acquire the context information of the current conversation scenario;

[0189] When the obtained context information has a historical scenario label, generate target prompt information based on the obtained context information and send the target prompt information to the target account.

[0190] Optionally, the selection information includes affirmative information; the request response module is specifically configured to execute:

[0191] When the selection information is affirmative information, obtain the target information corresponding to the information acquisition request according to the key information and return the target information to the target account.

[0192] Optionally, the selection information includes negative information; the request response module is specifically configured to execute:

[0193] Generate interaction information and send the interaction information to the target account; the interaction information is used to prompt the target account to input the information required for the current conversation scenario;

[0194] Based on the reply information of the target account for the interaction information, obtain the target information corresponding to the information acquisition request, and return the target information to the target account.

[0195] Optionally, it further includes:

[0196] A moment determination module, configured to determine the moment of receiving the information acquisition request before acquiring the key information in the historical conversation scenario of the target account;

[0197] When there is a historical conversation record corresponding to the target account within a preset time period before the moment, trigger the key information acquisition module to execute acquiring the key information in the historical conversation scenario of the target account.

[0198] Figure 5 It is a block diagram of a device 500 shown according to an exemplary embodiment. For example, the device 500 may be provided as a server. Refer to Figure 5 , the device 500 includes a processing component 522, which further includes one or more processors, and memory resources represented by a memory 532 for storing instructions executable by the processing component 522, such as application programs. The application programs stored in the memory 532 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 522 is configured to execute instructions to perform the information processing method described in the first aspect.

[0199] The device 500 may further include a power supply component 526 configured to perform power management of the device 500, a wired or wireless network interface 550 configured to connect the device 500 to a network, and an input / output (I / O) interface 558. The device 500 may operate based on an operating system stored in the memory 532, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or the like.

[0200] The technical solution provided by the embodiments of the present disclosure, after receiving an information acquisition request of a target account, determines the current conversation scenario corresponding to the information acquisition request, and acquires key information in the historical conversation scenario of the target account; and the historical conversation scenario is different from the current conversation scenario, that is to say, the target account has carried out cross-scenario conversation interaction with the server. By acquiring the key information in the historical conversation scenario and sending a target prompt message to the target account, the target account is prompted through the target prompt message to confirm whether to reuse the key information in the current conversation scenario. Finally, the selection information sent by the target account in response to the target prompt message is received, and the information acquisition request is responded based on the selection information.

[0201] As can be seen from the above description, when the target account and the server carry out cross-scenario conversation interaction, when the server reuses the key information in the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection information of the target account for the key information to be reused. Thus, the server can accurately respond to the information acquisition request and provide information with a relatively high accuracy rate for the user.

[0202] According to the fourth aspect of the embodiments of the present disclosure, a storage medium is provided. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method described in the first aspect.

[0203] The technical solution provided by the embodiments of the present disclosure, after receiving an information acquisition request of a target account, determines the current conversation scenario corresponding to the information acquisition request, and acquires key information in the historical conversation scenario of the target account; and the historical conversation scenario is different from the current conversation scenario, that is to say, the target account has carried out cross-scenario conversation interaction with the server. By acquiring the key information in the historical conversation scenario and sending a target prompt message to the target account, the target account is prompted through the target prompt message to confirm whether to reuse the key information in the current conversation scenario. Finally, the selection information sent by the target account in response to the target prompt message is received, and the information acquisition request is responded based on the selection information.

[0204] As can be seen from the above description, when the target account and the server carry out cross-scenario conversation interaction, when the server reuses the key information in the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection information of the target account for the key information to be reused. Thus, the server can accurately respond to the information acquisition request and provide information with a relatively high accuracy rate for the user.

[0205] According to the fifth aspect of the embodiments of the present disclosure, a computer program product is provided, characterized in that when it runs on a computer, the computer is enabled to implement the method described in the first aspect.

[0206] After receiving an information acquisition request of a target account, the technical solution provided by an embodiment of the present disclosure determines a current conversation scenario corresponding to the information acquisition request, and acquires key information in a historical conversation scenario of the target account; and the historical conversation scenario is different from the current conversation scenario, that is to say, the target account has performed cross-scenario conversation interaction with the server. By acquiring the key information in the historical conversation scenario and sending a target prompt message to the target account, the target account is prompted by the target prompt message to confirm whether to reuse the key information in the current conversation scenario. Finally, a selection message sent by the target account in response to the target prompt message is received, and the information acquisition request is responded based on the selection message.

[0207] As can be seen from the above description, when the target account performs cross-scenario conversation interaction with the server, when the server reuses the key information in the historical conversation scenario in the current conversation scenario, the server and the target account will clarify and confirm the key information to be reused, and respond to the information acquisition request based on the selection message of the target account for the key information to be reused. Therefore, the server can accurately respond to the information acquisition request, and the accuracy rate of the information provided to the user is relatively high.

[0208] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not disclosed herein. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0209] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. An information processing method, characterized in that, comprising: Receiving an information acquisition request of a target account; Determining a current conversation scenario corresponding to the information acquisition request; The current conversation scenario is determined after entity and selection recognition of the information acquisition request upon receiving the information acquisition request; Obtaining key information in the historical conversation scenario of the target account; The historical conversation scenario is different from the current conversation scenario; The key information is information of a target type, and the target type is the type of information required to be acquired by the information acquisition request; Sending a target prompt message to the target account; The target prompt message is used to prompt the target account to confirm whether to reuse the key information in the current conversation scenario; Responding to the information acquisition request based on the selection information sent by the target account for the target prompt message; After obtaining the key information in the historical conversation scenario of the target account, further comprising: Storing the key information as context information of the current conversation scenario, and setting a historical scenario label for the context information; the historical scenario label is used to represent that the context information belongs to the information obtained in the historical conversation scenario; The sending the target prompt message to the target account includes: Obtaining the context information of the current conversation scenario; Judging whether the context information of the current conversation scenario is set with a historical scenario label. When the context information of the current conversation scenario is set with a corresponding historical scenario label, generating a target prompt message based on the obtained context information and sending the target prompt message to the target account.

2. The method according to claim 1, characterized in that, The obtaining the key information in the historical conversation scenario of the target account includes: Determining the information type to which the information required to be acquired by the current conversation scenario belongs; Determining target information belonging to the information type in the context information of the historical conversation scenario; Determining the target information as the key information in the context information of the historical conversation scenario.

3. The method according to claim 1, characterized in that, The selection information includes affirmative information; the responding to the information acquisition request based on the selection information sent by the target account for the target prompt message includes: When the selection information is affirmative information, obtaining the target information corresponding to the information acquisition request according to the key information and returning the target information to the target account.

4. The method according to claim 1, characterized in that, The selection information includes negative information; the responding to the information acquisition request based on the selection information sent by the target account for the target prompt message includes: Generating interaction information and sending the interaction information to the target account; the interaction information is used to prompt the target account to input the information required to be acquired by the current conversation scenario; Based on the reply information of the target account for the interaction information, obtaining the target information corresponding to the information acquisition request and returning the target information to the target account.

5. The method according to any one of claims 1 to 4, characterized in that, before obtaining the key information in the historical conversation scenario of the target account, it further includes: determining the moment when the information acquisition request is received; when there is a historical conversation record corresponding to the target account within a preset time period before the moment, performing the acquisition of the key information in the historical conversation scenario of the target account.

6. An information processing device, characterized in that, it includes: a request acquisition module configured to receive an information acquisition request of a target account; a conversation scenario determination module configured to determine the current conversation scenario corresponding to the information acquisition request; the current conversation scenario is determined after entity and selection recognition of the information acquisition request after receiving the information acquisition request; a key information acquisition module configured to acquire the key information in the historical conversation scenario of the target account; the historical conversation scenario is different from the current conversation scenario; the key information is information of a target type, and the target type is the type of information required to be acquired by the information acquisition request; a prompt information sending module configured to send target prompt information to the target account; the target prompt information is used to prompt the target account to confirm whether to reuse the key information into the current conversation scenario; a request response module configured to respond to the information acquisition request based on the selection information sent by the target account for the target prompt information; it further includes: a scenario label setting module configured to, after acquiring the key information in the historical conversation scenario of the target account, store the key information as the context information of the current conversation scenario and set a historical scenario label for the context information; the historical scenario label is used to represent that the context information belongs to the information acquired in the historical conversation scenario; the prompt information sending module is specifically configured to: acquire the context information of the current conversation scenario; judge whether the context information of the current conversation scenario is set with a historical scenario label, and if the context information of the current conversation scenario is set with a corresponding historical scenario label, generate target prompt information based on the acquired context information and send the target prompt information to the target account.

7. An electronic device, characterized in that, it includes: a processor; a memory for storing executable instructions of the processor; wherein, the processor is configured to execute the instructions to implement the method according to any one of claims 1 to 5.

8. A storage medium, characterized in that, when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that, when it runs on a computer, the computer implements the method according to any one of claims 1 to 5.

Images