Method, device, storage medium and electronic equipment for processing information

Abstract

The present disclosure relates to a method and device for information processing, a storage medium and an electronic device, which can obtain a business graph representing the corresponding relationship between the business-related information of the target business in the dialogue information according to the dialogue information of the target user for the target business, sort a plurality of pre-established business to-do items according to the business graph, select a first target item to be executed by the target user according to the sorting result, and output first knowledge information associated with the plurality of first target items to assist in replying to the dialogue information. In this way, the business graph is established according to the user dialogue, the association between the dialogue information is established, the real-time performance and the adaptation degree of the subsequent output knowledge information are improved, and the user's multiple business processing intentions are actively predicted in combination with the business graph, the knowledge information corresponding to the business processing intention is actively preloaded, thereby realizing the active adaptation of the knowledge information, and improving the efficiency, accuracy and stability of the customer service.

Description

Technical Field

[0001] This disclosure relates to the field of information technology, and more specifically, to an information processing method, apparatus, storage medium, and electronic device. Background Technology

[0002] With the development of information technology, customer service, such as call center services, is now used in scenarios such as energy and digital operation and maintenance to provide services to customers based on the knowledge information output by dialogue assistance tools.

[0003] However, the knowledge information output by current dialogue assistance tools is usually obtained by keyword queries in user dialogues. Keyword queries lack connection with the dialogue context, resulting in large deviations in output results and insufficient adaptability, leading to low efficiency and stability of customer service. Summary of the Invention

[0004] To address the aforementioned problems, this disclosure provides an information processing method, apparatus, storage medium, and electronic device.

[0005] According to a first aspect of the present disclosure, an information processing method is provided, the method comprising: obtaining a business graph based on dialogue information of a target user regarding a target business, the business graph being used to characterize the correspondence between business-related information of the target business in the dialogue information, the business-related information including at least one of business requirements, business execution actions, and business execution elements; sorting a plurality of pre-established business to-do items according to the business graph, and selecting a plurality of first target items to be executed by the target user according to the sorting result; and outputting first knowledge information associated with the plurality of first target items, the first knowledge information being used to assist in responding to the dialogue information.

[0006] Optionally, obtaining the business graph based on the dialogue information of the target user for the target business includes: extracting the business association information and the correspondence between the business association information from the dialogue information based on a preset dialogue recognition model; updating the business nodes in the historical business graph and the connection relationship between the business nodes according to the business association information and the correspondence between the business association information, thereby obtaining the business graph.

[0007] Optionally, the historical business graph is pre-established by: acquiring historical dialogue information of multiple users performing various business tasks within a historical time period; extracting historical business association information and the correspondence between historical business association information from the historical dialogue information; using the historical business association information as business nodes and the correspondence between the historical business association information as the connection relationship of business nodes to obtain the historical business graph.

[0008] Optionally, the step of sorting a plurality of pre-established business to-do items according to the business graph and selecting a plurality of target users to execute a first target item according to the sorting result includes: determining a first business node corresponding to specified business association information from the business graph, wherein the specified business association information is the most recently updated business association information in the dialogue information; from the plurality of business to-do items, selecting business to-do items corresponding to a plurality of second business nodes adjacent to the first business node in the business graph as candidate items, wherein different second business nodes correspond to different business to-do items; sorting the plurality of candidate items by execution priority, and selecting a preset number of candidate items with the highest execution priority as the first target item.

[0009] Optionally, prioritizing the execution of multiple candidate items includes: determining the association probability of each candidate item with the specified business-related information; determining the output ratio of the knowledge information associated with each candidate item in the historical output information; and determining the execution priority corresponding to each candidate item based on the association probability and the output ratio to obtain the priority ranking result.

[0010] Optionally, the method further includes: filtering the candidate items based on the association probability and a preset association probability threshold; determining the execution priority corresponding to each candidate item includes: determining the execution priority corresponding to each filtered candidate item.

[0011] Optionally, the method further includes: obtaining a second target item input by the target user for the target service; and adjusting the execution priority of the first target item according to the second target item.

[0012] According to a second aspect of the present disclosure, an information processing apparatus is provided, the apparatus comprising: The acquisition module is used to acquire a business graph based on the dialogue information of the target user for the target business. The business graph is used to represent the correspondence between the business-related information of the target business in the dialogue information. The business-related information includes at least one of business requirements, business execution actions, and business execution elements. The analysis module is used to sort multiple pre-established business tasks according to the business graph, and select the first target task to be executed by multiple target users according to the sorting results. The output module is used to output first knowledge information associated with the plurality of first target items, and the first knowledge information is used to assist in responding to the dialogue information.

[0013] Optionally, the acquisition module is further configured to extract the business association information and the correspondence between the business association information from the dialogue information based on a preset dialogue recognition model; and update the business nodes in the historical business graph and the connection relationship between the business nodes according to the business association information and the correspondence between the business association information, thereby obtaining the business graph.

[0014] Optionally, the historical business graph is pre-established by: acquiring historical dialogue information of multiple users performing various business tasks within a historical time period; extracting historical business association information and the correspondence between historical business association information from the historical dialogue information; using the historical business association information as business nodes and the correspondence between the historical business association information as the connection relationship of business nodes to obtain the historical business graph.

[0015] Optionally, the analysis module is further configured to: determine a first business node corresponding to specified business association information from the business graph, wherein the specified business association information is the most recently updated business association information in the dialogue information; select, from the plurality of pending business items, the pending business items corresponding to a plurality of second business nodes adjacent to the first business node in the business graph as candidate items, wherein different second business nodes correspond to different pending business items; sort the plurality of candidate items by execution priority, and select a preset number of candidate items with the highest execution priority as the first target item.

[0016] Optionally, the analysis module is further configured to determine the association probability between each candidate item and the specified business-related information; determine the output ratio of the knowledge information associated with each candidate item in the historical output information; and determine the execution priority corresponding to each candidate item based on the association probability and the output ratio to obtain the priority ranking result.

[0017] Optionally, the analysis module is further configured to filter the candidate items based on the association probability and a preset association probability threshold; and determine the execution priority corresponding to each filtered candidate item.

[0018] Optionally, the acquisition module is further configured to acquire a second target item input by the target user for the target business; the analysis module is further configured to adjust the execution priority of the first target item based on the second target item.

[0019] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the information processing method provided in the first aspect of the present disclosure.

[0020] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the information processing method provided in the first aspect of this disclosure.

[0021] According to the above technical solution, a business graph can be obtained based on the dialogue information of the target user regarding the target business, representing the correspondence between business-related information in the dialogue information. Based on the business graph, multiple pre-established business tasks are sorted, and multiple first target tasks to be executed by the target user are selected according to the sorting results. First knowledge information associated with these first target tasks, used to assist in responding to the dialogue information, is then output. In this way, building a business graph based on user dialogue establishes connections between dialogue information, improving the real-time performance and adaptability of subsequent knowledge information output. Furthermore, by proactively predicting multiple business processing intentions of the user based on the business graph, and proactively pre-loading the knowledge information corresponding to these intentions, proactive adaptation of knowledge information is achieved, improving the efficiency, accuracy, and stability of customer service.

[0022] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating an information processing method according to an exemplary embodiment.

[0024] Figure 2 This is a flowchart illustrating another information processing method according to an exemplary embodiment.

[0025] Figure 3 This is a structural block diagram of an information processing apparatus according to an exemplary embodiment.

[0026] Figure 4 This is a block diagram of an electronic device provided according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0027] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0028] In the following description, the words "first" and "second" are used only to distinguish the purpose of the description and should not be interpreted as indicating or implying relative importance or order.

[0029] In related technologies, with the development of information technology, customer service in scenarios such as energy and digital operation and maintenance, such as call center services, typically provides services to customers based on knowledge information output by dialogue assistance tools. However, the knowledge information output by current dialogue assistance tools is usually obtained by keyword queries in user dialogues. Keyword queries lack relevance to the dialogue context, resulting in significant output deviations and insufficient adaptability, leading to low efficiency and stability of customer service.

[0030] To address the aforementioned problems, this disclosure provides an information processing method, apparatus, storage medium, and electronic device. This device can acquire a business graph representing the correspondence between business-related information in the dialogue information concerning target services, based on dialogue information from a target user regarding a target service. It then sorts multiple pre-established business tasks based on the business graph, selects multiple first target tasks to be executed by the target user according to the sorting result, and outputs first knowledge information associated with these first target tasks to assist in responding to the dialogue information. In this way, establishing a business graph based on user dialogue enables the establishment of relationships between dialogue information, improving the real-time performance and adaptability of subsequent knowledge information output. Furthermore, by proactively predicting various business processing intentions of users based on the business graph and proactively pre-loading knowledge information corresponding to these intentions, proactive adaptation of knowledge information is achieved, improving the efficiency, accuracy, and stability of customer service.

[0031] The present disclosure will now be described in conjunction with specific embodiments.

[0032] Figure 1 This is a flowchart illustrating an information processing method according to an exemplary embodiment, such as... Figure 1 As shown, the method may include the following steps: In step S101, a service graph is obtained based on the dialogue information of the target user for the target service. The service graph is used to represent the correspondence between the service-related information of the target service in the dialogue information.

[0033] The target service can be a service for which the target user expresses an intention to perform, such as "printing a fault report" or "password reset". The dialogue information can be the dialogue between the target user and the staff providing the service. This dialogue information can include voice information such as voice call streams, text information such as IM (Instant Messaging) chat logs, video information, and image information, etc., without specific limitations in this disclosure. Furthermore, the service-related information can include at least one of the following: service requirements, service execution actions, and service execution elements. For example, for the target service "password reset", the service requirement could be "set a new password", the service execution action could be "submit a password reset request", and the service execution element could be "username".

[0034] In one possible implementation, the business-related information and the correspondence between the business-related information can be extracted from the dialogue information based on a preset dialogue recognition model. Then, the business nodes in the historical business graph and the connection relationships between the business nodes are updated according to the business-related information and the correspondence between the business-related information, so as to obtain the business graph.

[0035] The preset dialogue recognition model can include an ASR (Automatic Speech Recognition) model, a DST (Dialogue State Tracking) model, and an artificial intelligence model. The historical business graph pre-records multiple historical business nodes and the node connection topology between them. For example, based on the artificial intelligence model, the dialogue information can be used for dialogue recognition, speech segmentation, business execution element identification, and business requirement classification, outputting business information units that include multiple business-related information. After determining the business-related information, it can be used as a new node in the historical business graph, and the interaction relationships of this business-related information can be mapped to connections between new nodes in the historical business graph, thus obtaining the business graph.

[0036] In step S102, based on the business map, the pre-established multiple business to-do items are sorted, and based on the sorting results, the first target item to be executed by multiple target users is selected.

[0037] Specifically, based on the business association information in the business graph, the system can select the business that the target user might intend to execute from among multiple pending business items, and then sort the selected business items to determine the first target item to be executed by the target user. For example, if the business execution element in the business graph is "username," the business that the target user might intend to execute could include "reset password," "modify information," and "cancel account." In this case, "reset password," "modify information," and "cancel account" can be sorted to obtain the sorting result.

[0038] In step S103, first knowledge information associated with the plurality of first target items is output, and the first knowledge information is used to assist in responding to the dialogue information.

[0039] Specifically, based on a pre-established knowledge mapping relationship, the system can output first knowledge information associated with the multiple first target items. This knowledge mapping relationship includes the association between multiple sets of to-do items and knowledge information. Each piece of knowledge information can include metadata such as the title of the to-do item, a content summary, and an operation guidance link. For example, based on this knowledge mapping relationship, the system can determine the first knowledge information associated with each first target item and display each set of first target items and the first knowledge information associated with each first target item in sorted order. For instance, if the sorted order of the first target items is "Reset Password," "Information Modification," and "Account Cancellation," the first knowledge information corresponding to "Reset Password" can be displayed at the top, the first knowledge information corresponding to "Information Modification" can be displayed in the center, and the first knowledge information corresponding to "Account Cancellation" can be displayed at the bottom.

[0040] The above technical solution can obtain a business graph representing the correspondence between business-related information in the dialogue information of target users regarding target services, based on the dialogue information of target users. Then, based on the business graph, it sorts multiple pre-established business tasks, selects multiple first-target tasks to be executed by target users according to the sorting results, and outputs first-level knowledge information associated with these first-target tasks to assist in responding to the dialogue information. In this way, building a business graph based on user dialogue establishes connections between dialogue information, improving the real-time performance and adaptability of subsequent knowledge information output. Furthermore, by proactively predicting multiple business processing intentions of users based on the business graph and proactively preloading the knowledge information corresponding to these intentions, it achieves proactive adaptation of knowledge information, improving the efficiency, accuracy, and stability of customer service.

[0041] In some embodiments, the aforementioned historical business graph can be pre-established in the following manner: obtaining historical dialogue information of multiple users performing various business tasks within a historical time period, extracting historical business association information and the correspondence between historical business association information from the historical dialogue information, using the historical business association information as business nodes, and using the correspondence between the historical business association information as the connection relationship of business nodes to obtain the historical business graph.

[0042] The historical time period can be determined based on the actual situation. The connection relationship of the business node can be determined based on the distribution frequency of the correspondence between the historical business-related information. For example, for the business-related information that appears for the first time in each historical dialogue information, the connection frequency distribution of each adjacent business-related information connected to this business-related information is obtained. The connection relationship between the high-frequency (e.g., frequency distribution greater than 10%) adjacent business-related information and this business-related information is established. This process continues to determine the connection distribution of the business-related information connected to each adjacent business-related information until all business-related information is connected, thus obtaining the historical business map. For example, printing fault reports, "fault location", and "repair dispatch" can be considered as a set of connection relationships for business-related information.

[0043] The above technical solution can obtain historical dialogue information of multiple users performing various business tasks within a historical time period, and establish a historical business graph based on the historical business association information in the historical dialogue information and the correspondence between the historical business association information. This allows for the subsequent establishment of a business graph by updating historical business information, establishing associations between dialogue information, and proactively preloading knowledge information corresponding to business processing intentions to achieve proactive adaptation of knowledge information.

[0044] In some embodiments, step S102 may include: S121. Determine the first business node corresponding to the specified business association information from the business graph.

[0045] The designated business association information is the most recently updated business association information in the dialogue information. For example, if the target user raises business association information such as "forgot password" and "username" in the dialogue information, "username" can be used as the designated business association information to determine the first business node.

[0046] S122. From the multiple pending business items, select the pending business items corresponding to the multiple second business nodes adjacent to the first business node in the business graph as candidate items.

[0047] Different second business nodes correspond to different business pending items. For example, among these multiple business pending items, the multiple second business nodes adjacent to the first business node "username" may include "reset password", "modify information", "account cancellation", "SMS verification" and "fingerprint verification". The business pending items corresponding to these multiple second business nodes can be used as candidate items.

[0048] S123. Rank the execution priorities of multiple candidate items and select the preset number of candidate items with the highest execution priority as the first target item.

[0049] In one possible implementation, the association probability between each candidate item and the specified business-related information can be determined, and the output ratio of the knowledge information associated with each candidate item in the historical output information can be determined. Based on the association probability and the output ratio, the execution priority corresponding to each candidate item can be determined, and the priority ranking result can be obtained.

[0050] For example, the state vector h of the business graph can first be established based on GAT (Graph Attention Network) or Transformer-based Path Encoder. t The probability of association between each candidate item and the specified business-related information is determined using the following formula:

[0051] in, This is the first business node. This is the second business node. This is the state vector of the business graph. The embedding vector representing the second service node. The concatenated vector representing the state vector and the embedding vector. The evaluation value representing the concatenated vector. The probability representing the evaluation value of the concatenated vector. This represents the probability of association between the candidate item and the specified business-related information.

[0052] Meanwhile, this historical output information includes output information for business nodes in the historical business graph within a historical time period. The output ratio of knowledge information associated with each candidate item in the historical output information can be determined using the following formula:

[0053] in, This represents the proportion of knowledge information associated with the candidate item in the historical output information. The amount of knowledge information associated with this candidate item. The total amount of knowledge information output for this historical information.

[0054] After determining the correlation probability and the output ratio, the execution priority evaluation value for each candidate item can be determined using the following formula:

[0055] in, This is the execution priority evaluation value corresponding to the candidate item. This is the probability of the association. This is the output ratio. This is the first weight corresponding to the association probability. This is the second weight corresponding to the output ratio. and The sum of is 1. and The value can be determined based on the actual situation.

[0056] After calculating the execution priority evaluation value for each candidate item, the candidate items can be sorted according to the size of the priority evaluation value. The larger the priority evaluation value, the higher the execution priority. A preset number of candidate items with the largest priority evaluation value can be used as the first target item. The preset number can be determined according to the actual situation. For example, the preset number can be 3.

[0057] The above technical solution can sort multiple pre-established business tasks based on the business graph, and select multiple target tasks to be executed by target users based on the sorting results. This can proactively predict multiple business processing intentions of users by combining the business graph, so as to proactively preload the knowledge information corresponding to the business processing intentions, thereby realizing the proactive adaptation of knowledge information and improving the efficiency, accuracy and stability of subsequent customer service.

[0058] In some embodiments, the method may further include: filtering the candidate items based on the association probability and a preset association probability threshold. Determining the execution priority for each candidate item may further include: determining the execution priority for each filtered candidate item.

[0059] The preset association probability threshold can be determined according to the actual situation. For example, the preset association probability threshold can be 65%. For instance, candidate items with an association probability lower than the preset association probability threshold can be eliminated to improve the accuracy of knowledge information adaptation. For example, if the business graph obtained from the dialogue information includes business association information "submit password reset application" and "username", the association probability of the candidate item "account cancellation" is 1%, and the candidate item "account cancellation" can be eliminated.

[0060] The above technical solutions can also determine the correlation probability between candidate items to be executed and the business graph to filter candidate items, thereby improving the prediction accuracy of various business processing intentions of users in the future, so as to proactively preload the knowledge information corresponding to the business processing intentions, thereby realizing the proactive adaptation of knowledge information.

[0061] In some embodiments, the above method may further include: obtaining a second target item input by the target user for the target business, and adjusting the execution priority of the first target item according to the second target item.

[0062] Specifically, it can acquire supplementary information from the target user regarding the dialogue, as well as the target user's interactive operations related to the first target item, including clicking, accepting, ignoring, and manual searching. For example, if the target user selects any item from the first target items as a to-do item, the execution priority of the selected item can be set to the highest. Furthermore, if the first target items do not include a second target item input by the target user for the target business, the execution priority of the second target item can be set to the highest, the second knowledge information associated with the second target item can be output, and the first knowledge information corresponding to the first target item can be output simultaneously.

[0063] The above technical solutions can also adjust the execution priority of target items according to the user's selection, thereby improving the prediction accuracy of various business processing intentions of users in the future, so as to proactively preload the knowledge information corresponding to the business processing intentions, thereby realizing the proactive adaptation of knowledge information.

[0064] Figure 2 This is a flowchart illustrating another information processing method according to an exemplary embodiment, see reference. Figure 2 The method may include: S201. Based on a preset dialogue recognition model, extract business-related information and the correspondence between the business-related information from the dialogue information of the target user regarding the target business.

[0065] The preset dialogue recognition model may include ASR model, DST model and artificial intelligence model, etc. The target business may be the business that proposes the execution intention of the target user. The dialogue information may be the dialogue information between the target user and the staff who provide services to the target user. The business-related information may include at least one of business requirements, business execution actions and business execution elements.

[0066] S202. According to the business association information and the correspondence between the business association information, update the business nodes in the historical business graph and the connection relationship between the business nodes to obtain the business graph.

[0067] The historical business graph can be pre-established in the following way: obtain historical dialogue information of multiple users when performing various business tasks within a historical time period, extract historical business association information and the correspondence between historical business association information from the historical dialogue information, use the historical business association information as business nodes, and use the correspondence between the historical business association information as the connection relationship of business nodes to obtain the historical business graph.

[0068] S203. Determine the first business node corresponding to the specified business association information from the business graph.

[0069] The specified business association information refers to the most recently updated business association information in the dialogue information.

[0070] S204. From the multiple pending business items, select the pending business items corresponding to the multiple second business nodes adjacent to the first business node in the business graph as candidate items.

[0071] Different second business nodes correspond to different pending business items.

[0072] S205. Determine the probability of association between each candidate item and the specified business-related information.

[0073] S206. Filter the candidate items based on the association probability and the preset association probability threshold.

[0074] The preset association probability threshold can be determined according to the actual situation. For example, the preset association probability threshold can be 65%, and candidate items with an association probability lower than the preset association probability threshold can be eliminated.

[0075] S207. Determine the output ratio of the knowledge information associated with each filtered candidate item in the historical output ratio information.

[0076] This historical output information includes output information for business nodes in the historical business graph within a historical time period.

[0077] S208. Based on the association probability and the output ratio, determine the execution priority of each candidate item and obtain the priority ranking result.

[0078] S209. Select the first target item to be executed for multiple target users based on the priority ranking result.

[0079] Among them, a preset number of candidate items with the highest execution priority can be selected as the first target item. The preset number can be determined according to the actual situation, for example, the preset number can be 3.

[0080] S210. Output first knowledge information associated with the plurality of first target items, the first knowledge information being used to assist in responding to the dialogue information.

[0081] Specifically, based on a pre-established knowledge mapping relationship, the system can output first knowledge information associated with the multiple first target items. This knowledge mapping relationship includes the association between multiple sets of to-do items and knowledge information. Each piece of knowledge information can include metadata such as the title of the to-do item, a content summary, and an operation guide link.

[0082] By employing the above technical solution, business-related information and the correspondence between these information can be extracted from the dialogue information of target users regarding target services. Based on this business-related information and their correspondence, the business nodes in the historical business graph and the connections between them are updated, resulting in a business graph representing the correspondence between business-related information related to target services in the dialogue information. Candidate items adjacent to the nodes corresponding to specified business-related information are determined based on this business graph, and these candidate items are ranked. Multiple first target items to be executed by the target users are selected based on the ranking results, and first knowledge information associated with these first target items to assist in responding to the dialogue information is output. In this way, building a business graph based on user dialogue establishes connections between dialogue information, improving the real-time performance and adaptability of subsequent knowledge information output. Furthermore, by proactively predicting various business processing intentions of users based on the business graph and pre-loading the knowledge information corresponding to these intentions, proactive adaptation of knowledge information is achieved, improving the efficiency, accuracy, and stability of customer service.

[0083] It should be noted that the above Figure 2 The descriptions of each step in the illustrated embodiments can be found in the descriptions of the relevant steps in the foregoing embodiments, and will not be repeated here.

[0084] Figure 3 This is a structural block diagram of an information processing apparatus 300 according to an exemplary embodiment, such as... Figure 3 As shown, the information processing apparatus 300 may include: The acquisition module 301 is used to acquire a business graph based on the dialogue information of the target user for the target business. The business graph is used to represent the correspondence between the business-related information of the target business in the dialogue information. The business-related information includes at least one of business requirements, business execution actions, and business execution elements. Analysis module 302 is used to sort multiple pre-established business tasks according to the business graph, and select the first target task to be executed by multiple target users according to the sorting results. The output module 303 is used to output first knowledge information associated with the plurality of first target items, and the first knowledge information is used to assist in responding to the dialogue information.

[0085] Optionally, the acquisition module 301 is further configured to extract the business association information and the correspondence between the business association information from the dialogue information based on a preset dialogue recognition model; and update the business nodes in the historical business graph and the connection relationship between the business nodes according to the business association information and the correspondence between the business association information, thereby obtaining the business graph.

[0086] Optionally, the historical business graph is pre-established by: acquiring historical dialogue information of multiple users performing various business tasks within a historical time period; extracting historical business association information and the correspondence between historical business association information from the historical dialogue information; using the historical business association information as business nodes and the correspondence between the historical business association information as the connection relationship of business nodes to obtain the historical business graph.

[0087] Optionally, the analysis module 302 is further configured to: determine a first business node corresponding to specified business association information from the business graph, wherein the specified business association information is the most recently updated business association information in the dialogue information; select, from the multiple pending business items, the pending business items corresponding to multiple second business nodes adjacent to the first business node in the business graph as candidate items, wherein different second business nodes correspond to different pending business items; sort the multiple candidate items by execution priority, and select a preset number of candidate items with the highest execution priority as the first target item.

[0088] Optionally, the analysis module 302 is further configured to determine the association probability of each candidate item with the specified business-related information; determine the output ratio of the knowledge information associated with each candidate item in the historical output information; and determine the execution priority corresponding to each candidate item based on the association probability and the output ratio, thereby obtaining the priority ranking result.

[0089] Optionally, the analysis module 302 is further configured to filter the candidate items based on the association probability and a preset association probability threshold; and determine the execution priority corresponding to each filtered candidate item.

[0090] Optionally, the acquisition module 301 is further configured to acquire a second target item input by the target user for the target business; the analysis module 302 is further configured to adjust the execution priority of the first target item based on the second target item.

[0091] Using the aforementioned apparatus, business-related information and the correspondence between these information can be extracted from the dialogue information of a target user regarding a target service. Based on this business-related information and their correspondence, the business nodes in the historical business graph and the connections between these nodes are updated, resulting in a business graph representing the correspondence between business-related information related to the target service in the dialogue information. Candidate items adjacent to the nodes corresponding to specified business-related information are determined based on this business graph, and these candidate items are ranked. Multiple first target items to be executed by the target user are selected based on the ranking result, and first knowledge information associated with these first target items to assist in responding to the dialogue information is output. In this way, establishing a business graph based on user dialogue establishes connections between dialogue information, improving the real-time performance and adaptability of subsequent knowledge information output. Furthermore, by proactively predicting various business processing intentions of users based on the business graph and proactively pre-loading knowledge information corresponding to these intentions, proactive adaptation of knowledge information is achieved, improving the efficiency, accuracy, and stability of customer service.

[0092] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0093] Figure 4 This is a block diagram of an electronic device 400 provided according to an exemplary embodiment of the present disclosure. Figure 4 As shown, the electronic device 400 may include a processor 401 and a memory 402. The electronic device 400 may also include one or more of a multimedia component 403, an input / output (I / O) interface 404, and a communication component 405.

[0094] The processor 401 controls the overall operation of the electronic device 400 to complete all or part of the steps in the information processing method described above. The memory 402 stores various types of data to support the operation of the electronic device 400. This data may include, for example, instructions for any application or method operating on the electronic device 400, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 403 may include a screen and audio components. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 402 or transmitted via communication component 405. The audio component also includes at least one speaker for outputting audio signals. I / O interface 404 provides an interface between processor 401 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 405 is used for wired or wireless communication between the electronic device 400 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G technologies, or combinations thereof, is not limited here. Therefore, the corresponding communication component 405 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0095] In an exemplary embodiment, the electronic device 400 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the information processing method described above.

[0096] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the information processing method described above. For example, the computer-readable storage medium may be the memory 402 including program instructions described above, which may be executed by the processor 401 of the electronic device 400 to complete the information processing method described above.

[0097] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0098] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0099] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A method of information processing, characterized by, The method includes: Based on the dialogue information of the target user regarding the target business, a business graph is obtained. The business graph is used to represent the correspondence between business-related information of the target business in the dialogue information. The business-related information includes at least one of business requirements, business execution actions, and business execution elements. Based on the business graph, multiple pre-established business tasks are sorted, and the first target task to be executed by multiple target users is selected based on the sorting results. Output first knowledge information associated with the plurality of first target items, the first knowledge information being used to assist in responding to the dialogue information.

2. The method of claim 1, wherein, The step of obtaining a business graph based on the dialogue information of the target user regarding the target business includes: Based on a preset dialogue recognition model, the business-related information and the correspondence between the business-related information are extracted from the dialogue information; Based on the business association information and the correspondence between the business association information, update the business nodes in the historical business graph and the connection relationships between the business nodes to obtain the business graph.

3. The method of claim 2, wherein, The historical business map is pre-established in the following ways: Retrieve historical conversation information of multiple users when performing various business tasks within a historical time period; Extract historical business association information and the correspondence between historical business association information from the historical dialogue information; The historical business association information is used as business nodes, and the correspondence between the historical business association information is used as the connection relationship of the business nodes to obtain the historical business map.

4. The method of claim 1, wherein, The step of sorting multiple pre-established business tasks according to the business graph, and selecting multiple target users to perform first target tasks based on the sorting results, includes: The first business node corresponding to the specified business association information is determined from the business graph, wherein the specified business association information is the most recently updated business association information in the dialogue information; From the multiple pending business items, the pending business items corresponding to multiple second business nodes adjacent to the first business node in the business graph are selected as candidate items, with different second business nodes corresponding to different pending business items. The execution priority of multiple candidate items is sorted, and the preset number of candidate items with the highest execution priority are selected as the first target item.

5. The method of claim 4, wherein, The process of prioritizing the execution of multiple candidate items includes: Determine the association probability between each candidate item and the specified business-related information; Determine the proportion of knowledge information associated with each candidate item in the historical output information; Based on the association probability and the output ratio, the execution priority of each candidate item is determined, and the priority ranking result is obtained.

6. The method of claim 5, wherein, The method further includes: The candidate items are filtered based on the association probability and the preset association probability threshold; Determining the execution priority for each candidate item includes: Determine the execution priority for each filtered candidate item.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Obtain the second target item input by the target user for the target business; Adjust the execution priority of the first objective based on the second objective.

8. An apparatus for information processing, characterized by comprising: The device includes: The acquisition module is used to acquire a business graph based on the dialogue information of the target user for the target business. The business graph is used to represent the correspondence between the business-related information of the target business in the dialogue information. The business-related information includes at least one of business requirements, business execution actions, and business execution elements. The analysis module is used to sort multiple pre-established business tasks according to the business graph, and select the first target task to be executed by multiple target users according to the sorting results. The output module is used to output first knowledge information associated with the plurality of first target items, and the knowledge information is used to assist in responding to the dialogue information.

9. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1-7.

10. An electronic device, comprising: include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1-7.

Images