Response method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a response method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a target knowledge theme corresponding to a business consultation request; determining a target knowledge image corresponding to the target knowledge theme in a preset image set according to the target knowledge theme; the target knowledge image comprises a plurality of nodes, and each node corresponds to at least one piece of knowledge information; obtaining additional consultation information, wherein the additional consultation information is supplementary information of the business consultation request; and determining a target node in the target knowledge image according to the additional consultation information, and performing response processing according to knowledge information corresponding to the target node. Therefore, the response efficiency can be improved when the service consultation request is responded.

Description

Technical Field

[0001] The present application relates to the field of communications, and in particular to a response method, device, electronic device and storage medium. Background Art

[0002] In the scenario of consulting and answering, manual answering is a very common answering method. Through manual answering, we can quickly understand the needs of the consultant, accurately perceive the consultant's patience state and adaptively adjust the answering method. Summary of the invention

[0003] The embodiments of the present application provide a response method, device, electronic device and storage medium to improve response efficiency.

[0004] In a first aspect, an embodiment of the present application provides a response method, including:

[0005] Determine the target knowledge topic corresponding to the business consultation request;

[0006] According to the target knowledge subject, determining a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information;

[0007] Acquire additional consulting information, where the additional consulting information is supplementary information of the business consulting request;

[0008] A target node is determined in the target knowledge image according to the additional inquiry information, and a response process is performed according to the knowledge information corresponding to the target node.

[0009] In a second aspect, an embodiment of the present application provides a response device, the device comprising:

[0010] A determination unit, used to determine a target knowledge topic corresponding to a business consulting request;

[0011] The determination unit is further configured to determine, according to the target knowledge subject, a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information;

[0012] An acquiring unit, configured to acquire additional consulting information, wherein the additional consulting information is supplementary information of the business consulting request;

[0013] A response unit is used to determine a target node in the target knowledge image according to the additional consultation information, and perform response processing according to the knowledge information corresponding to the target node.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor; and a memory configured to store computer-executable instructions, wherein the computer-executable instructions, when executed, cause the processor to execute the response method as described in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing computer-executable instructions, which, when executed by a processor, implement the response method as described in the first aspect.

[0016] It can be seen that in an embodiment of the present application, first, the target knowledge topic corresponding to the business consultation request is determined; secondly, based on the target knowledge topic, the target knowledge image corresponding to the target knowledge topic is determined in a preset image set; the target knowledge image includes multiple nodes, each node corresponds to at least one knowledge information; then, additional consultation information is obtained, and the additional consultation information is supplementary information for the business consultation request; then, the target node is determined in the target knowledge image based on the additional consultation information, and response processing is performed based on the knowledge information corresponding to the target node. In this way, on the one hand, since the target knowledge image corresponding to the target knowledge topic includes multiple nodes, and each node corresponds to at least one knowledge information, a target knowledge image can cover a large amount of knowledge relied upon for answering business consultation requests. By using the knowledge information corresponding to each node in the target knowledge image to answer, it is possible to avoid manually switching back and forth between multiple knowledge documents to view the knowledge required for the answer, thereby improving the answer efficiency and reducing the error rate of manual answers. On the other hand, when answering business consultation requests, it may be necessary to reply to the sender of the business consultation request after supplementing some additional consultation information. By locating the current corresponding target node in the target knowledge image and answering through the additional consultation information, the sender can get a targeted answer for each speech during the consultation process, thereby improving the consultation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor.

[0018] Figure 1 A schematic diagram of an implementation environment of a response method provided in an embodiment of the present application;

[0019] Figure 2 A processing flow chart of a response method provided in an embodiment of the present application;

[0020] Figure 3 A configuration interface diagram of a knowledge image provided in an embodiment of the present application;

[0021] Figure 4 A display interface diagram of a knowledge image provided in an embodiment of the present application;

[0022] Figure 5 A node information configuration page for a judgment node provided in an embodiment of the present application;

[0023] Figure 6 A node information configuration page for an image node provided in an embodiment of the present application;

[0024] Figure 7 A processing flow chart of another response method provided in an embodiment of the present application;

[0025] Figure 8 A schematic diagram of a response device provided in an embodiment of the present application;

[0026] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0028] The response method provided in one or more embodiments of this specification may be applicable to the implementation environment of the response method, such as Figure 1 As shown, the implementation environment at least includes a server 101 for responding to a business consultation request and a terminal device 102 for sending a business consultation request.

[0029] The server 101 may be a single server, or a server cluster consisting of a plurality of servers, or one or more cloud servers in a cloud computing platform, for responding to business consultation requests.

[0030] The terminal device 102 may be a mobile phone, a personal computer, a tablet computer, an e-book reader, a device for information interaction based on VR (Virtual Reality, virtual reality technology), a vehicle-mounted terminal, an IoT device, a wearable smart device, a laptop computer and a desktop computer, etc.; the terminal device 102 may be configured with an application client, and the specific form of the client may be an application, a subroutine within an application, a service module within an application, or a web program; the client may send a business consultation request to the server 101.

[0031] In this implementation environment, the server 101 determines the target knowledge topic corresponding to the business consulting request in the process of responding to the business consulting request sent by the terminal device 102; determines the target knowledge image corresponding to the target knowledge topic in the preset image set according to the target knowledge topic; the target knowledge image includes multiple nodes, each node corresponds to at least one knowledge information; obtains additional consulting information, which is supplementary information of the business consulting request; determines the target node in the target knowledge image according to the additional consulting information, and performs response processing according to the knowledge information corresponding to the target node. In this way, since the target knowledge image corresponding to the target knowledge topic includes multiple nodes, and each node corresponds to at least one knowledge information, a target knowledge image can cover a large amount of knowledge relied upon for answering business consultation requests. By using the knowledge information corresponding to each node in the target knowledge image to answer, it is possible to avoid manually switching back and forth between multiple knowledge documents to view the knowledge required for the answer, thereby improving the answer efficiency and reducing the error rate of manual answers. On the other hand, when answering business consultation requests, it may be necessary to reply to the sender of the business consultation request after supplementing some additional consultation information. By locating the current corresponding target node in the target knowledge image and answering through the additional consultation information, the sender can get a targeted answer for each speech during the consultation process, thereby improving the consultation experience.

[0032] This specification provides an embodiment of a response method:

[0033] In the scenario of manual consultation and response, the staff often needs to remember a large amount of business information in order to provide good consulting services to the consultant in a timely manner. In the case of a large amount of detailed data in the business information, the staff may rely on the query of knowledge documents. For example, the staff first searches for the reply information required by the consultant in one or more knowledge documents corresponding to the business consulted by the consultant, and then provides the reply information to the consultant. However, the consultant's patience will decrease significantly as the waiting time increases. In the case where the information that the consultant is concerned about involves multiple knowledge documents, the staff may need to switch back and forth between multiple knowledge documents, which takes a long time and may make mistakes, reducing the efficiency of the reply and bringing a bad experience to the consultant. In order to solve the above problems, an embodiment of the present application provides a reply method.

[0034] Figure 2 A processing flow chart of a response method provided in an embodiment of the present application. Figure 2 The response method provided in this embodiment specifically includes steps S202 to S208.

[0035] Step S202: determine the target knowledge topic corresponding to the business consultation request.

[0036] The business consultation request may be a request sent by a consultant for consulting on a target business. Consultants include but are not limited to: a person who is interested in the target business but has not yet applied for the target business, and a person who has already applied for the target business.

[0037] The target business can be a business that can provide services or products to users. For example, in a resource lending scenario, the target business can be a business that settles the repayment in advance; in a transaction scenario, the target business can be a business that sells in a discount area, and so on.

[0038] Before determining the target knowledge topic corresponding to the business execution request, a business consultation request may be obtained first. The business consultation request may be obtained by receiving a business consultation text and generating a corresponding business consultation request according to the business consultation text.

[0039] The business consultation text may be the text entered by the consultant when asking questions regarding the target business.

[0040] For example, business consultation text 1 is: Hello, I want to settle the repayment in advance, what should I do?

[0041] Business consultation text 2 is: The product A I purchased has quality problems. What proof is required for return or exchange?

[0042] Business consultation text 3 is: When will my order 1 be shipped?

[0043] The business consultation text may also be the text entered by the consultant when applying for / terminating the target business.

[0044] For example, business consultation text 4 is: Hello, how do I handle business A?

[0045] Business consultation text 5 is: Hello, I want to handle business A.

[0046] Referring to the business consultation text 5 in the above example, when the consultant expresses the intention to apply for or terminate the target business, the business consultation text implicitly contains questions about how to apply for the target business or how to terminate the target business.

[0047] The business consultation request in this embodiment may also be replaced by a business handling request or a business suspension request. The business handling request may be a request submitted by an applicant to apply for handling a target business. The business suspension request may be a request submitted by an applicant to apply for suspension of a target business. The response process for a business handling request or a business suspension request may refer to the response process for a business consultation request in this embodiment.

[0048] Obtaining the business consultation request may also be receiving a business consultation voice and generating a corresponding business consultation request according to the business consultation voice.

[0049] The business consultation voice may be the voice uttered by the consultant when asking questions about the target business, or the voice uttered by the consultant when applying for / terminating the target business.

[0050] In a specific implementation method, a business consulting request carries a request text; determining a target knowledge topic corresponding to the business consulting request includes: performing text similarity calculations based on the request text and each knowledge topic in a preset knowledge topic set to obtain the text similarity of each knowledge topic; and determining the target knowledge topic of the business consulting request based on the text similarity of each knowledge topic.

[0051] The request text carried in the business consultation request may be a business consultation text, or may be a text obtained by performing speech-to-text processing on the business consultation voice.

[0052] During specific implementation, a preset knowledge topic set may be configured in the knowledge middle platform, and the preset knowledge topic set includes multiple knowledge topics.

[0053] In information technology, for an organization or individual, knowledge refers to the possession of information or the ability to quickly locate information.

[0054] The knowledge center can be an institution-level knowledge management platform. All the knowledge of the institution is entered into the knowledge center. Any business systems or platforms of various departments of the institution that require relevant knowledge can obtain it through the knowledge center.

[0055] The knowledge entered in the knowledge center can be stored in the form of documents, that is, the knowledge center stores multiple knowledge documents, each of which stores different knowledge. Each knowledge document can correspond to a knowledge topic. The knowledge topic corresponding to each knowledge document can be obtained by refining the central idea of ​​the content of the knowledge document.

[0056] Exemplarily, a knowledge document for storing business consulting process knowledge of business A may correspond to the knowledge topic “business A consulting process”.

[0057] The text similarity calculation process is performed based on the request text and each knowledge topic in the preset knowledge topic set to obtain the text similarity between the request text and each knowledge topic. For each knowledge topic, the text similarity between the knowledge topic and the request text is used as the text similarity of the knowledge topic.

[0058] The target knowledge topic of the business consulting request is the same concept as the target knowledge topic corresponding to the previous business consulting request.

[0059] Based on the text similarity of each knowledge topic, the target knowledge topic of the business consulting request is determined. The text similarity of each knowledge topic can be sorted according to its numerical value to obtain a sorting result, and the text similarity with the largest value is determined based on the sorting result, and the knowledge topic corresponding to the text similarity with the largest value is determined as the target knowledge topic.

[0060] For example, request text 1: How to handle business A?

[0061] Knowledge Topic 1 is "A Business Process", Knowledge Topic 2 is "B Business Process", Knowledge Topic C is "C Business Termination Process", and so on.

[0062] Perform text similarity calculation on request text 1 and knowledge topic 1 to obtain text similarity s1 of knowledge topic 1;

[0063] Perform text similarity calculation on request text 1 and knowledge topic 2 to obtain text similarity s2 of knowledge topic 2;

[0064] The request text 1 and the knowledge topic 3 are subjected to text similarity calculation processing to obtain the text similarity s3 of the knowledge topic 3.

[0065] s1>s2>s3, the knowledge topic 1 corresponding to s1 is determined as the target knowledge topic.

[0066] In actual applications, users may express the same question in a variety of different ways.

[0067] For example, request text 1: How to handle business A?

[0068] Request text 2: I want to handle business A, what do I need to do?

[0069] Request text 3: Hello, I would like to consult how to handle business A.

[0070] Taking the above situation into consideration, multiple corresponding consulting statements can be pre-configured for each knowledge topic, and text similarity calculation can be performed based on the request text and each consulting statement of each knowledge topic in the preset knowledge topic set to obtain the text similarity of each knowledge topic; based on the text similarity of each knowledge topic, the target knowledge topic of the business consulting request is determined.

[0071] The text similarity calculation is performed based on the request text and each consulting sentence of each knowledge topic in the preset knowledge topic set to obtain the text similarity of each knowledge topic. This can be achieved in the following way:

[0072] A text similarity calculation is performed based on the request text and each consulting sentence of each knowledge topic in the preset knowledge topic set to obtain the text similarity between the request text and each consulting sentence of each knowledge topic; for each knowledge topic, the one with the largest text similarity between each consulting sentence of the knowledge topic and the second semantic recognition result is determined as the text similarity of the knowledge topic.

[0073] For example, knowledge topic A has three pre-configured consulting statements: statement X1, statement X2, and statement X3.

[0074] Perform text similarity calculation on the request text 1 and the sentence X1 to obtain the text similarity S1 between the request text 1 and the sentence X1;

[0075] Perform text similarity calculation on the request text 1 and the sentence X2 to obtain the text similarity S2 between the request text 1 and the sentence X2;

[0076] The text similarity calculation process is performed based on the request text 1 and the sentence X1 to obtain the text similarity S3 between the request text 1 and the sentence X3.

[0077] S1>S2>S3, S3 is determined as the text similarity of knowledge topic A.

[0078] Step S204, according to the target knowledge subject, determining a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each node corresponding to at least one piece of knowledge information.

[0079] The target knowledge image corresponding to the target knowledge subject may be a knowledge image used to describe a business consulting process, and the business consulting process is determined by the target knowledge subject.

[0080] The target knowledge topic may determine one or more corresponding business consulting processes from among multiple candidate business consulting processes.

[0081] The business consultation process may be a consultation processing process pre-configured for a target business.

[0082] The business consultation process includes but is not limited to: business processing sub-process, business termination sub-process and business information consultation sub-process.

[0083] The business processing sub-process is used to represent the various steps that need to be performed for the user to process the target business; the business termination sub-process is used to represent the various steps that need to be performed for the user to terminate the target business; the business information consultation sub-process is used to represent the various steps that need to be performed to answer the user's questions about the relevant information of the target business.

[0084] The preset image set may include a plurality of knowledge images.

[0085] A knowledge image can be knowledge represented in a knowledge document in the form of an image.

[0086] Each knowledge image may be one of a flowchart, a mind map, a swim lane diagram, and other image types.

[0087] Each knowledge document has a knowledge topic, and one or more knowledge images included in each knowledge document may correspond to the knowledge topic of the knowledge document.

[0088] In addition, each knowledge topic may have multiple subtopics, and each knowledge document may be divided into multiple subdocuments, wherein each subdocument corresponds to a subtopic of the knowledge topic of the knowledge document. One or more knowledge images included in each subdocument may correspond to the subtopic corresponding to the subdocument.

[0089] According to the target knowledge subject, the target knowledge image corresponding to the target knowledge subject is determined in the preset image set. It can be that according to the target knowledge subject, the business consulting process corresponding to the target knowledge subject is determined among multiple candidate business consulting processes, and then, the target knowledge image describing the business consulting process is determined in the preset image set.

[0090] The target knowledge graph includes a plurality of nodes, each of which corresponds to one or more pieces of knowledge information.

[0091] Each piece of knowledge information includes a knowledge text or a knowledge image. Knowledge text is the knowledge in the form of text in the knowledge document. Knowledge image is the knowledge in the form of image in the knowledge document.

[0092] In a specific implementation, the target knowledge image includes a text node and an image node; the preset knowledge topic set includes a target knowledge topic and a first knowledge topic; the knowledge information corresponding to the text node includes a sub-topic text of the target knowledge topic and a sub-topic text of the first knowledge topic; the knowledge information corresponding to the image node includes a sub-topic text of the target knowledge topic and an associated knowledge image describing a sub-process of a business consulting process corresponding to the target knowledge topic.

[0093] The target knowledge image includes multiple nodes, which may include text nodes, image nodes, and other types of nodes. The knowledge information corresponding to the other types of nodes may include knowledge represented by other data forms besides text and images.

[0094] The preset knowledge topic set may include a target knowledge topic and a first knowledge topic, where the first knowledge topic is other knowledge topics in the preset knowledge topic set except the target knowledge topic.

[0095] In a specific implementation, in a part of text nodes, the text nodes may correspond to multiple knowledge information, among which at least one knowledge information includes a sub-topic text of the target knowledge topic, and at least one knowledge information includes a sub-topic text of the first knowledge topic.

[0096] The number of the first knowledge topics can be one or more, and the number of the subtopic texts can be one or more.

[0097] The sub-topic text may be text information in a sub-document corresponding to the sub-topic in the knowledge document.

[0098] In another part of the text nodes, the text node may correspond to a plurality of knowledge information, each of which includes a sub-topic text of the target knowledge topic.

[0099] In a specific implementation, an image node may correspond to multiple pieces of knowledge information, among which at least one piece of knowledge information includes a sub-topic text of a target knowledge topic, and at least one piece of knowledge information includes an associated knowledge image describing a sub-process of a business consulting process corresponding to the target knowledge topic.

[0100] The associated knowledge image describing the sub-process of the business consulting process corresponding to the target knowledge subject refers to the associated knowledge image describing the sub-process, which belongs to the business consulting process corresponding to the target knowledge subject.

[0101] The business consultation process may have multiple sub-processes, and each image node may correspond to one of the multiple sub-processes.

[0102] Step S206, obtaining additional consulting information; the additional consulting information is supplementary information of the business consulting request.

[0103] The additional consultation information may be text input by the sender of the business consultation request after the business consultation request has been triggered, or voice uttered by the sender of the business consultation request after the business consultation request has been triggered.

[0104] The sender of the business consultation request is the consultant mentioned above who consults on the target business.

[0105] In the scenario where business consultation is answered by an agent, on the one hand, the sender of the business consultation request is making a business consultation without understanding the target business, and may not be clear about what information he needs to provide during the consultation process, such as the last four digits of the identity number, the communication number, the product type, etc. Therefore, after triggering the business consultation request, the sender usually needs to supplement the business consultation request with user information, product information or service information; on the other hand, in various situations such as when the sender applies for the target business or applies to terminate the target business, there may be some situations where the user's intention needs to be confirmed. For example, before the agent performs the processing confirmation operation for the sender, the sender needs to say "agree to process" first. Therefore, the sender needs to supplement the business consultation request with intention confirmation information under the guidance of the agent.

[0106] An agent is a job position in a call center or customer service department of an organization that provides target services. For example, agents can be divided into the following categories based on their job content:

[0107] General consultation: provide users with relevant information and record or transfer user information.

[0108] Technical support: providing technical support to users and arranging related services, etc.

[0109] Marketing and sales: outbound telephone marketing.

[0110] Information collection: screening, filtering, collecting or verifying large amounts of data.

[0111] Follow-up complaint category: handle user complaints and suggestions, and conduct follow-up visits to some users.

[0112] Other affairs: Complete various other work contents via telephone.

[0113] The agent can be a robot agent or a human agent.

[0114] In specific implementation, after step S204 is executed and before step S206 is executed, response processing can also be performed according to the intermediate response information. The intermediate response information can be determined by the business consultation process corresponding to the target knowledge image.

[0115] The intermediate response information may be a preset response script, for example, "Hello, please provide your mobile phone number", "Hello, what is your product type?".

[0116] The intermediate response information can also be an operation guidance phrase, for example, "Hello, you need to open XXX to perform XX operation."

[0117] The additional consultation information may include feedback information from the sender of the service consultation request to the intermediate response information.

[0118] Step S208, determining the target node in the target knowledge image according to the additional inquiry information, and performing response processing according to the knowledge information corresponding to the target node.

[0119] The target knowledge image is used to describe the business consultation process. According to the additional consultation information, the operation step corresponding to the additional consultation information in the business consultation process can be determined to determine the target node.

[0120] In the scenario where an agent performs business consultation responses, the business consultation process may include multiple operation steps of the agent. For some of the multiple operation steps, the agent may need to wait for the sender of the consultation service request to send additional consultation information before continuing to execute other operation steps after the operation step. Each operation step may correspond to a node in the target knowledge graph.

[0121] The node corresponding to the additional consulting information in the business consulting process may be a node corresponding to an operation step performed by the agent after receiving the additional consulting information.

[0122] In addition, the multiple nodes included in the target knowledge node may include a judgment node.

[0123] After the operation step corresponding to the judgment node, if the additional consultation information obtained is different, the next operation step to be performed by the agent is also different.

[0124] For example, in the target knowledge graph, node 1 is connected to node 2 and node 3 respectively, node 1 is located before node 2, node 1 is located before node 3, and node 1 is a judgment node. Node 1 corresponds to operation step 1, node 2 corresponds to operation step 2, and node 3 corresponds to operation step 3.

[0125] After the agent performs operation step 1, the additional consultation information 1 of the consultant is obtained, and the agent performs operation step 2. Alternatively, after the agent performs operation step 1, the additional consultation information 2 of the consultant is obtained, and the agent performs operation step 3.

[0126] The knowledge information corresponding to the target node may include preset response words, and the response processing is performed according to the knowledge information corresponding to the target node, which may be broadcasting the preset response words.

[0127] The knowledge information corresponding to the target node may include a preset knowledge image, which is used to describe a sub-process of the business consulting process. The response processing is performed according to the knowledge information corresponding to the target node, and the response processing may be performed according to the sub-process corresponding to the preset knowledge image.

[0128] In a specific implementation, response processing is performed based on the knowledge information corresponding to the target node, including: when the target node is a text node, generating first response information based on the sub-topic text of the target knowledge topic and the sub-topic text of the first knowledge topic; and performing response processing based on the first response information.

[0129] The sub-topic text of the target knowledge topic and the sub-topic text of the first knowledge topic are concatenated to generate a response text, which is used as the first response information. The response processing according to the first response information can be generating a response voice according to the response text and broadcasting it, or displaying the response text, etc.

[0130] By concatenating the sub-topic text of the target knowledge topic and the sub-topic text of the first knowledge topic to obtain the response text, the purpose of knowledge integration of different knowledge documents can be achieved without manual switching back and forth between multiple knowledge documents, thereby improving the response efficiency.

[0131] In a specific implementation, the target knowledge image also includes node description information of each node; determining the target node in the target knowledge image according to the additional consulting information includes: performing semantic recognition processing according to the additional consulting information to obtain the semantic recognition result of the additional consulting information; in the target knowledge image, determining the corresponding first node according to the additional consulting information; the additional consulting information includes feedback information of the sender of the business consulting request to the preset response words of the first node; determining the additional text corresponding to the additional consulting information according to the semantic recognition result and the node description information of the first node; performing text similarity calculation processing according to the additional text and the node description information of at least one child node of the first node to obtain the text similarity of each child node; determining the target node in the target knowledge image according to the text similarity of each child node.

[0132] In practical applications, when the additional consulting information is the feedback information of the sender of the business consulting request to the intermediate response information, on the one hand, the sender is much more likely to give a brief reply when sending the additional consulting information than to say a complete sentence. On the other hand, when expressing different choices, the sender's additional consulting information may not be a fixed sentence. Therefore, before performing text similarity calculation processing, the sender's feedback information and node description information can be combined to determine the sender's intention, so as to more accurately locate the target node.

[0133] For example, the robot agent provides the consultant with a plan "I suggest you do xxx first and then xxx", and the consultant says "OK", which means that the consultant agrees with the plan. The consultant says "Is there any other way", which means that the consultant disagrees with the plan.

[0134] Semantic recognition processing is performed according to the additional consulting information to obtain a semantic recognition result of the additional consulting information.

[0135] In the target knowledge image, the corresponding first node is determined according to the additional consulting information, and the node corresponding to the operation step before obtaining the additional consulting information is determined as the first node. The additional consulting information includes feedback information of the sender of the business consulting request on the preset response words of the first node.

[0136] The description information of each node includes the preset response words of the node.

[0137] The additional text corresponding to the additional consultation information is determined based on the semantic recognition result and the node description information of the first node. The user intention corresponding to the semantic recognition result can be determined based on the semantic recognition result and the preset response words of the first node, and the additional text corresponding to the additional consultation information is generated based on the user intention and the preset response words of the first node.

[0138] For example, the preset response of the first node is: We suggest that you can...

[0139] The additional consultation information is: It is inconvenient, is there any other method?

[0140] The semantic recognition result of the additional consulting information is X.

[0141] Based on the semantic recognition result X and the preset response words "It is recommended that you can...", it can be determined that the user intention corresponding to the semantic recognition result includes the user rejecting the suggestion. Then, based on the user intention and the preset response words "It is recommended that you can...", the additional text "User rejects..." corresponding to the additional consultation information is generated.

[0142] The first node may be a judgment node, and the first node may be connected to at least one child node, and the first node is in front and the child node is in the back. That is, each of the at least one child node of the first node may be a node located after the first node in the target knowledge image.

[0143] The description information of a node may include a judgment question and may also include branch conditions of the judgment question.

[0144] For example, the judgment node is node 1, node 1 is connected to node 2 and node 3 respectively, node 1 is located before node 2, and node 1 is located before node 3, then node 2 and node 3 are both child nodes of node 1.

[0145] The description information of node 1 includes the judgment question of node 1: whether to agree with suggestion A.

[0146] The description information of node 2 includes branch condition 1 of the judgment problem of node 1: agree with suggestion A.

[0147] The description information of node 3 includes branch condition 2 of the judgment problem of node 1: reject suggestion A.

[0148] A text similarity calculation process is performed based on the additional text and the node description information of at least one child node of the first node to obtain the text similarity of each child node. Alternatively, a text similarity calculation process is performed based on the additional text and the node description information of each child node in the at least one child node to obtain the text similarity between the node description information of each child node and the additional text, and the text similarity is used as the text similarity of each child node.

[0149] For example, at least one child node of the first node includes node 1, node 2, and node 3.

[0150] The text similarity calculation process is performed based on the additional text and the node description information of node 1 to obtain the text similarity between the additional text and the node description information of node 1, and the text similarity between the additional text and the node description information of node 1 is used as the text similarity of node 1.

[0151] The text similarity calculation process is performed based on the additional text and the node description information of node 2 to obtain the text similarity between the additional text and the node description information of node 2, and the text similarity between the additional text and the node description information of node 2 is used as the text similarity of node 2.

[0152] The text similarity calculation process is performed based on the additional text and the node description information of node 3 to obtain the text similarity between the additional text and the node description information of node 3, and the text similarity between the additional text and the node description information of node 3 is used as the text similarity of node 3.

[0153] According to the text similarity of each child node, the target node is determined in the target knowledge image. The text similarity of each child node can be sorted according to the numerical value to obtain the sorting result, and the text similarity with the largest value is determined according to the sorting result, and the child node corresponding to the text similarity with the largest value is determined as the target node.

[0154] For example, the appended text is "User rejects suggestion A".

[0155] The first node is node 1, and at least one child node of the first node includes node 2 and node 3.

[0156] The description information of node 2 includes: agreeing with suggestion A.

[0157] The description information of node 3 includes: rejecting suggestion A.

[0158] Calculate the text similarity between the appended text and the description information of node 2 to obtain the text similarity S1 corresponding to node 2;

[0159] The text similarity between the additional text and the description information of node 3 is calculated to obtain the text similarity S2 corresponding to node 3.

[0160] S2>S1, so the node 3 corresponding to S2 can be determined as the target node.

[0161] In a specific implementation, the business consultation request carries a user identifier; after determining the target knowledge image corresponding to the target knowledge topic in a preset image set according to the target knowledge topic, the response method also includes: querying the corresponding user business data according to the user identifier; performing response processing according to the knowledge information corresponding to the target node, including: performing response processing according to the user business data and the knowledge information corresponding to the target node.

[0162] In a specific implementation, the business consultation request may be obtained when the sender of the business consultation request logs in to the answering system, and the business consultation request carries the user identification of the sender.

[0163] The user identifier is used to identify the unique corresponding user in the user data set corresponding to the answering system.

[0164] The answering system may pre-configure a user data set of registered users, the user data set including a user ID and user service data of each of the multiple registered users. The request triggered by each registered user when logging into the answering system carries the user ID of the registered user.

[0165] According to the user ID, the user service data corresponding to the user ID can be queried in the user data set.

[0166] The response processing is performed according to the user business data and the knowledge information corresponding to the target node. It can be that the sender of the business consultation request is judged according to the user business data whether the judgment condition corresponding to the target node is met, and the response processing is performed according to the judgment result and the knowledge information corresponding to the target node.

[0167] By performing response processing based on the user business data determined by the user identifier and the knowledge information corresponding to the target node, on the one hand, the authenticity of the user business data can be guaranteed to prevent illegal users from impersonating others to handle or terminate the target business. On the other hand, the sender of the business consultation request may not remember his or her own user business data, which led to the consultation need, or the user business data provided by the sender of the business consultation request based on memory may be inaccurate and interfere with the response system.

[0168] In a specific implementation, the answering method also includes: receiving a node selection input in a target knowledge image; determining a corresponding node to be processed based on the node selection input; and displaying the target knowledge image and knowledge information corresponding to the node to be processed in an associated manner.

[0169] In a scenario where business consultation is answered by an agent, a node selection input from an artificial agent in a target knowledge image may be received, and a node to be processed selected by the artificial agent may be determined based on the node selection input.

[0170] The node to be processed can be a text node or an image node.

[0171] When the node to be processed is a text node, the target knowledge image and the knowledge information corresponding to the node to be processed are displayed in association, and the target knowledge image can be displayed in a first preset area, and at least one knowledge text corresponding to the node to be processed can be displayed in a second preset area. The associated display of the target knowledge image and the at least one knowledge text is conducive to the human agent to view the image in accordance with the knowledge text.

[0172] When the node to be processed is an image node, the target knowledge image and the knowledge information corresponding to the node to be processed are displayed in association, and the target knowledge image can be displayed in the first preset area, and the knowledge image corresponding to the node to be processed can be displayed in the second preset area. The associated display of the target knowledge image and the knowledge image corresponding to the node to be processed is conducive to the manual seat to view multiple images at the same time and compare them.

[0173] Compared with viewing a graph in a knowledge document and jumping to another knowledge document to view the knowledge text or knowledge image when the node to be processed in the graph involves another knowledge document, the data query efficiency in this implementation is higher and the errors that may be caused by human agents switching back and forth between multiple knowledge documents are reduced.

[0174] In such Figure 2 In the embodiment shown, the target knowledge subject corresponding to the business consultation request is determined; according to the target knowledge subject, the target knowledge image corresponding to the target knowledge subject is determined in the preset image set; the target knowledge image includes multiple nodes, each node corresponds to at least one knowledge information; additional consultation information is obtained, and the additional consultation information is supplementary information of the business consultation request; the target node is determined in the target knowledge image according to the additional consultation information, and the answer processing is performed according to the knowledge information corresponding to the target node. In this way, since the target knowledge image corresponding to the target knowledge subject includes multiple nodes, and each node corresponds to at least one knowledge information, a target knowledge image can cover a large amount of knowledge relied on for answering the business consultation request, and the knowledge information corresponding to each node in the target knowledge image is used for answering, which can avoid manually switching back and forth in multiple knowledge documents to view the knowledge required for the answer, improve the answer efficiency, and reduce the error rate of manual answering; on the other hand, when answering the business consultation request, it may be necessary to reply to the sender of the business consultation request after supplementing some additional consultation information. By locating the current corresponding target node in the target knowledge image and answering by adding additional consultation information, the sender can get a targeted answer every time he speaks during the consultation process, which improves the consultation experience.

[0175] Corresponding to the above-described response method, based on the same technical concept, the present application embodiment also provides a method for generating a knowledge image. The knowledge image generated by the knowledge image generation method can be used in the response method provided by the aforementioned method embodiment. The knowledge image generation method provided by the present application embodiment can be specifically referred to in Figure 3-Figure 6 .

[0176] Figure 3 A configuration interface diagram of a knowledge image provided in an embodiment of the present application.

[0177] like Figure 3As shown, in the image information area of ​​the knowledge image configuration interface, to the right of the image name is an image name input box 302, to the right of the image type is an image type selection box 304, to the right of the associated knowledge is a knowledge adding control 306, and to the right of the knowledge adding control 306 is a control "Start Drawing".

[0178] The image name input box 302 is pre-configured to have a maximum input length of 50 characters.

[0179] The image type selection box 304 is pre-configured with multiple image type selection items, such as basic image, flow chart, mind map, swim lane diagram, etc.

[0180] For example, the knowledge adding control 306 may be a button control. After clicking the button control, a pop-up window of the knowledge list may pop up, and the pop-up window displays a knowledge list consisting of multiple knowledge topics. Figure 2 Corresponding description of the embodiments.

[0181] If a knowledge topic (represented as knowledge A) needs to add a knowledge image, you can click Add Process on the drawing tool page of the knowledge platform to open the image workbench.

[0182] The drawing tool page of the knowledge platform can be the page of the open source image engine antv-x6 (X6 for short). In addition, the image drawing can be performed by Kmp-x6. Kmp-x6 is an editable and previewable component package based on the data of X6 and the knowledge platform. If other projects of the same organization want to view the drawn images, they only need to import the package to call for viewing and preview.

[0183] The knowledge platform may include an image platform, which may display Figure 3 The configuration interface of the knowledge image shown.

[0184] Fill in the image name in the image name input box 302 in the image workbench, select the image type in the image type selection box 304 in the image workbench, and click the knowledge addition control 306. Select "Knowledge A" in the knowledge list displayed in the pop-up window and then click the control "Start Drawing".

[0185] Selecting "knowledge A" as the associated knowledge indicates that the knowledge image to be drawn serves knowledge A, that is, knowledge A is associated with the entire knowledge image to be drawn.

[0186] like Figure 3 As shown, the toolbar area of ​​the knowledge image configuration interface includes a start icon, a common node icon, a judgment node icon, a sub-image icon and an end icon.

[0187] like Figure 3 As shown, in the image drawing area of ​​the knowledge image configuration interface:

[0188] The start icon 308 indicates the start of the business consultation process, and the end icon 322 indicates the result of the business consultation process. The mind map of the business consultation process includes multiple nodes, wherein the first node is the common node 1 corresponding to the common node icon 310, and the common node 1 points to the judgment node corresponding to the judgment node icon 312. The output of the judgment node includes two branches, one of which points to the common node 2 corresponding to the common node icon 314, and the other branch points to the common node 3 corresponding to the common node icon 318. The common node 2 points to the image node 1 corresponding to the sub-image icon 316. The common node 3 points to the image node 2 corresponding to the sub-image icon 320.

[0189] Below the image drawing area are controls: Cancel, Save as Draft, Preview, and Save and Publish.

[0190] like Figure 3 As shown, the associated display area 324 of the configuration interface of the knowledge image includes knowledge text 1, knowledge text 2, knowledge text 3 and knowledge text 4.

[0191] After the knowledge image configuration is completed, the image workbench can display the display interface of the knowledge image.

[0192] Figure 4 A display interface diagram of a knowledge image provided in an embodiment of the present application.

[0193] like Figure 4 As shown, the start icon 402 indicates the start of the business consultation process, and the end icon 416 indicates the result of the business consultation process. The mind map of the business consultation process includes multiple nodes, wherein the first node is the common node 1 corresponding to the common node icon 404, and the common node 1 points to the judgment node corresponding to the judgment node icon 406. The output of the judgment node includes two branches, one of which points to the common node 2 corresponding to the common node icon 408, and the other branch points to the common node 3 corresponding to the common node icon 412. The common node 2 points to the image node 1 corresponding to the sub-image icon 410. The common node 3 points to the image node 2 corresponding to the sub-image icon 414.

[0194] The display interface diagram also includes an associated display area 418 , which includes knowledge text 1 , knowledge text 2 , knowledge text 3 , and knowledge text 4 .

[0195] The associated display area 418 can display the knowledge information corresponding to the selected node to be processed.

[0196] For example, when the human agent selects the ordinary node 1 corresponding to the ordinary node icon 404 as the node to be processed, the knowledge information corresponding to the ordinary node 1 can be seen in the associated display area 418, including knowledge text 1, knowledge text 2, knowledge text 3 and knowledge text 4, thereby achieving the effect of associating the display of the node to be processed and the knowledge information corresponding to the node to be processed.

[0197] For another example, when the human agent selects the image node 2 corresponding to the sub-image icon 414 as the node to be processed, the knowledge information corresponding to the image node 2 can be seen in the associated display area 418, including the knowledge text 5 and the associated knowledge image ( Figure 3 (not shown in the figure), thereby achieving the effect of associating and displaying the node to be processed with the knowledge information corresponding to the node to be processed.

[0198] Since the technical concept is the same, the description in this embodiment is relatively simple, and the relevant parts may refer to the corresponding description of the method embodiment provided above.

[0199] In the image drawing process, taking the brain map type image as an example, the nodes that can be added to the image include ordinary nodes, judgment nodes and image nodes. Figure 5 and Figure 6 The node configuration method is illustrated as an example. Figure 5 A node information configuration page for a judgment node provided in an embodiment of the present application. Figure 6 A node information configuration page for an image node provided in an embodiment of the present application.

[0200] i. Ordinary nodes:

[0201] 1. Double-click the node to configure the node information:

[0202] a)Node name

[0203] b) Node Description

[0204] c) Related knowledge

[0205] The node description may include the node description information of the common node. The definition of node description information can refer to Figure 2 Corresponding description of the embodiments.

[0206] It should be noted that, in the process of configuring the node information of the common node, the associated knowledge of the common node is different from the associated knowledge “knowledge A” of the knowledge image to be drawn.

[0207] The associated knowledge of a common node is used to indicate that the node is related to knowledge text in other knowledge documents. For example, when a human agent queries knowledge A and is at the current node, he needs to check the content of knowledge B. Knowledge B is another knowledge topic different from knowledge A.

[0208] A common node may be configured with one or more associated knowledge, each of which may be a sub-topic text of a first knowledge topic, where the first knowledge topic is a knowledge topic other than the target knowledge topic (ie, knowledge A).

[0209] 2. Click to select, click a chapter (recorded as chapter P1) of "Knowledge A" in the associated display area 324, and then the common node can be associated with "Chapter P1" of "Knowledge A", which means that the common node is related to "Chapter P1" of "Knowledge A". In the image drawing stage, the associated display area 324 can display multiple chapters of "Knowledge A".

[0210] "Chapter P1" can be a sub-topic text of the target knowledge topic (ie, knowledge A).

[0211] In addition, a normal node can be a text node. For the definition of a text node, see Figure 2 The embodiments correspond to the description section.

[0212] ii. Judgment Node

[0213] 1. Double-click the node to configure the node information:

[0214] a)Node name

[0215] b) Node Description

[0216] c) Related knowledge

[0217] The configuration of the node description and associated knowledge of the judgment node can refer to the configuration of the common node.

[0218] 2. Click to select, and click on a chapter (recorded as chapter P2) of "Knowledge A" in the associated display area 324, then the judgment node can be associated with the "Chapter P2" of "Knowledge A", which means that the judgment node involves the "Chapter P2" of "Knowledge A".

[0219] "Chapter P2" can be a sub-topic text of the target knowledge topic (ie, knowledge A).

[0220] In addition, the judgment node can be a text node. For the definition of judgment node, see Figure 2 The embodiments correspond to the description section.

[0221] like Figure 5 As shown, in the node information configuration interface 500 of the judgment node, the right side of the node name is a node name input box 502, and the right side of the node description is a node description input box 504. The right side of the associated knowledge is a knowledge adding control 506. The currently displayed associated knowledge includes: knowledge file 508, knowledge text 510 and knowledge text 512.

[0222] Among them, knowledge text 508, namely knowledge text a1, is in an expired state;

[0223] The knowledge text 510, i.e., the knowledge text a2, is in a deleted state;

[0224] The knowledge text 512, namely the knowledge text a3, is neither invalid nor deleted, so the knowledge text 512 is normally associated with the judgment node.

[0225] iii. Image Node

[0226] 1. Double-click the node to configure the node information:

[0227] a)Node name

[0228] b) Node Description

[0229] c) Associated images

[0230] d) Associated knowledge (same as ordinary nodes)

[0231] The associated image of an image node is used to indicate that the image node is related to knowledge images in other knowledge documents. For example, when an agent queries knowledge A and is at the current node, he needs to view another knowledge image.

[0232] In the case where the knowledge image to be drawn is a flowchart, the associated image of the image node in the knowledge image to be drawn may be a flowchart of a sub-process of the business consultation process described by the flowchart.

[0233] 2. Click to select, and click on a chapter (recorded as chapter P3) of "Knowledge A" in the associated display area 324, then the image node can be associated with "Chapter P3" of "Knowledge A", which means that the image node involves "Chapter P3" of "Knowledge A".

[0234] like Figure 6 As shown, in the node information configuration interface 600 of the image node, the right side of the node name is a node name input box 602, and the right side of the node description is a node description input box 604. The right side of the associated image is an image search box 606. The currently displayed associated images include: knowledge image 608, knowledge image 610 and knowledge image 612.

[0235] Among them, the knowledge image 608, namely the knowledge image b1, is in an expired state;

[0236] The knowledge image 610, i.e., the knowledge image b2, is in a deleted state;

[0237] The knowledge image 612, namely the knowledge image b3, is neither invalid nor deleted, so the knowledge image 612 is normally associated with the image node.

[0238] In addition, during the image drawing process, you can implement custom connections through the registered connections provided by X6, and X6 supports the deletion of edges and the addition of text.

[0239] The registered connections provided by X6 include but are not limited to: basic image connections, mind map connections, swim lane diagram connections and annotation connections.

[0240] During the image drawing process, you can implement custom nodes through the registered nodes provided by X6 and support associating data on the nodes.

[0241] Custom nodes include but are not limited to: common nodes, judgment nodes, and image nodes.

[0242] During the image drawing process, business processing can be completed through the node events provided by X6.

[0243] Node events include but are not limited to: node selection event, node addition event, image change event, node deletion event, edge addition event, edge deletion event, node dragging event, node size change event, and so on.

[0244] Through the above customized edges, nodes and monitoring of X6's internal events and the data of the knowledge center, the editing and viewing engine of image knowledge such as Kmp-x6 public component package can be completed. When using the Kmp-x6 component package, different image types can be viewed and edited through defined parameters, such as basic images, mind maps, swim lane diagrams, etc. Kmp-x6 also supports external monitoring. When a business button or an action event is triggered, the user can perform different businesses through the parameters and events transmitted.

[0245] After the image drawing is completed in the image drawing area and the node information of each node is configured, the drawn knowledge image can be saved.

[0246] Exemplarily, by clicking the control “Save as Draft”, verification of the image name may be triggered.

[0247] The image name verification method may include: verifying whether the image name is empty, and verifying whether the image name is the same as a saved image name.

[0248] For example, by clicking the control "Save and Publish", the following verification methods can be triggered:

[0249] (1) Perform mandatory and duplicate image name checks within the specified business domain. If the check fails, a prompt will be displayed below the field: "Image name cannot be empty" or "Image name cannot be repeated."

[0250] (2) If the image is a basic image type, check whether the image contains a start icon and an end icon, and generate a warning prompt based on the check result: "The image has no start / end nodes and cannot be saved."

[0251] (3) Perform image integrity check. All nodes must be connected and associated with other nodes. They cannot be isolated nodes. Otherwise, a warning message will be generated: "There is an isolated node in the image, please check."

[0252] (4) Check whether the knowledge text associated with each node exists. If not, automatically disconnect the link between the node and the knowledge text.

[0253] (5) Check whether the knowledge text associated with each node is abnormal. The knowledge status includes but is not limited to: expired, deleted or other abnormal status. If there is an abnormality, a warning prompt is generated: "The node-associated knowledge text is abnormal, please check."

[0254] (6) Check whether the knowledge image associated with each node is abnormal. The knowledge status includes but is not limited to: expired, deleted or other abnormal status. If there is an abnormality, a warning prompt is generated: "The knowledge image associated with the node is abnormal, please check", and the node is automatically selected.

[0255] (7) Check whether the image status of each node-associated knowledge image is abnormal. The image status includes published and unpublished. If the image status is unpublished, a warning prompt is generated: "The node-associated sub-image status is abnormal, please check", and the node is automatically selected.

[0256] After the verification is completed, the "mask" layer page can be displayed, and the page text prompts: "Image is being published, please wait!" After the publishing is completed, a prompt can be generated: "Published successfully" and stay on the current page.

[0257] After successful publishing, you can view the knowledge image, such as Figure 4 shown.

[0258] Figure 7 A processing flow chart of another response method provided in an embodiment of the present application.

[0259] In execution Figure 7 Before the answering method shown, a method for generating a knowledge image can be executed by the knowledge middle station, and the method for generating a knowledge image includes the following steps:

[0260] Step S702, adding drawing.

[0261] Step S704 is image association knowledge.

[0262] Associating knowledge with an image may be performed by configuring a corresponding knowledge topic for the knowledge image.

[0263] Step S706, drawing an image.

[0264] Step S708, associating knowledge with image nodes.

[0265] Associating knowledge with image nodes may be performed by configuring corresponding knowledge information for each node in the knowledge image. The knowledge information may include knowledge text or knowledge image.

[0266] Step S710, publishing image knowledge.

[0267] Publish the knowledge image that has been drawn and the node information is configured.

[0268] Steps S702 to S710 may refer to Figure 3-Figure 6 An embodiment of a method for generating a knowledge image is shown.

[0269] like Figure 7 As shown, the response method executed by the customer service system may include the following steps:

[0270] Step S712, query image knowledge.

[0271] The agent sends a request to obtain the target knowledge image to the knowledge center through the customer service system.

[0272] The knowledge center returns the target knowledge image to the customer service system.

[0273] Step S714, click each image node.

[0274] It should be noted that clicking each image node refers to clicking each node in the target knowledge image, which node may be a text node or an image node, rather than just clicking the image node.

[0275] Step S716, checking the knowledge associated with each node.

[0276] The knowledge associated with each node may include knowledge text or knowledge image.

[0277] Since the technical concept is the same, the description in this embodiment is relatively simple, and the relevant parts may refer to the corresponding description of the method embodiment provided above.

[0278] In the above-mentioned embodiment, a response method is provided. Correspondingly, based on the same technical concept, an embodiment of the present application also provides a response device, which will be described below in conjunction with the accompanying drawings.

[0279] Figure 8 A schematic diagram of a response device provided in an embodiment of the present application.

[0280] This embodiment provides a response device 800, including:

[0281] A determination unit 802, configured to determine a target knowledge topic corresponding to the business consultation request;

[0282] The determining unit 802 is further configured to determine, according to the target knowledge subject, a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information;

[0283] An acquiring unit 804 is configured to acquire additional consulting information, where the additional consulting information is supplementary information of the business consulting request;

[0284] The response unit 806 is used to determine a target node in the target knowledge image according to the additional query information, and perform response processing according to the knowledge information corresponding to the target node.

[0285] Optionally, the target knowledge image further includes node description information of each node; the response unit 806 is specifically configured to:

[0286] Performing semantic recognition processing according to the additional consulting information to obtain a semantic recognition result of the additional consulting information;

[0287] In the target knowledge image, a corresponding first node is determined according to the additional consultation information; the additional consultation information includes feedback information of a sender of the business consultation request on a preset response script of the first node;

[0288] Determining, according to the semantic recognition result and the node description information of the first node, the additional text corresponding to the additional consulting information;

[0289] Performing text similarity calculation processing based on the additional text and node description information of at least one child node of the first node to obtain text similarity of each child node;

[0290] The target node is determined in the target knowledge image according to the text similarity of each of the sub-nodes.

[0291] Optionally, the service consultation request carries a user identifier; the answering device 800 further includes:

[0292] A query unit, used to query corresponding user service data according to the user identifier;

[0293] The response unit 806 is specifically configured to:

[0294] Response processing is performed according to the user service data and the knowledge information corresponding to the target node.

[0295] Optionally, the business consultation request carries a request text; the determining unit 802 is specifically configured to:

[0296] Performing text similarity calculation processing on the request text and each knowledge topic in the preset knowledge topic set to obtain the text similarity of each knowledge topic;

[0297] According to the text similarity of each of the knowledge topics, a target knowledge topic of the business consulting request is determined.

[0298] Optionally, the response device 800 further includes:

[0299] A receiving unit, configured to receive a node selection input in the target knowledge image;

[0300] The determination unit 802 is further configured to determine a corresponding node to be processed according to the node selection input;

[0301] A display unit is used to display the target knowledge image and the knowledge information corresponding to the node to be processed in an associated manner.

[0302] Optionally, the target knowledge image includes text nodes and image nodes;

[0303] The preset knowledge topic set includes a target knowledge topic and a first knowledge topic, and the knowledge information corresponding to the text node includes a sub-topic text of the target knowledge topic and a sub-topic text of the first knowledge topic;

[0304] The knowledge information corresponding to the image node includes the sub-topic text of the target knowledge topic and an associated knowledge image describing the sub-process of the business consulting process corresponding to the target knowledge topic.

[0305] Optionally, the response unit 806 is specifically configured to:

[0306] In the case where the target node is the text node, generating first response information according to the sub-topic text of the target knowledge topic and the sub-topic text of the first knowledge topic;

[0307] Perform response processing according to the first response information.

[0308] The response device provided in the embodiment of the present application includes: a determination unit, used to determine the target knowledge topic corresponding to the business consultation request; the determination unit is also used to determine the target knowledge image corresponding to the target knowledge topic in a preset image set according to the target knowledge topic; the target knowledge image includes multiple nodes, each node corresponds to at least one knowledge information; an acquisition unit, used to obtain additional consultation information, the additional consultation information is supplementary information of the business consultation request; the response unit is used to determine the target node in the target knowledge image according to the additional consultation information, and perform response processing according to the knowledge information corresponding to the target node. In this way, since the target knowledge image corresponding to the target knowledge topic includes multiple nodes, and each node corresponds to at least one knowledge information, a target knowledge image can cover a large amount of knowledge relied upon for answering business consultation requests. By using the knowledge information corresponding to each node in the target knowledge image to answer, it is possible to avoid manually switching back and forth between multiple knowledge documents to view the knowledge required for the answer, thereby improving the answer efficiency and reducing the error rate of manual answers. On the other hand, when answering business consultation requests, it may be necessary to reply to the sender of the business consultation request after supplementing some additional consultation information. By locating the current corresponding target node in the target knowledge image and answering through the additional consultation information, the sender can get a targeted answer for each speech during the consultation process, thereby improving the consultation experience.

[0309] Corresponding to the above-described response method, based on the same technical concept, an embodiment of the present application further provides an electronic device, which is used to execute the above-provided response method. Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0310] like Fig. 9 As shown, the electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors 901 and memory 902, and the memory 902 may store one or more storage applications or data. Among them, the memory 902 can be a short-term storage or a persistent storage. The application stored in the memory 902 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the electronic device. Furthermore, the processor 901 can be configured to communicate with the memory 902 and execute a series of computer executable instructions in the memory 902 on the electronic device. The electronic device may also include one or more power supplies 903, one or more wired or wireless network interfaces 904, one or more input / output interfaces 905, one or more keyboards 906, etc.

[0311] In a specific embodiment, the electronic device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions for the electronic device, and the one or more programs configured to be executed by one or more processors include the following computer executable instructions:

[0312] Determine the target knowledge topic corresponding to the business consultation request;

[0313] According to the target knowledge subject, determining a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information;

[0314] Acquire additional consulting information, where the additional consulting information is supplementary information of the business consulting request;

[0315] A target node is determined in the target knowledge image according to the additional inquiry information, and a response process is performed according to the knowledge information corresponding to the target node.

[0316] An embodiment of a computer-readable storage medium provided in this specification is as follows:

[0317] Corresponding to the response method described above, based on the same technical concept, an embodiment of the present application also provides a computer-readable storage medium.

[0318] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions. When the computer-executable instructions are executed by a processor, the following process is implemented:

[0319] Determine the target knowledge topic corresponding to the business consultation request;

[0320] According to the target knowledge subject, determining a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information;

[0321] Acquire additional consulting information, where the additional consulting information is supplementary information of the business consulting request;

[0322] A target node is determined in the target knowledge image according to the additional inquiry information, and a response process is performed according to the knowledge information corresponding to the target node.

[0323] It should be noted that the embodiment of the computer-readable storage medium in this specification and the embodiment of the response method in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the corresponding method mentioned above, and the repeated parts will not be repeated.

[0324] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0325] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0326] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable device to generate a machine, so that the instructions executed by the processor of the computer or other programmable device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0327] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0328] These computer program instructions may also be loaded onto a computer or other programmable device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0329] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0330] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0331] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0332] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0333] Embodiments of the present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0334] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0335] The above description is only an embodiment of this document and is not intended to limit this document. For those skilled in the art, this document may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this document should be included in the scope of the claims of this document.

Claims

1. A response method, characterized in that: include: Determine the target knowledge topic corresponding to the business consultation request; According to the target knowledge subject, determining a target knowledge image corresponding to the target knowledge subject in a preset image set; The target knowledge graph includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information; Acquire additional consulting information, where the additional consulting information is supplementary information of the business consulting request; A target node is determined in the target knowledge image according to the additional inquiry information, and a response process is performed according to the knowledge information corresponding to the target node.

2. The method according to claim 1, characterized in that: The target knowledge image also includes node description information of each of the nodes; and determining the target node in the target knowledge image according to the additional consulting information includes: Performing semantic recognition processing according to the additional consulting information to obtain a semantic recognition result of the additional consulting information; In the target knowledge image, a corresponding first node is determined according to the additional consultation information; the additional consultation information includes feedback information of a sender of the business consultation request on a preset response script of the first node; Determining, according to the semantic recognition result and the node description information of the first node, the additional text corresponding to the additional consulting information; Performing text similarity calculation processing based on the additional text and node description information of at least one child node of the first node to obtain text similarity of each child node; The target node is determined in the target knowledge image according to the text similarity of each of the sub-nodes.

3. The method according to claim 1, characterized in that The business consultation request carries a user identifier; after determining the target knowledge image corresponding to the target knowledge topic in a preset image set according to the target knowledge topic, the method further includes: According to the user identifier, query the corresponding user service data; The answering process according to the knowledge information corresponding to the target node includes: Response processing is performed according to the user service data and the knowledge information corresponding to the target node.

4. The method according to claim 1, characterized in that The business consulting request carries a request text; the step of determining a target knowledge topic corresponding to the business consulting request includes: Performing text similarity calculation processing on the request text and each knowledge topic in the preset knowledge topic set to obtain the text similarity of each knowledge topic; According to the text similarity of each of the knowledge topics, a target knowledge topic of the business consulting request is determined.

5. The method according to claim 1, characterized in that Also includes: Receiving a node selection input in the target knowledge image; Determine the corresponding node to be processed according to the node selection input; The target knowledge image and the knowledge information corresponding to the node to be processed are displayed in association.

6. The method according to claim 1, characterized in that The target knowledge image includes text nodes and image nodes; The preset knowledge topic set includes a target knowledge topic and a first knowledge topic, and the knowledge information corresponding to the text node includes a sub-topic text of the target knowledge topic and a sub-topic text of the first knowledge topic; The knowledge information corresponding to the image node includes the sub-topic text of the target knowledge topic and an associated knowledge image describing the sub-process of the business consulting process corresponding to the target knowledge topic.

7. The method according to claim 6, characterized in that The answering process according to the knowledge information corresponding to the target node includes: In the case where the target node is the text node, generating first response information according to the sub-topic text of the target knowledge topic and the sub-topic text of the first knowledge topic; Perform response processing according to the first response information.

8. A response device, characterized in that: include: A determination unit, used to determine a target knowledge topic corresponding to a business consulting request; The determination unit is further configured to determine, according to the target knowledge subject, a target knowledge image corresponding to the target knowledge subject in a preset image set; the target knowledge image includes a plurality of nodes, each of which corresponds to at least one piece of knowledge information; An acquiring unit, configured to acquire additional consulting information, wherein the additional consulting information is supplementary information of the business consulting request; A response unit is used to determine a target node in the target knowledge image according to the additional consultation information, and perform response processing according to the knowledge information corresponding to the target node.

9. An electronic device, characterized in that: The device comprises: A processor; and a memory configured to store computer executable instructions, which, when executed, cause the processor to execute the response method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store computer-executable instructions, and the computer-executable instructions, when executed by a processor, implement the response method according to any one of claims 1 to 7.

Images