Information processing method and device, equipment, medium and product

Abstract

The invention provides an information processing method and apparatus, a device, a medium and a product. The method comprises the steps of obtaining a first text sent by a first user in a first question and answer dialogue; sending a first prompt word including the target question to the first language model, and receiving first recommendation information returned by the first language model; when the second language model is utilized to determine that the first question-answer dialogue is related to the target object, determining second recommendation information; and presenting the first recommendation information and the second recommendation information to the second user. In the method, the natural language processing capability of the language model is utilized to perform information recommendation for the technical support personnel, the technical support personnel can refer to the automatically generated recommendation information, quickly reply to the user and improve the efficiency of solving the technical problem, and when the method is applied to the IT operation and maintenance scene, the IT operation and maintenance processing efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an information processing method, device, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] Technical support personnel, also known as technical support engineers, technical support consultants, etc., are usually used to help users solve technical problems. In different scenarios, technical support personnel can provide different technical support services. For example, when users use chemical systems to produce chemical products, if they encounter problems related to chemical equipment, technical support personnel can provide technical support services related to chemical technology. For another example, when users use information technology (IT) infrastructure, if they encounter problems related to IT operations and maintenance, technical support personnel can provide technical support services related to IT operations and maintenance.

[0003] Typically, users and technical support personnel communicate through question-and-answer dialogues, that is, users seek solutions remotely, and technical support personnel provide technical support services remotely.

[0004] In the related art, when a technical support staff receives a message text from a user, it is necessary to combine the historical conversation text with the user, refine the user's technical problem, and query information related to the user and the technical problem from multiple different systems in order to reply to the user's technical problem. At the same time, the technical support staff often need to handle the technical problems of multiple users at the same time, which results in low efficiency and poor flexibility in solving technical problems. Summary of the invention

[0005] The present application provides an information processing method. The method uses the natural language processing capability of a language model to recommend information to technical support personnel, so that technical support personnel can quickly reply to users and improve the efficiency of solving technical problems. The present application also provides a device, an electronic device, a computer-readable storage medium, and a computer program product corresponding to the above method.

[0006] In a first aspect, the present application provides an information processing method, the method comprising:

[0007] Obtaining a first text sent by a first user in a first question-and-answer dialogue; wherein the first question-and-answer dialogue is a dialogue between the first user and a second user asking questions, and the first text is used to ask questions regarding a target question;

[0008] Sending a first prompt word including the target question to a first language model, and receiving first recommendation information returned by the first language model; wherein the first recommendation information is used to assist the second user in replying to the target question;

[0009] When it is determined by using the second language model that the first question-and-answer dialogue is related to the target object, second recommendation information is determined; wherein the second recommendation information includes attribute information of the target object;

[0010] The first recommendation information and the second recommendation information are presented to the second user.

[0011] In a second aspect, the present application provides an information processing device, the device comprising:

[0012] an acquisition module, configured to acquire a first text sent by a first user in a first question-and-answer dialogue; wherein the first question-and-answer dialogue is a dialogue between the first user and a second user asking questions, and the first text is used to ask questions regarding a target question;

[0013] A first determination module is configured to send a first prompt word including the target question to a first language model, and receive first recommendation information returned by the first language model; wherein the first recommendation information is used to assist the second user in replying to the target question;

[0014] A second determination module is configured to determine second recommendation information when determining that the first question-answer dialogue is related to the target object using a second language model; wherein the second recommendation information includes attribute information of the target object;

[0015] A presentation module is used to present the first recommendation information and the second recommendation information to the second user.

[0016] In a third aspect, the present application provides an electronic device, the electronic device comprising a processor and a memory. The processor and the memory communicate with each other. The processor is used to execute instructions stored in the memory, so that the electronic device performs the information processing method in the first aspect or any implementation of the first aspect.

[0017] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions instruct an electronic device to execute the information processing method described in the first aspect or any one of the implementations of the first aspect.

[0018] In a fifth aspect, the present application provides a computer program product comprising instructions, which, when executed on an electronic device, enables the electronic device to execute the information processing method described in the first aspect or any one of the implementations of the first aspect.

[0019] Based on the implementations provided in the above aspects, this application can also be further combined to provide more implementations.

[0020] It can be seen from the above technical solutions that this application has the following advantages:

[0021] The present application provides an information processing method, which first obtains a first text sent by a first user in a first question-and-answer dialogue, wherein the first question-and-answer dialogue is a dialogue between the first user and a second user in which the first text is used to ask questions regarding a target question, then sends a first prompt word including the target question to a first language model, receives first recommendation information returned by the first language model, wherein the first recommendation information is used to assist the second user in replying to the target question, determines second recommendation information when the first question-and-answer dialogue is determined to be related to a target object using a second language model, and the second recommendation information includes attribute information of the target object, and presents the first recommendation information and the second recommendation information to the second user.

[0022] In this method, for a scenario where a user (i.e., the first user) and a technical support staff (i.e., the second user) have a conversation related to technical support, the natural language processing capability of the language model is used to recommend information to the technical support staff. On the one hand, when the text sent by the user is used to ask a question, the language model is used to generate recommended information for assisting the reply. On the other hand, the language model is used to identify the target object related to the text sent by the user, and generate recommended information including attributes related to the target object. In this way, there is no need for technical support staff to manually refine questions and query information. Technical support staff can refer to the automatically generated recommended information, quickly reply to users, and improve the efficiency of solving technical problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical method of the embodiments of the present application, the drawings required for use in the embodiments are briefly introduced below.

[0024] Figure 1 A schematic diagram of the architecture of an information processing system provided in an embodiment of the present application;

[0025] Figure 2 A flowchart of an information processing method provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of a process for obtaining a target knowledge text provided in an embodiment of the present application;

[0027] Figure 4 A schematic diagram of a process for determining a target object provided in an embodiment of the present application;

[0028] Figure 5A schematic diagram of a process for updating the name of a target object provided in an embodiment of the present application;

[0029] Figure 6 A schematic diagram of a question-and-answer processing page provided in an embodiment of the present application;

[0030] Figure 7 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application;

[0031] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] The terms "first" and "second" in the embodiments of the present application are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0033] First, some technical terms and application scenarios involved in the embodiments of this application are introduced.

[0034] Technical support personnel, also known as technical support engineers, technical support consultants, etc., are usually used to help users solve technical problems. In different scenarios, technical support personnel can provide different technical support services. For example, when users use chemical systems to produce chemical products, if they encounter problems related to chemical equipment, technical support personnel can provide technical support services related to chemical technology. For another example, when users use information technology (IT) infrastructure, if they encounter problems related to IT operations and maintenance, technical support personnel can provide technical support services related to IT operations and maintenance.

[0035] Taking the technical support personnel as IT operation and maintenance personnel as an example, in some enterprises, considering the large number of employees, they are usually equipped with information technology (IT) operation and maintenance personnel. IT operation and maintenance personnel, also known as IT operation and maintenance engineers, are responsible for matters related to IT operation and maintenance in the enterprise. For example, IT operation and maintenance personnel can be responsible for hardware equipment (such as personal computers (PCs), monitors, peripherals, etc.), software applications and network status troubleshooting, hardware equipment distribution and replacement, software application installation and update, etc.

[0036] Typically, users communicate with technical support personnel in the form of question-and-answer dialogues, that is, users seek solutions remotely, and technical support personnel provide technical support services remotely. In some examples, enterprises use office automation (OA) systems to assist in office work, wherein office automation can realize the automated processing of office affairs, greatly improving the work efficiency of individual or group office affairs. OA systems typically include a plurality of functional modules for providing different functions, and technical support can be used as a functional module of the OA system to provide technical support services. Users can trigger instant messaging with technical support personnel in the technical support module, ask technical questions, and receive replies from technical support personnel. In addition, technical support sites are also usually set up in the office area of ​​the enterprise. When technical problems need to be handled offline, such as the need to issue or replace hardware equipment, or the need to check or repair hardware equipment offline, users can also handle offline with technical support personnel at the technical support site.

[0037] In the related art, since users and technical support personnel conduct question-and-answer conversations in the form of instant messaging messages, the message texts sent by users are usually fragmented and will not fully explain the technical problem in one message text. Instead, the technical support personnel will ask questions by sending multiple message texts. Therefore, when the technical support personnel receive the user's message text, they need to combine the historical conversation texts with the user to refine the user's technical problem.

[0038] In addition, IT operation and maintenance personnel need to query information related to the user and the technical problem from multiple different systems in order to respond to the user's technical problem. Taking the IT operation and maintenance scenario as an example, a user consults about an IT operation and maintenance problem. When the IT operation and maintenance problem involves "the computer cannot connect to the network", the IT operation and maintenance personnel need to query the PC information used by the user from the equipment asset library, query the work area where the user is located from the employee database, in order to determine the office network used by the user, and then query the IT operation and maintenance knowledge base for solutions to the problem of "cannot connect to the network", and then integrate the information queried from different systems to respond to the user's IT operation and maintenance problem.

[0039] At the same time, technical support personnel often need to deal with technical problems of multiple users at the same time. Handling technical problems for each user involves multiple rounds of dialogue interactions, which results in low efficiency and poor flexibility in solving technical problems.

[0040] In view of this, the present application provides an information processing method, which first obtains a first text sent by a first user in a first question-and-answer dialogue, wherein the first question-and-answer dialogue is a dialogue between the first user and the second user in which a question is answered, and the first text is used to ask questions regarding a target question. Next, a first prompt word including the target question is sent to a first language model, and first recommendation information returned by the first language model is received, wherein the first recommendation information is used to assist the second user in replying to the target question. When the second language model is used to determine that the first question-and-answer dialogue is related to the target object, second recommendation information is determined, the second recommendation information including attribute information of the target object, and the first recommendation information and the second recommendation information are presented to the second user.

[0041] In this method, for a scenario where a user (i.e., the first user) and a technical support staff (i.e., the second user) have a conversation related to technical support, the natural language processing capability of the language model is used to recommend information to the technical support staff. On the one hand, when the text sent by the user is used to ask a question, the language model is used to generate recommended information for assisting the reply. On the other hand, the language model is used to identify the target object related to the text sent by the user, and generate recommended information including attributes related to the target object. In this way, there is no need for technical support staff to manually refine questions and query information. Technical support staff can refer to the automatically generated recommended information, quickly reply to users, and improve the efficiency of solving technical problems.

[0042] To facilitate understanding of the technical solution provided by the embodiments of the present application, it will be described below with reference to the accompanying drawings.

[0043] See also Figure 1 An architectural diagram of an information processing system is shown. In an embodiment of the present application, the information processing system can be used to execute an information processing method. Specifically, the information processing system is used to perform question analysis and skill identification on a first text sent by a first user in a first question-and-answer dialogue. During the question analysis process, the information processing system identifies a target question from the first text, analyzes the target question in combination with a first language model, and then generates first recommendation information. During the skill identification process, the information processing system identifies a target object related to the first question-and-answer dialogue with the help of a second language model, and then generates second recommendation information including attribute information of the target object. In this way, technical support personnel can refer to the first recommendation information and the second recommendation information, and make quick and targeted responses to the first text in the first question-and-answer dialogue, thereby improving the efficiency of solving technical problems.

[0044] See also Figure 2 The flowchart of an information processing method provided by an embodiment of the present application is shown, and the method specifically includes:

[0045] S201: Obtain a first text sent by a first user in a first question-and-answer dialogue.

[0046] The first question-and-answer dialogue is a dialogue between the first user and the second user asking questions. For example, the first question-and-answer dialogue can be a dialogue between the first user and the second user about technical issues. The first user can be understood as a user who encounters a technical problem and needs to consult, and the second user can be understood as a user who solves the technical problem and provides technical support, that is, the second user can be a technical support staff.

[0047] In the IT operation and maintenance scenario, the second user may be an IT operation and maintenance personnel, and the first text may be related to IT operation and maintenance. In other words, the first text may be understood as a message text sent by the first user to the IT operation and maintenance personnel for consulting on issues related to IT operation and maintenance.

[0048] In some embodiments, the technical support personnel provide technical support services through the technical support module in the OA system. In other words, the information processing system can be deployed in the technical support module in the OA system. In this case, the first text can be an instant messaging message sent by a first user (e.g., an employee other than the technical support personnel in the enterprise) using the OA system to the technical support personnel in the technical support module.

[0049] Considering that the first user and the second user conduct a conversation in the form of instant messaging messages, the message text sent by the first user is usually fragmented, and a complete sentence may be sent through multiple message texts. Therefore, in an embodiment of the present application, question analysis is performed on the message text sent by the first user in the first question-and-answer conversation to determine whether the message text asks questions regarding the target question. Only when the message text is used to ask questions regarding the target question, the message text is determined as the first text.

[0050] Among them, the target question can be understood as the technical question raised by the first user through the first text. In other words, when the semantic representation of the first text indicates that the first user has raised a new technical question, the first text can be used to ask questions regarding the target question, and the "new technical question" is the target question.

[0051] For example, in an IT operation and maintenance scenario, the target question may include consulting questions related to IT operation and maintenance. When the first text is "Hello" or "Can you help me take a look?", the first text does not raise a new technical question, and the first text is not used to ask questions about the target question. When the first text is "My computer cannot open application A", the first text raises a new technical question, and the first text is used to ask questions about the target question. In this case, the target question may be "The computer cannot open application A."

[0052] In some possible implementations, the process of determining whether the message text sent by the first user asks a question regarding the target question can be accomplished with the aid of a language model. Specifically, during the first question-and-answer dialogue, the message text sent by the first user is received, a second prompt word is generated, the second prompt word is sent to a third language model, an analysis result returned by the language model is received, and in response to the analysis result characterizing that the message text is used to ask a question regarding the target question, the message text is determined to be the first text.

[0053] The third language model can be understood as a language model used to determine whether the message text asks questions about the target question. The third language model has natural language processing capabilities, can understand the meaning of natural language, and process different types of natural language tasks. For example, the third language model can be a deep learning model trained using text data.

[0054] That is to say, considering that the message text is described by natural language and represents the natural language content, the natural language processing capability of the third language model is used to determine whether the message text asks questions regarding the target question.

[0055] The third language model determines whether the message text asks questions about the target question based on prompt learning. Among them, prompt words can be used to guide the language model to perform specific outputs in generative tasks (such as text generation tasks, question-answering tasks, and dialogue tasks). By configuring prompt words, the language model can understand the background and requirements of the task, so that the language model can handle different types of natural language processing tasks without retraining the language model, increasing the scalability and flexibility of the language model.

[0056] In the embodiment of the present application, the second prompt word may include: message text, historical dialogue text of the first question-and-answer dialogue, historical questions in the first question-and-answer dialogue, and information indicating relevant questions for analyzing the message text. The historical questions in the first question-and-answer dialogue can be understood as technical questions that have been raised in the first question-and-answer dialogue.

[0057] Since the second prompt word includes the above information, the third language model can analyze whether the message text raises new technical issues based on the prompt capability of the second prompt word, combined with the conversation text and historical questions, and output the analysis result. In this way, each message text sent by the first user is analyzed to identify whether the first user raises a new technical issue through the message text and whether it is necessary to perform the subsequent process of generating the first recommended information.

[0058] In some embodiments, the analysis result can be output in a lightweight data exchange format (javascript object notation, JSON). Specifically, the second prompt word can be as follows:

[0059] "Please judge whether the user has raised a new question based on the chat history and historical questions. Please return it in the JSON format of {"has_new_question":, "new_questions":}.

[0060] Chat log:

[0061] User: Hello

[0062] IT Operations Engineer: Hello

[0063] User: My computer won’t turn on.

[0064] IT operation and maintenance engineer: Why not try restarting it?

[0065] User: It still doesn't work. I wonder if the power cord is broken. How do I replace it if it is broken?

[0066] Historical problem: Computer won’t turn on”

[0067] Among them, "User: It still doesn't work. I don't know if the power cord is broken. How do I replace a broken power cord?" is the message text, "User: Hello", "IT operation and maintenance engineer: Hello", "User: My computer won't turn on" and "IT operation and maintenance engineer: How about restarting it?" are the historical dialogue texts of the first question and answer dialogue, "Computer won't turn on" is the historical question of the first question and answer dialogue, and "Please determine whether the user has raised a new question based on the chat records and historical questions" is information indicating relevant issues for analyzing the message text.

[0068] The analysis results returned by the third language model can be as follows:

[0069]

[0070] Among them, "true" represents the message text used to ask questions about the target question, and "the power cord is broken" is the target question.

[0071] It should be noted that the embodiments of the present application do not limit the method for determining whether the message text asks questions regarding the target question. In other possible implementations, it is also possible to determine whether the message text asks questions regarding the target question by matching with pre-set question keywords, using a question recognition algorithm, and the like.

[0072] When the message text is used to ask a question regarding a target question, the message text may be determined as the first text.

[0073] S202: Sending a first prompt word including the target question to a first language model, and receiving first recommendation information returned by the first language model.

[0074] In the embodiment of the present application, the first recommendation information is generated with the help of the first language model. The first recommendation information can be used to assist the second user in replying to the target question, that is, the first recommendation information can be understood as reference information for the second user to reply to the target question. Since the target question usually involves technical issues, the second user can solve the target question with the help of the first recommendation information when replying to the target question. Therefore, the first recommendation information can also be understood as information related to the solution to the target question.

[0075] Similarly, the first language model can be understood as a language model used to generate the first recommendation information. The first language model has natural language processing capabilities, can understand the meaning of natural language, and process different types of natural language tasks. For example, the first language model can be a deep learning model trained using text data. In other words, considering that the target question is described in natural language and represents the natural language content, the natural language processing capabilities of the first language model are used to generate reference information that can be used to answer the target question.

[0076] The first language model can also generate the first recommendation information based on the prompt learning method. In the embodiment of the present application, the first prompt word can include the target question. In addition, the first prompt word can also include information indicating the first auxiliary information for generating the target question. In this way, the first language model can analyze the target question based on the prompt ability of the first prompt word, generate reference information for answering the target question, and output the first recommendation information.

[0077] In some embodiments, Figure 1 As shown, the first language model can also generate the first recommendation information with the help of existing knowledge text. Figure 3 A flow chart of obtaining a target knowledge text is shown, in which a target knowledge text related to a target question is obtained from a knowledge base, and then a first prompt word is generated. The first prompt word may include: a target question, a target knowledge text, and information indicating generation of first recommendation information. The first prompt word is sent to a first language model, and the first recommendation information returned by the first language model is received.

[0078] Among them, the knowledge base stores knowledge texts. For example, in an IT operation and maintenance scenario, the knowledge base can be an IT operation and maintenance knowledge base. The IT operation and maintenance knowledge base can store solutions to different IT operation and maintenance problems, standard operating procedure (SOP) documents for different IT operation and maintenance problems, historical processing records of different IT operation and maintenance problems, etc.

[0079] In this way, by querying the target knowledge text related to the target question in the knowledge base and adding the target knowledge text to the first prompt word, the first language model can learn the knowledge related to the target question based on the prompt ability of the first prompt word, and then combine the knowledge related to the target question to generate the first recommendation information of the target question in a targeted manner, so that the first recommendation information is more matched with the target question.

[0080] In some embodiments, the knowledge text in the knowledge base can be stored in the form of semantic vectors. In specific implementation, the knowledge text is obtained from different data sources (such as the internal database of the OA system or an external database), the knowledge text is segmented and vectorized, and the semantic vector of the knowledge text is stored in the knowledge base.

[0081] Correspondingly, when obtaining the target knowledge text, we can first determine the semantic vector of the target question, calculate the similarity between the semantic vector of the target question and the semantic vectors of multiple knowledge texts in the knowledge base, and determine the knowledge text whose similarity meets the set conditions as the target knowledge text related to the target question.

[0082] That is to say, considering that the knowledge texts in the knowledge base are stored in the form of semantic vectors, the target text is vectorized, and the similarity between the semantic vector of the target text and the semantic vector of the knowledge text in the knowledge base is calculated, and the knowledge text that meets the set conditions (for example, the similarity is greater than the similarity threshold) is determined as the target knowledge text. In this way, the knowledge text similar to the target question is screened out from the knowledge base, thereby improving the matching degree between the first recommendation information generated by the first language model and the target question.

[0083] The embodiments of the present application do not limit the method of vectorizing the target question and the knowledge text. For example, the natural language processing capability of the language model can be used to obtain the semantic vector of the target question or the knowledge text, or the vectorization model can be used to obtain the semantic vector of the target question or the knowledge text.

[0084] S203: When it is determined by using the second language model that the first question-answer dialogue is related to the target object, second recommendation information is determined.

[0085] In the embodiment of the present application, the target object can be understood as the object involved in the first question-and-answer dialogue. Among them, the object can be understood as the entity involved in the process of solving the technical problem. For example, in the IT operation and maintenance scenario, the target object can be the target IT object, and in the chemical industry scenario, the target object can be the target chemical industry object.

[0086] In some embodiments, objects may have different object types. For example, in an IT operation and maintenance scenario, object types may include networks, hardware, software, and IT operation and maintenance sites. For example, when the object type is hardware, the target object may be a PC, a monitor, etc., and when the object type is an IT operation and maintenance site, the target object may be an IT operation and maintenance site A, an IT operation and maintenance site B, etc. For another example, in a chemical industry scenario, object types may include heat exchange equipment, separation equipment, and fluid transport equipment. For example, when the object type is a heat exchange equipment, the target object may be a shell and tube heat exchanger, a plate heat exchanger, when the object type is a separation equipment, the target object may be a distillation tower, an extraction tower, etc., and when the object type is a fluid transport equipment, the target object may be a centrifugal pump, a reciprocating pump, etc.

[0087] In this way, the process of determining the target object related to the first question-and-answer dialogue can be divided into two parts: identifying the object type involved in the first question-and-answer dialogue and extracting the target object under the object type. Among them, identifying the object type involved in the first question-and-answer dialogue can also be called skill identification, and extracting the target object under the object type can also be called slot extraction.

[0088] In the embodiment of the present application, the process of determining the target object related to the first question-answering dialogue with the help of the second language model. Figure 4 As shown, a third prompt word is generated, the third prompt word is sent to the second language model, a recognition result returned by the second language model is received, and in response to the recognition result characterizing that the first question-and-answer dialogue is related to the target object, second recommendation information is determined.

[0089] Among them, the second language model can be understood as a language model used to determine the target object related to the first question-and-answer dialogue. The second language model has natural language processing capabilities, can understand the meaning of natural language, and process different types of natural language tasks. For example, the second language model can be a deep learning model trained using text data. That is to say, considering that the dialogue text of the first question-and-answer dialogue is described by natural language and represents the natural language content, the natural language processing capabilities of the second language model are used to determine the target object related to the first question-and-answer dialogue.

[0090] The second language model can also generate recognition results based on prompt learning. In an embodiment of the present application, the third prompt word may include: the first text, the historical dialogue text of the first question-and-answer dialogue, and information indicating the identification of the object type and the extraction of the target object under the object type from the first question-and-answer dialogue. The second language model can perform skill recognition from the first question-and-answer dialogue based on the prompt ability of the third prompt word, identify the target object type involved in the first question-and-answer dialogue, and perform slot extraction from the first question-and-answer dialogue, extract the target object under the target object type, and return the recognition result.

[0091] When the recognition result represents that the first question-and-answer dialogue is related to the target object, it indicates that the first user and the second user are having a dialogue regarding the target object. Therefore, the second user is likely to need the attribute information of the target object when solving a technical problem. In an embodiment of the present application, the second recommendation information includes the attribute information of the target object. The second user can use the attribute information of the target object as reference information to give a comprehensive and detailed reply to the first text.

[0092] In some embodiments, the recognition result may be output in JSON format. For example, the recognition result may be as follows:

[0093] {"skill_set_1_dedtected":true,

[0094] "key_parameters":

[0095] {"parameter_1":"abc",

[0096] "parameter_2":null}}

[0097] Among them, the target object is "abc".

[0098] It should be noted that, in the embodiment of the present application, the first language model, the second language model and the third language model may be different language models or the same language model, and the embodiment of the present application does not limit this.

[0099] The attribute information of the target object can be understood as information about at least one attribute related to the target object. For example, in an IT operation and maintenance scenario, when the target object is IT operation and maintenance site A, the attribute information may include the location of IT operation and maintenance site A, the name of IT operation and maintenance site A, the operating hours of IT operation and maintenance site A, IT operation and maintenance personnel information of IT operation and maintenance site A, etc.

[0100] Specifically, continue as Figure 1 As shown, since the recognition result returned by the second language model can also characterize the target object type of the target object, the attribute information of the target object is obtained from the object system of the target object type, and the second recommendation information is determined according to the attribute information of the target object.

[0101] That is to say, considering that the attribute information of different objects is usually stored in different object systems and the information is relatively scattered, therefore, by identifying the target object type involved in the first question-and-answer dialogue, the attribute information of the target object is obtained from the object system corresponding to the target object type, so as to smoothly generate the second recommendation information.

[0102] In some embodiments, considering that there are many object systems, obtaining the attribute information of objects of different object types from different object systems requires obtaining the permissions of different object systems, which is costly and error-prone. Therefore, a general system can be set up to integrate the information from different object systems. In this way, the attribute information of the target object can be obtained by calling the information acquisition interface of the general system.

[0103] Furthermore, the first question-and-answer dialogue between the first user and the second user usually includes a lot of spoken words or general names of objects, and the spoken words or general names of objects are usually inconsistent with the names of the target objects stored in the object system. For example, in an IT operation and maintenance scenario, the first text is "My computer won't turn on", where the target object extracted by the second language model is "computer", which is a general name for PCs. When it is necessary to query the attribute information of the target object from the object system of the target object type (for example, a hardware asset library), it is necessary to search with the standard name of the target object. Therefore, before determining the second recommended information, it is necessary to modify the name of the target object.

[0104] Specifically, the name of the target object is standardized and the name of the target object is updated. The name standardization process can be understood as a process of changing the name of the target object returned by the second language model to the name of the target object stored in the object system of the target object type. In this way, by converting the target object into a standard name, it is ensured that the attribute information can be obtained smoothly.

[0105] The embodiment of the present application supports updating the name of the target IT object in different ways, such as Figure 5 As shown, in some embodiments, at least one candidate object associated with the first user is obtained, and according to the similarity between the name of the target object and the name of the at least one candidate object, the name of the target object is updated to the name of the target candidate object.

[0106] In other embodiments, a target object type of a target object is determined, at least one candidate object corresponding to the target object type is obtained from a standardized name library, and based on the similarity between the name of the target object and the name of at least one candidate object, the name of the target object is updated to the name of the target candidate object.

[0107] The target candidate object is one of the at least one candidate object.

[0108] In an embodiment of the present application, a candidate object is generally an object with a standardized name. In some possible implementations, the candidate object is associated with the first user, that is, the account identifier of the first user is used as an index to determine the candidate object of the first user under the target object type from the object system of the target object type. For example, in an IT operation and maintenance scenario, the target object determined by the second language model is "computer". In the object system of the target object type (for example, a hardware asset library), the account identifier of the first user is used as an index to search for multiple hardware devices of the first user, such as a PC of brand A, a monitor of brand B, and a keyboard of brand C. The multiple hardware devices of the first user are candidate objects. Since the candidate object is determined from the object system of the target object type, the candidate object has a standardized name.

[0109] In some other possible implementations, the candidate object is associated with the target object type, that is, the target object type is used as an index to determine the candidate object under the target object type from the standardized name library. For example, in an IT operation and maintenance scenario, the target object determined by the second language model is "browser". In the standardized name library, the target object type (for example, software) is used as an index to search for multiple objects belonging to the software, such as xx browser, yy office software, and zz compiler. The multiple objects belonging to the software are candidate objects. Since the candidate objects are determined from the standardized name library, the candidate objects have standardized names.

[0110] After the candidate object is determined, the target object name is updated by the similarity between the name of the target object and the name of the candidate object. In specific implementation, the name of the target object can be updated to the name of the target candidate object whose similarity meets the set conditions (for example, the largest similarity), so that the name of the target object meets the standardization requirements.

[0111] In this way, the target object with the standardized name is used as an index, and the attribute information of the target object can be queried from the object system of the target object type, thereby generating the second recommendation information. Since the second recommendation information includes the attribute information of the target object, and the target object is related to the first question-and-answer dialogue, the second recommendation information can provide detailed information of the target object when the first user only mentions the target object.

[0112] S204: Present the first recommendation information and the second recommendation information to the second user.

[0113] Since the first recommendation information provides reference information for replying to the target question and the second recommendation information provides detailed information of the target object, by presenting the first recommendation information and the second recommendation information to the second user (i.e., the technical support personnel), the second user can quickly obtain information related to solving technical problems, and without the need for the second user to manually search for various information, a reply to the first text can be achieved by combining the first recommendation information and the second recommendation information.

[0114] For example, in an IT operation and maintenance scenario, when the first recommendation information indicates a solution to a target problem, the second user can quickly provide a solution to the target problem of the first user by copying the first recommendation information and sending the first recommendation information. For another example, when the second recommendation information includes attribute information of IT operation and maintenance site A, the second user can quickly reply to the user's first text by copying the second recommendation information and, based on the second recommendation information, adding content for informing the first user to go to the IT operation and maintenance site for offline services.

[0115] In a specific implementation, a question and answer processing page is presented in the client of the second user; wherein the question and answer processing page includes a recommended information presentation area, in which the first recommended information and the second recommended information are presented.

[0116] In addition, the question and answer processing page also includes a message list area and a conversation area. In the message list area, conversation information of at least one question and answer conversation is presented, and in the conversation area, the first question and answer conversation is presented.

[0117] The question-and-answer processing page can be understood as the page used by the second user to handle technical issues. For example, in an IT operation and maintenance scenario, the question-and-answer processing page can be the front-end page of the technical support module. By splitting the question-and-answer processing page into multiple different areas, the second user can view conversation information with different first users in the message list area, view conversation text with a certain first user in the conversation area, and view automatically generated recommended information in the recommended information presentation area.

[0118] In this way, when the technical support personnel are solving technical problems, recommended information is presented in real time for the message text sent by the first user, providing support for the technical support personnel to reply to the first user's message and solve technical problems.

[0119] like Figure 6As shown, in the IT operation and maintenance scenario, the question and answer processing page 60 includes a message list area 601, a conversation area 602, and a recommended information presentation area 603. In the message list area 601, the conversation information between the second user and multiple different first users (for example, AAA, BBB, CCC, and DDD) is presented in the form of a message list, and in the conversation area 602, the conversation text between the second user and DDD is presented. At the same time, the second user can also reply to the message in the conversation area 602.

[0120] In the recommendation information presentation area 603, the first recommendation information 6032 and the second recommendation information 6031 are presented. The first text is "My abc is not working", the second recommendation information 6031 includes the attribute information of the abc software in the first user's device, and the first recommendation information 6032 includes the solution to the abc software expiration. In this way, the second user can quickly obtain the attribute information of the target object and the reference information for answering the target problem by viewing the first recommendation information and the second recommendation information, thereby improving the efficiency of IT operation and maintenance processing.

[0121] In this method, for a scenario where a user (i.e., the first user) and a technical support staff (i.e., the second user) have a conversation related to technical support, the natural language processing capability of the language model is used to recommend information to the technical support staff. On the one hand, when the text sent by the user is used to ask a question, the language model is used to generate recommended information for assisting the reply. On the other hand, the language model is used to identify the target object related to the text sent by the user, and generate recommended information including attributes related to the target object. In this way, there is no need for technical support staff to manually refine questions and query information. Technical support staff can refer to the automatically generated recommended information, quickly reply to users, and improve the efficiency of solving technical problems.

[0122] Combination of the above Figures 1 to 6 The information processing method provided in the embodiment of the present application is introduced in detail. The device and equipment provided in the embodiment of the present application will be introduced in conjunction with the accompanying drawings.

[0123] See also Figure 7 The structural diagram of the information processing device shown in FIG. 70 includes:

[0124] An acquisition module 701 is used to acquire a first text sent by a first user in a first question-and-answer dialogue; wherein the first question-and-answer dialogue is a dialogue between the first user and a second user asking questions, and the first text is used to ask questions regarding a target question;

[0125] A first determination module 702 is configured to send a first prompt word including the target question to a first language model, and receive first recommendation information returned by the first language model; wherein the first recommendation information is used to assist the second user in replying to the target question;

[0126] A second determination module 703 is configured to determine second recommendation information when determining that the first question-answer dialogue is related to the target object using a second language model; wherein the second recommendation information includes attribute information of the target object;

[0127] The presentation module 704 is configured to present the first recommendation information and the second recommendation information to the second user.

[0128] In some possible implementations, the first question-and-answer dialogue is related to information technology IT operation and maintenance, the target questions include consulting questions related to IT operation and maintenance, and the target objects include target IT objects.

[0129] In some possible implementations, the acquisition module 701 is specifically used to:

[0130] During the first question-and-answer dialogue, receiving a message text sent by the first user;

[0131] Generate a second prompt word; wherein the second prompt word includes: the message text, the historical dialogue text of the first question-and-answer dialogue, the historical questions in the first question-and-answer dialogue, and information indicating relevant questions for analyzing the message text;

[0132] Sending the second prompt word to a third language model, and receiving an analysis result returned by the third language model;

[0133] In response to the analysis result characterizing that the message text is used to ask a question regarding a target question, the message text is determined as a first text.

[0134] In some possible implementations, the first determining module 702 is further configured to:

[0135] Acquire target knowledge text related to the target problem from the knowledge base;

[0136] The first determining module 702 is specifically configured to:

[0137] Generate a first prompt word; wherein the first prompt word includes: the target question, the target knowledge text and information indicating the generation of the first recommended information;

[0138] The first prompt word is sent to a first language model, and first recommendation information returned by the first language model is received.

[0139] In some possible implementations, the first determining module 702 is specifically configured to:

[0140] Determining a semantic vector of the target question;

[0141] Calculating the similarity between the semantic vector of the target question and the semantic vectors of multiple knowledge texts in the knowledge base;

[0142] The knowledge text whose similarity meets the set conditions is determined as the target knowledge text related to the target problem.

[0143] In some possible implementations, the second determining module 703 is specifically configured to:

[0144] Generate a third prompt word; wherein the third prompt word includes: the first text, the historical dialogue text of the first question-and-answer dialogue, and information indicating the identification of an object type and the extraction of a target object under the object type from the first question-and-answer dialogue;

[0145] Sending the third prompt word to the second language model, and receiving a recognition result returned by the second language model;

[0146] In response to the recognition result indicating that the first question-and-answer dialogue is related to the target object, second recommendation information is determined.

[0147] In some possible implementations, the recognition result further represents the target object type of the target object; and the second determination module 703 is specifically configured to:

[0148] Acquiring attribute information of the target object from an object system of the target object type;

[0149] Determine second recommendation information according to the attribute information of the target object.

[0150] In some possible implementations, the second determining module 703 is further configured to:

[0151] Perform name standardization processing on the target object and update the name of the target object.

[0152] In some possible implementations, the second determining module 703 is specifically configured to:

[0153] Acquire at least one candidate object associated with the first user;

[0154] According to the similarity between the name of the target object and the name of the at least one candidate object, the name of the target object is updated to the name of the target candidate object; wherein the target candidate object is one of the at least one candidate object.

[0155] In some possible implementations, the second determining module 703 is specifically configured to:

[0156] Determining a target object type of the target object;

[0157] Acquire at least one candidate object under the target object type from a standardized name library;

[0158] According to the similarity between the name of the target object and the name of the at least one candidate object, the name of the target object is updated to the name of the target candidate object; wherein the target candidate object is one of the at least one candidate object.

[0159] In some possible implementations, the presenting module 704 is specifically configured to:

[0160] In the client of the second user, a question and answer processing page is presented; wherein the question and answer processing page includes a recommended information presentation area;

[0161] In the recommendation information presentation area, the first recommendation information and the second recommendation information are presented.

[0162] In some possible implementations, the question-and-answer processing page further includes a message list area and a conversation area; the presentation module 704 is further configured to:

[0163] In the message list area, presenting dialogue information of at least one question-and-answer dialogue;

[0164] In the dialogue area, the first question-and-answer dialogue is presented.

[0165] The information processing device 70 according to the embodiment of the present application may correspond to executing the method described in the embodiment of the present application, and the above and other operations and / or functions of each module / unit of the information processing device 70 are respectively to implement Figure 2 For the sake of brevity, the corresponding processes of each method in the illustrated embodiment are not described in detail here.

[0166] The present application also provides an electronic device. The electronic device is specifically used to implement Figure 7 The functions of the information processing device 70 in the illustrated embodiment.

[0167] Figure 8 A schematic diagram of the structure of an electronic device 800 is provided. Figure 8 As shown, the electronic device 400 includes a bus 801, a processor 802, a communication interface 803 and a memory 804. The processor 802, the memory 804 and the communication interface 803 communicate with each other via the bus 801.

[0168] The bus 801 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0169] The processor 802 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0170] The communication interface 803 is used for communicating with the outside. For example, the communication interface 803 can be used for communicating with a terminal.

[0171] The memory 804 may include a volatile memory, such as a random access memory (RAM). The memory 804 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0172] The memory 804 stores executable codes, and the processor 802 executes the executable codes to perform the aforementioned information processing method.

[0173] Specifically, in implementing Figure 7 In the case of the embodiment shown, and Figure 7 When each module or unit of the information processing device 70 described in the embodiment is implemented by software, the execution Figure 7 The software or program code required for the functions of each module / unit in the system may be partially or completely stored in the memory 804. The processor 802 executes the program code corresponding to each unit stored in the memory 804 to perform the aforementioned information processing method.

[0174] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute the above-mentioned information processing method applied to the information processing device 70.

[0175] The embodiment of the present application further provides a computer program product, which includes one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part.

[0176] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer or data center to another website, computer or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0177] When the computer program product is executed by a computer, the computer executes any of the aforementioned information processing methods. The computer program product may be a software installation package, and when any of the aforementioned information processing methods is needed, the computer program product may be downloaded and executed on a computer.

[0178] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to each embodiment of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0179] The units involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the unit / module does not, in some cases, constitute a limitation on the unit itself.

[0180] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0181] In the context of the present application embodiment, machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable medium can include but is not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0182] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0183] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0184] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0185] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0186] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An information processing method, characterized in that: The method comprises: Obtaining a first text sent by a first user in a first question-and-answer dialogue; wherein the first question-and-answer dialogue is a dialogue between the first user and a second user asking questions, and the first text is used to ask questions regarding a target question; Sending a first prompt word including the target question to a first language model, and receiving first recommendation information returned by the first language model; wherein the first recommendation information is used to assist the second user in replying to the target question; When it is determined by using the second language model that the first question-and-answer dialogue is related to the target object, second recommendation information is determined; wherein the second recommendation information includes attribute information of the target object; The first recommendation information and the second recommendation information are presented to the second user.

2. The method according to claim 1, characterized in that The first question-and-answer dialogue is related to information technology IT operation and maintenance, the target questions include consulting questions related to IT operation and maintenance, and the target objects include target IT objects.

3. The method according to claim 1, characterized in that The obtaining of the first text sent by the first user in the first question-and-answer dialogue includes: During the first question-and-answer dialogue, receiving a message text sent by the first user; Generate a second prompt word; wherein the second prompt word includes: the message text, the historical dialogue text of the first question-and-answer dialogue, the historical questions in the first question-and-answer dialogue, and information indicating relevant questions for analyzing the message text; Sending the second prompt word to a third language model, and receiving an analysis result returned by the third language model; In response to the analysis result characterizing that the message text is used to ask a question regarding a target question, the message text is determined as a first text.

4. The method according to claim 1, characterized in that The method further comprises: Acquire target knowledge text related to the target problem from the knowledge base; The step of sending the first prompt word including the target question to the first language model and receiving the first recommendation information returned by the first language model includes: Generate a first prompt word; wherein the first prompt word includes: the target question, the target knowledge text and information indicating the generation of the first recommended information; The first prompt word is sent to a first language model, and first recommendation information returned by the first language model is received.

5. The method according to claim 4, characterized in that The step of acquiring target knowledge text related to the target problem from the knowledge base includes: Determining a semantic vector of the target question; Calculating the similarity between the semantic vector of the target question and the semantic vectors of multiple knowledge texts in the knowledge base; The knowledge text whose similarity meets the set conditions is determined as the target knowledge text related to the target problem.

6. The method according to claim 1, characterized in that When determining that the first question-and-answer dialogue is related to the target object by using the second language model, determining the second recommendation information includes: Generate a third prompt word; wherein the third prompt word includes: the first text, the historical dialogue text of the first question-and-answer dialogue, and information indicating the identification of an object type and the extraction of a target object under the object type from the first question-and-answer dialogue; Sending the third prompt word to the second language model, and receiving a recognition result returned by the second language model; In response to the recognition result indicating that the first question-and-answer dialogue is related to the target object, second recommendation information is determined.

7. The method according to claim 6, characterized in that The recognition result also represents the target object type of the target object; and determining the second recommendation information includes: Acquiring attribute information of the target object from an object system of the target object type; Determine second recommendation information according to the attribute information of the target object.

8. The method according to claim 1, characterized in that When it is determined by using the second language model that the first question-answer dialogue is related to the target object, before determining the second recommendation information, the method further includes: Perform name standardization processing on the target object and update the name of the target object.

9. The method according to claim 8, characterized in that The step of performing name standardization on the target object and updating the name of the target object includes: Acquire at least one candidate object associated with the first user; According to the similarity between the name of the target object and the name of the at least one candidate object, the name of the target object is updated to the name of the target candidate object; wherein the target candidate object is one of the at least one candidate object.

10. The method according to claim 8, characterized in that The step of performing name standardization on the target object and updating the name of the target object includes: Determining a target object type of the target object; Acquire at least one candidate object under the target object type from a standardized name library; According to the similarity between the name of the target object and the name of the at least one candidate object, the name of the target object is updated to the name of the target candidate object; wherein the target candidate object is one of the at least one candidate object.

11. The method according to any one of claims 1 to 10, characterized in that: The presenting the first recommendation information and the second recommendation information to the second user includes: In the client of the second user, a question and answer processing page is presented; wherein the question and answer processing page includes a recommended information presentation area; In the recommendation information presentation area, the first recommendation information and the second recommendation information are presented.

12. The method according to claim 11, characterized in that The question-and-answer processing page also includes a message list area and a conversation area; the method also includes: In the message list area, presenting dialogue information of at least one question-and-answer dialogue; In the dialogue area, the first question-and-answer dialogue is presented.

13. An information processing device, characterized in that: The device comprises: an acquisition module, configured to acquire a first text sent by a first user in a first question-and-answer dialogue; wherein the first question-and-answer dialogue is a dialogue between the first user and a second user asking questions, and the first text is used to ask questions regarding a target question; A first determination module is configured to send a first prompt word including the target question to a first language model, and receive first recommendation information returned by the first language model; wherein the first recommendation information is used to assist the second user in replying to the target question; A second determination module is configured to determine second recommendation information when determining that the first question-answer dialogue is related to the target object using a second language model; wherein the second recommendation information includes attribute information of the target object; A presentation module is used to present the first recommendation information and the second recommendation information to the second user.

14. An electronic device, characterized in that: The electronic device comprises a processor and a memory; The processor is configured to execute instructions stored in the memory, so that the electronic device performs the method according to any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that: The method comprises instructions, wherein the instructions instruct an electronic device to execute the method according to any one of claims 1 to 12.

16. A computer program product, characterized in that The computer program product comprises computer readable instructions for implementing the method according to any one of claims 1 to 12.

Images