Method and apparatus for generating business quality inspection models

Abstract

This invention discloses a method and apparatus for generating a business quality inspection model. The method includes: acquiring business processing information in a target domain; analyzing the business processing information according to the business processing methods of the target domain to obtain knowledge documents; performing structured processing on the knowledge documents to obtain structured information corresponding to the knowledge documents; generating multiple quality inspection model nodes based on the structured information; and generating a quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes. Upon receiving content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content. This invention solves the technical problem in related technologies where the construction of quality inspection models for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures relies on manual operation.

Description

Technical Field

[0001] This invention relates to the field of natural language processing technology, and more specifically, to a method and apparatus for generating a business quality inspection model. Background Technology

[0002] The ability to monitor whether business personnel have followed standard procedures and responded as required when encountering specific scenarios or customer issues requires the capabilities of a quality inspection model.

[0003] Traditional manual quality inspection methods rely entirely on human inspectors reviewing historical records and comparing each item for compliance, which is extremely time-consuming, labor-intensive, and limits the number of items inspected (usually no more than 5%). With the development of artificial intelligence technology, quality inspection methods are gradually becoming more intelligent. However, most current intelligent quality inspection solutions focus on speech-to-text conversion and matching text with the quality inspection model. The construction of the quality inspection model itself still heavily relies on manual operation by sales personnel (high learning curve, difficult to learn, high labor / time costs, and unable to be updated in a timely manner to keep pace with business changes).

[0004] Regarding the issue that the construction of quality inspection models for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures in the aforementioned related technologies relies on manual operation, no effective solution has yet been proposed. Summary of the Invention

[0005] This invention provides a method and apparatus for generating a business quality inspection model, which at least solves the technical problem in the related art that the construction of a quality inspection model for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures relies on manual operation.

[0006] According to one aspect of the present invention, a method for generating a business quality inspection model is provided, comprising: acquiring business processing information of a target domain; analyzing the business processing information according to the business processing method of the target domain to obtain a knowledge document, wherein the knowledge document is used to record quality inspection information required for business quality inspection of the target domain, the quality inspection information including at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; performing structured processing on the knowledge document to obtain structured information corresponding to the knowledge document; generating multiple quality inspection model nodes based on the structured information, wherein the quality inspection model nodes are nodes in the quality inspection model used for quality inspection of content to be inspected; generating the quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes, wherein after receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content to be inspected.

[0007] Optionally, the knowledge document is subjected to structured processing to obtain structured information corresponding to the knowledge document, including at least one of the following: marking the natural language description corresponding to the knowledge document as a predetermined array, wherein the predetermined array includes at least: business-related products, business type, and revenue of the business-related products; marking the natural language description corresponding to the knowledge document as key-value pairs.

[0008] Optionally, after performing structuring processing on the knowledge document to obtain the structured information corresponding to the knowledge document, the method for generating the business quality inspection model further includes: classifying the structured information according to the knowledge document to obtain the classified structured information.

[0009] Optionally, generating multiple quality inspection model nodes based on the structured information includes: determining the title type of the categorized structured information; and generating multiple quality inspection model nodes according to each of the title types.

[0010] Optionally, after generating multiple quality inspection model nodes based on the structured information, the method for generating the business quality inspection model further includes: adding control attributes to each of the multiple quality inspection model nodes, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model nodes according to the quality inspection requirements of the content to be inspected.

[0011] Optionally, after generating the quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes, the method for generating the business quality inspection model further includes: obtaining historical session data of the target domain; and performing quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model.

[0012] Optionally, the quality inspection model is used to perform a quality inspection operation on the historical call data to test the quality inspection model, including: obtaining the quality inspection result of the quality inspection model on the historical call data; sending the quality inspection result to a predetermined terminal device to evaluate the quality inspection result through the predetermined terminal device; and obtaining the evaluation result of the quality inspection result fed back by the predetermined terminal device.

[0013] Optionally, the method for generating the business quality inspection model further includes: when the evaluation result indicates that the erroneous quality inspection information in the quality inspection result exceeds a predetermined threshold, optimizing the quality inspection model based on the quality inspection result.

[0014] According to another aspect of the present invention, a business quality inspection model generation apparatus is also provided, comprising: a first acquisition unit, configured to acquire business processing information of a target domain; an analysis unit, configured to analyze the business processing information according to the business processing method of the target domain to obtain a knowledge document, wherein the knowledge document is used to record quality inspection information required for business quality inspection of the target domain, the quality inspection information including at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; a structuring processing unit, configured to perform structuring processing on the knowledge document to obtain structured information corresponding to the knowledge document; a first generation unit, configured to generate multiple quality inspection model nodes based on the structured information, wherein the quality inspection model nodes are nodes in the quality inspection model used for quality inspection of content to be inspected; and a second generation unit, configured to generate the quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes, wherein after receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content to be inspected.

[0015] Optionally, the structured processing unit includes at least one of the following: a first tagging module, used to tag the natural language description corresponding to the knowledge document as a predetermined array, wherein the predetermined array includes at least: business-related products, business types, and revenue of the business-related products; and a second tagging module, used to tag the natural language description corresponding to the knowledge document as key-value pairs.

[0016] Optionally, the device for generating the business quality inspection model further includes a classification unit, used to classify the structured information according to the knowledge document after performing structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document, so as to obtain the classified structured information.

[0017] Optionally, the first generation unit includes: a determining module, configured to determine the title type of the categorized structured information; and a generation module, configured to generate multiple quality inspection model nodes based on each of the title types.

[0018] Optionally, the device for generating the business quality inspection model further includes an adding unit, used to add control attributes to each of the multiple quality inspection model nodes after generating multiple quality inspection model nodes based on the structured information, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model nodes according to the quality inspection requirements of the content to be inspected.

[0019] Optionally, the device for generating the service quality inspection model further includes: a second acquisition unit, configured to acquire historical session data of the target domain after generating the quality inspection model according to the service processing order corresponding to the multiple quality inspection model nodes; and a quality inspection unit, configured to perform quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model.

[0020] Optionally, the quality inspection unit includes: a first acquisition module, used to acquire the quality inspection results of the historical call data by the quality inspection model; a sending module, used to send the quality inspection results to a predetermined terminal device for evaluation by the predetermined terminal device; and a second acquisition module, used to acquire the evaluation results of the quality inspection results fed back by the predetermined terminal device.

[0021] Optionally, the apparatus for generating the business quality inspection model further includes an optimization unit, configured to optimize the quality inspection model based on the quality inspection results when the evaluation result indicates that the erroneous quality inspection information in the quality inspection results exceeds a predetermined threshold.

[0022] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes the method for generating a business quality inspection model as described in any one of the foregoing embodiments.

[0023] According to another aspect of the present invention, a processor is also provided, the processor being configured to run a program, wherein the program, when running, executes the method for generating a business quality inspection model as described in any of the preceding embodiments.

[0024] In this embodiment of the invention, business processing information of the target domain is obtained; the business processing information is analyzed according to the business processing method of the target domain to obtain a knowledge document, wherein the knowledge document is used to record the quality inspection information required for business quality inspection of the target domain, and the quality inspection information includes at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; the knowledge document is processed in a structured manner to obtain the structured information corresponding to the knowledge document; multiple quality inspection model nodes are generated based on the structured information, wherein the quality inspection model node is a node in the quality inspection model used to inspect the content to be inspected; a quality inspection model is generated according to the business processing order corresponding to the multiple quality inspection model nodes, wherein after receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content to be inspected. The technical solution provided by this invention enables the structured processing of knowledge documents, the generation of multiple quality inspection model nodes based on the structured information, and the generation of quality inspection models by sequentially generating quality inspection models. This achieves the technical effect of automating the quality inspection modeling process, lowers the barrier to entry for business personnel, and allows them to easily get started by relying on their familiar business logic. It also solves the technical problem in related technologies where the construction of quality inspection models for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures relies on manual operation. Attached Figure Description

[0025] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0026] Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of generating a business quality inspection model according to an embodiment of the present invention.

[0027] Figure 2 This is a flowchart of a method for generating a business quality inspection model according to an embodiment of the present invention;

[0028] Figure 3 This is a schematic diagram of a device for generating a business quality inspection model according to an embodiment of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] As described in the background section, business quality inspection typically involves the following steps: 1) Manual structuring of business documents: Business personnel offline review existing business documents, structuring each quality inspection standard into standard keywords or key phrases, and uploading them to the system as material for building the quality inspection model; 2) Manual expansion of keywords / key phrases: Based on the personal experience of business personnel, or through viewing and annotating a large amount of historical conversation data, keywords or key phrases expressing similar semantics are collected as expansions of standard phrases, thereby improving the coverage of the phrases; 3) Adding logical operations: If the quality inspection content involves contextual logical relationships, some logical operators need to be added on top of this (such as keyword A and keyword B, indicating that both words in a sentence must be mentioned) to form complete quality inspection points; 4) Building a quality inspection framework: According to different business scenarios (such as loss reporting scenarios, financial management scenarios, etc.), a quality inspection framework that conforms to the scenario is manually built, and the quality inspection points completed in the above steps are configured as framework nodes to build the final usable quality inspection model. This method relies heavily on manual processing. To address the shortcomings of the aforementioned related technologies, which rely on manual operation in constructing quality inspection models to monitor whether business personnel handle specific scenarios or problems in accordance with standard procedures, this invention provides a method and apparatus for generating business quality inspection models, a computer-readable storage medium, and a processor.

[0032] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0033] The methods and embodiments provided in this invention can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of generating a business quality inspection model according to an embodiment of the present invention. For example... Figure 1 As shown, a mobile terminal may include one or more ( Figure 1Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0034] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the business quality inspection model generation method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-described networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the above-described networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] According to an embodiment of the present invention, a method embodiment for generating a business quality inspection model is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0036] Figure 2 This is a flowchart of a method for generating a business quality inspection model according to an embodiment of the present invention, such as... Figure 2 As shown, the method for generating this business quality inspection model includes the following steps:

[0037] Step S202: Obtain business processing information for the target domain.

[0038] In this embodiment, for a certain field, the business processing method of that field and the standard operating procedure corresponding to that business processing method can be obtained first.

[0039] Step S204: Analyze the business processing information according to the business processing method of the target domain to obtain a knowledge document. The knowledge document is used to record the quality inspection information required for business quality inspection of the target domain. The quality inspection information includes at least: Q&A information of the business to be inspected, quality inspection reference information, and business process information.

[0040] In this embodiment, the business processing information can be analyzed according to the business processing method determined in the above steps to obtain the knowledge documents used in the business processing of this domain. These knowledge documents are used to record quality inspection information in the target domain, such as question-and-answer specifications, quality inspection specifications, and business process specifications.

[0041] Step S206: Perform structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document.

[0042] In this embodiment, the knowledge document can be structured using the aforementioned business knowledge structuring module to obtain the structured information corresponding to the knowledge document for further processing.

[0043] Step S208: Generate multiple quality inspection model nodes based on structured information. The quality inspection model node is a node in the quality inspection model used to perform quality inspection on the content to be inspected.

[0044] In this embodiment, the node generation module can be used to classify the titles of structured information, and automatically generate a quality inspection model node for each category. The content of the structured parsing (keywords, key-value pairs, triples, etc.) serves as the skeleton of the node text generation, and the content generated in this way serves as the flesh and blood of the node. It is used to perform semantic matching with the agent call text for quality inspection, replacing the traditional quality inspection steps of "manual expansion of keywords / key sentences" and "adding logical operations".

[0045] Step S210: Generate a quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes. After receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content.

[0046] In this embodiment, a model framework generation module can be used to create a model topic, select the classification information contained in the model, and automatically build a complete quality inspection model by sequentially constructing the quality inspection nodes involved according to the classification information. Here, each node is generated according to the title classification, and the parent category of the title classification is the document classification. Therefore, after selecting the document classification information, the nodes of its subordinate topic classification are automatically constructed in sequence, replacing the traditional quality inspection step of "building a quality inspection framework".

[0047] As described above, in this embodiment of the invention, business processing information of the target domain can be obtained; then, the business processing information is analyzed according to the business processing method of the target domain to obtain a knowledge document. The knowledge document is used to record the quality inspection information required for business quality inspection of the target domain. The quality inspection information includes at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; then, the knowledge document is structured to obtain the structured information corresponding to the knowledge document; then, multiple quality inspection model nodes are generated based on the structured information. The quality inspection model node is a node in the quality inspection model used to inspect the content to be inspected; and the quality inspection model is generated according to the business processing order corresponding to the multiple quality inspection model nodes. After receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content to be inspected. This achieves the purpose of structuring the knowledge document, generating multiple quality inspection model nodes according to the structured information, and generating the quality inspection model in sequence. This achieves the technical effect of automating the quality inspection modeling process, reducing the threshold for business personnel to use, and allowing them to easily get started by relying only on their familiar business logic.

[0048] Therefore, the technical solution provided by the embodiments of the present invention solves the technical problem in the related art that the construction of a quality inspection model for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures relies on manual operation.

[0049] According to the above embodiments of the present invention, the knowledge document is subjected to structured processing to obtain the structured information corresponding to the knowledge document, including at least one of the following: marking the natural language description corresponding to the knowledge document as a predetermined array, wherein the predetermined array includes at least: business-related products, business types, and revenue of business-related products; marking the natural language description corresponding to the knowledge document as key-value pairs.

[0050] In this embodiment, this can be achieved through a business knowledge structuring module. Specifically, knowledge documents can be uploaded through the business knowledge structuring module, and the module will automatically perform structured parsing and classification (with a small amount of manual annotation during the parsing process to ensure the accuracy of the parsing results). For example, natural language descriptions can be marked as triples (financial product, type, return) or key-value pairs (handling fee: 5%), and the structured information can be classified according to the document content (classified by document and title), thus replacing the traditional quality inspection step of "manual structuring of business documents".

[0051] According to the above embodiments of the present invention, after performing structuring processing on the knowledge document to obtain the structured information corresponding to the knowledge document, the method for generating the business quality inspection model further includes: classifying the structured information according to the knowledge document to obtain the classified structured information.

[0052] In this embodiment, structured information can be categorized, for example, natural language descriptions can be labeled as triples (financial product, type, return) or key-value pairs (handling fee: 5%), and the structured information can be categorized according to the document content (categorized by document and title), thereby replacing the traditional quality inspection step of "manually structuring business documents".

[0053] According to the above embodiments of the present invention, generating multiple quality inspection model nodes based on structured information may include: determining the title type of the categorized structured information; and generating multiple quality inspection model nodes according to each of the title types.

[0054] In this embodiment, the quality inspection model node can be automatically generated for each category based on the title of the structured information.

[0055] According to the above embodiments of the present invention, after generating multiple quality inspection model nodes based on structured information, the method for generating the business quality inspection model further includes: adding control attributes to each of the multiple quality inspection model nodes, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model nodes according to the quality inspection requirements of the content to be inspected.

[0056] In this embodiment, the attribute control module can be used to add control attributes to the content generated by nodes, ensuring that the content generated by nodes is controllable and can meet the personalized needs of different scenarios (such as a complaint scenario that tends to focus on negative emotions, and a financial management scenario that tends to focus on the authenticity of the script). Based on the quality inspection characteristics of agent calls, the control module includes the following adjustable attributes.

[0057] For example, sentiment: happy, like, neutral, disgusting, sad, fearful, angry; used to control the sentiment of the content generated by the node; facts: adding key-value pairs and triples; used to enhance the authenticity of the content generated by the node and avoid false information; topic: inputting a topic to make the generated content more closely related to the input topic; format: dialogue, description; choosing the dialogue format, the node generates content in the form of a one-to-one interactive dialogue between the agent and the user; choosing the description format, the node generates content as a description (such as the agent giving a longer informational introduction); style: general The format can be categorized as: serious, formal, or lively. A serious style generates more technical terms and formal language, while a lively style leans towards conversational language (e.g., agent responses in a questionnaire interview will use more formal language, while user replies will be more conversational and varied in format). Weighting: Different information can be weighted (this could be added factual content or keywords). Increasing weight increases the frequency of that information or keyword in the generated content (e.g., certain terms may be required when confirming insurance purchase). Scope: Word count and sentence count. This constrains the length of the generated content. When "Format" is set to "Description," the scope constraint is in words, limiting the word count range. When "Format" is set to "Conversation," the scope constraint is in sentences, limiting the sentence count range.

[0058] According to the above embodiments of the present invention, after generating a quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes, the method for generating the business quality inspection model may further include: obtaining historical session data of the target domain; and performing quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model.

[0059] In this embodiment, the model testing module can be used to upload session data to form a test set, which can perform real-time and batch testing on the quality inspection model. The test results can be viewed by clicking to view the result details.

[0060] According to the above embodiments of the present invention, a quality inspection model is used to perform a quality inspection operation on historical call data to test the quality inspection model, including: obtaining the quality inspection results of the quality inspection model on the historical call data; sending the quality inspection results to a predetermined terminal device for evaluation by the predetermined terminal device; and obtaining the evaluation results of the quality inspection results fed back by the predetermined terminal device.

[0061] In this embodiment, the quality inspection results of the quality inspection model on historical session data can be sent to the auditor to determine the reliability of the quality inspection results. The evaluation results of the determined quality inspection results can determine whether the quality inspection model needs to be optimized.

[0062] According to the above embodiments of the present invention, the method for generating the business quality inspection model further includes: optimizing the quality inspection model based on the quality inspection results when the evaluation results indicate that the erroneous quality inspection information in the quality inspection results exceeds a predetermined threshold.

[0063] In this embodiment, semi-automatic verification can be performed based on test results, which include: matching and non-matching. For non-matching parts, manual review can be used to confirm which of the following categories the problem falls into: duplicate statements, disjointed content, or factual errors. Duplicate statements: For statements marked as duplicated, the system can automatically optimize the generated content by deduplication, reducing semantic similarity between generated statements and making the generated content more concise. Disjointed content: For statements marked as disjointed, the system can automatically correct its attribute control module (increasing weight in the weight module, narrowing the scope in the range module, adjusting the style to serious, etc.). If the correction still does not meet the requirements, factual information and topics can be manually added. Factual errors: Manual correction of factual information is required (e.g., changing "interest rate: 5%" to "interest rate: 3%").

[0064] The technical approach provided by this invention can automate the quality inspection modeling process, greatly reducing the barrier to entry for business personnel. They can easily get started by simply relying on their familiar business logic. The attribute control module enables personalized control of the model node content, which is more in line with the quality inspection characteristics of agent calls. Semi-automatic verification and optimization are achieved after model testing, reducing the manpower and time costs of model building and model deployment. This allows the model to be iterated and updated in a timely manner along with business adjustments, reducing the lag in model updates.

[0065] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0066] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0067] According to embodiments of the present invention, an apparatus for generating a business quality inspection model for implementing the above-described method for generating a business quality inspection model is also provided. Figure 3 This is a schematic diagram of a business quality inspection model generation device according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes: a first acquisition unit 31, an analysis unit 33, a structured processing unit 35, a first generation unit 37, and a second generation unit 39. The generation device for this business quality inspection model will be described below.

[0068] The first acquisition unit 31 is used to acquire business processing information in the target domain.

[0069] Analysis unit 33 is used to analyze business processing information according to the business processing method of the target domain to obtain knowledge documents. The knowledge documents are used to record the quality inspection information required for business quality inspection of the target domain. The quality inspection information includes at least: Q&A information of the business to be inspected, quality inspection reference information, and business process information.

[0070] The structured processing unit 35 is used to perform structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document.

[0071] The first generation unit 37 is used to generate multiple quality inspection model nodes based on structured information. The quality inspection model node is a node in the quality inspection model used to perform quality inspection on the content to be inspected.

[0072] The second generation unit 39 is used to generate a quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes. After receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content.

[0073] It should be noted that the first acquisition unit 31, analysis unit 33, structured processing unit 35, first generation unit 37 and second generation unit 39 mentioned above correspond to steps S202 to S210 in the above embodiments. The three modules and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments.

[0074] As can be seen from the above, in the solution described in the above embodiments of the present invention, the first acquisition unit can be used to acquire business processing information of the target domain; then, the analysis unit can be used to analyze the business processing information according to the business processing method of the target domain to obtain a knowledge document, wherein the knowledge document is used to record the quality inspection information required for business quality inspection of the target domain, and the quality inspection information includes at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; then, the structured processing unit can be used to perform structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document; and the first generation unit can be used to generate multiple quality inspection model nodes based on the structured information, wherein the quality inspection model... The type node is a node in the quality inspection model used to inspect the content to be inspected; and the second generation unit generates the quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes. After receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content. This realizes the structured processing of knowledge documents, generates multiple quality inspection model nodes based on the structured information, and generates the quality inspection model in sequence. This achieves the technical effect of automating the quality inspection modeling process, reduces the threshold for business personnel to use, and allows them to easily get started by relying on their familiar business logic.

[0075] Therefore, the technical solution provided by the embodiments of the present invention solves the technical problem in the related art that the construction of a quality inspection model for supervising whether business personnel's handling of specific scenarios or problems conforms to standard procedures relies on manual operation.

[0076] Optionally, the structured processing unit includes at least one of the following: a first tagging module, used to tag the natural language description corresponding to the knowledge document as a predetermined array, wherein the predetermined array includes at least: business-related products, business types, and revenue of business-related products; and a second tagging module, used to tag the natural language description corresponding to the knowledge document as key-value pairs.

[0077] Optionally, the device for generating the business quality inspection model further includes a classification unit, which, after performing structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document, classifies the structured information according to the knowledge document to obtain the classified structured information.

[0078] Optionally, the first generation unit includes: a determining module for determining the title type of the categorized structured information; and a generation module for generating multiple quality inspection model nodes based on each of the title types.

[0079] Optionally, the device for generating the business quality inspection model further includes an adding unit, used to add control attributes to each of the multiple quality inspection model nodes after generating multiple quality inspection model nodes based on structured information, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model nodes according to the quality inspection requirements of the content to be inspected.

[0080] Optionally, the device for generating the business quality inspection model further includes: a second acquisition unit, used to acquire historical session data of the target domain after generating the quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes; and a quality inspection unit, used to perform quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model.

[0081] Optionally, the quality inspection unit includes: a first acquisition module for acquiring the quality inspection results of the quality inspection model on historical call data; a sending module for sending the quality inspection results to a predetermined terminal device for evaluation of the quality inspection results by the predetermined terminal device; and a second acquisition module for acquiring the evaluation results of the quality inspection results fed back by the predetermined terminal device.

[0082] Optionally, the device for generating the business quality inspection model further includes an optimization unit, used to optimize the quality inspection model based on the quality inspection results when the evaluation results indicate that the erroneous quality inspection information in the quality inspection results exceeds a predetermined threshold.

[0083] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes the method for generating a business quality inspection model as described above.

[0084] Optionally, in this embodiment, the computer-readable storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any communication device in a group of communication devices.

[0085] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: acquiring business processing information of the target domain; analyzing the business processing information according to the business processing method of the target domain to obtain a knowledge document, wherein the knowledge document is used to record the quality inspection information required for business quality inspection of the target domain, and the quality inspection information includes at least: question and answer information of the business to be inspected, quality inspection reference information, and business process information; performing structured processing on the knowledge document to obtain structured information corresponding to the knowledge document; generating multiple quality inspection model nodes based on the structured information, wherein the quality inspection model node is a node in the quality inspection model used for quality inspection of the content to be inspected; generating a quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes, wherein after receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content to be inspected.

[0086] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: marking the natural language description corresponding to the knowledge document as a predetermined array, wherein the predetermined array includes at least: business-related products, business types, and revenue of business-related products; marking the natural language description corresponding to the knowledge document as key-value pairs.

[0087] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after performing structuring processing on the knowledge document to obtain the structured information corresponding to the knowledge document, classifying the structured information according to the knowledge document to obtain the classified structured information.

[0088] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining the header type of the categorized structured information; generating multiple quality inspection model nodes based on each of the header types.

[0089] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after generating multiple quality inspection model nodes based on structured information, adding control attributes to each of the multiple quality inspection model nodes, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model nodes according to the quality inspection requirements of the content to be inspected.

[0090] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after generating a quality inspection model according to the business processing order corresponding to multiple quality inspection model nodes, obtaining historical session data of the target domain; and performing quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model.

[0091] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: obtaining the quality inspection results of the quality inspection model on the historical session data; sending the quality inspection results to a predetermined terminal device for evaluation of the quality inspection results by the predetermined terminal device; and obtaining the evaluation results of the quality inspection results fed back by the predetermined terminal device.

[0092] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: when the evaluation result indicates that the erroneous quality inspection information in the quality inspection result exceeds a predetermined threshold, the quality inspection model is optimized based on the quality inspection result.

[0093] According to another aspect of the present invention, a processor is also provided, which is used to run a program, wherein the program executes the method for generating a business quality inspection model as described above.

[0094] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0095] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0096] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0097] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0099] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0100] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for generating a business quality inspection model, characterized in that, include: Obtain business processing information in the target domain; The business processing information is analyzed according to the business processing method of the target domain to obtain a knowledge document. The knowledge document is used to record the quality inspection information required for business quality inspection of the target domain. The quality inspection information includes at least: Q&A information of the business to be inspected, quality inspection reference information, and business process information. The knowledge document is processed in a structured manner to obtain the structured information corresponding to the knowledge document; Multiple quality inspection model nodes are generated based on the structured information, wherein the quality inspection model node is a node in the quality inspection model used to perform quality inspection on the content to be inspected; The quality inspection model is generated according to the business processing order corresponding to the multiple quality inspection model nodes. Upon receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content. After performing structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document, the method further includes: classifying the structured information according to the knowledge document to obtain the classified structured information. Generating multiple quality inspection model nodes based on the structured information includes: determining the title type of the categorized structured information; automatically generating one quality inspection model node for each category; and generating multiple quality inspection model nodes according to each of the title types.

2. The method for generating a business quality inspection model according to claim 1, characterized in that, The knowledge document is subjected to structured processing to obtain the structured information corresponding to the knowledge document, including at least one of the following: The natural language description corresponding to the knowledge document is marked as a predetermined array, wherein the predetermined array includes at least: business-related products, business type, and revenue of the business-related products; The natural language descriptions corresponding to the knowledge documents are marked as key-value pairs.

3. The method for generating a business quality inspection model according to claim 1, characterized in that, After generating multiple quality inspection model nodes based on the structured information, the process also includes: Add control attributes to each of the plurality of quality inspection model nodes, wherein the control attributes are used to adjust the attributes of predetermined fields in the quality inspection model node according to the quality inspection requirements of the content to be inspected.

4. The method for generating a business quality inspection model according to claim 1, characterized in that, After generating the quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes, the process further includes: Obtain historical session data for the target domain; The quality inspection model is used to perform quality inspection operations on the historical session data in order to test the quality inspection model.

5. The method for generating a business quality inspection model according to claim 4, characterized in that, Perform quality inspection operations on the historical session data using the quality inspection model to test the quality inspection model, including: Obtain the quality inspection results of the historical session data by the quality inspection model; The quality inspection results are sent to a predetermined terminal device for evaluation. Obtain the evaluation results of the quality inspection results fed back by the predetermined terminal device.

6. The method for generating a business quality inspection model according to claim 5, characterized in that, Also includes: When the evaluation result indicates that the erroneous quality inspection information in the quality inspection result exceeds a predetermined threshold, the quality inspection model is optimized based on the quality inspection result.

7. A device for generating a business quality inspection model, characterized in that, include: The first acquisition unit is used to acquire business processing information in the target domain; The analysis unit is used to analyze the business processing information according to the business processing method of the target domain to obtain a knowledge document. The knowledge document is used to record the quality inspection information required for business quality inspection of the target domain. The quality inspection information includes at least: Q&A information of the business to be inspected, quality inspection reference information, and business process information. The structured processing unit is used to perform structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document; The first generation unit is used to generate multiple quality inspection model nodes based on the structured information, wherein the quality inspection model node is a node in the quality inspection model used to perform quality inspection on the content to be inspected. The second generation unit is used to generate the quality inspection model according to the business processing order corresponding to the multiple quality inspection model nodes. Upon receiving the content to be inspected, the quality inspection model sends the content to be inspected to the corresponding quality inspection model node for business quality inspection based on the type of the content. The device for generating the business quality inspection model further includes a classification unit, used to classify the structured information according to the knowledge document after performing structured processing on the knowledge document to obtain the structured information corresponding to the knowledge document, thereby obtaining the classified structured information. The first generation unit includes: a determining module, used to determine the title type of the categorized structured information; and a generation module, used to automatically generate a quality inspection model node for each category, and to generate multiple quality inspection model nodes according to each of the title types.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program executes the method for generating a business quality inspection model according to any one of claims 1 to 6.

Images