Work order processing method and device, medium and program product

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a work order processing method and device, a medium and a program product. The method comprises the steps of obtaining a target feature set corresponding to a user feedback work order, and obtaining a service type corresponding to the user feedback work order according to the target feature set through a pre-trained service classification model; according to the service type, transferring the user feedback work order to the first processing end, and obtaining communication voice between the first processing end and the target user; performing text conversion on the communication voice to obtain a feedback description text, and recognizing the feedback description text by adopting a named entity recognition method to obtain a target service subclass; and according to the target service subclass, continuously transferring the user feedback work order to a second processing end, generating a solution according to the feedback description text through a pre-trained question and answer model, and sending the solution to the second processing end. Automatic and accurate circulation of the user feedback work order can be realized, and the processing efficiency of the user feedback work order can be improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a work order processing method, device, medium, and program product. Background Technology

[0002] With the rapid expansion of the financial industry's service scale, the number of user feedback tickets is growing exponentially. Efficiently processing user feedback tickets is of great significance for improving service quality and customer satisfaction.

[0003] Currently, existing work order processing methods typically involve each processing unit manually determining the next processing unit based on communication with the user, and then manually routing user feedback work orders. However, this manual approach to handling user feedback work orders is not only inefficient but also prone to inaccurate identification and matching of processing units, leading to the blind routing of user feedback work orders and further impacting processing efficiency. Summary of the Invention

[0004] This invention provides a work order processing method, device, medium, and program product, which can realize the automatic and accurate flow of user feedback work orders and improve the processing efficiency of user feedback work orders.

[0005] According to one aspect of the present invention, a work order processing method is provided, comprising: Obtain the target feature set corresponding to the user feedback work order, and obtain the business type corresponding to the user feedback work order based on the target feature set using a pre-trained business classification model. Based on the business type, the user feedback work order is transferred to the first processing terminal, and the communication voice between the first processing terminal and the target user is obtained; The communication voice is converted into text to obtain feedback description text, and the feedback description text is identified using a named entity recognition method to obtain the target business subclass. Based on the target business subclass, the user feedback work order is further transferred to the second processing end, and a solution is generated based on the feedback description text using a pre-trained question-and-answer model, and the solution is sent to the second processing end.

[0006] According to another aspect of the present invention, a work order processing apparatus is provided, comprising: The business type acquisition module is used to acquire the target feature set corresponding to the user feedback work order, and to acquire the business type corresponding to the user feedback work order based on the target feature set using a pre-trained business classification model. The communication voice acquisition module is used to transfer the user feedback work order to the first processing terminal according to the business type, and acquire the communication voice between the first processing terminal and the target user; The business subclass acquisition module is used to convert the communication voice into text to obtain feedback description text, and to use the named entity recognition method to recognize the feedback description text to obtain the target business subclass. The solution generation module is used to continue the flow of the user feedback work order to the second processing terminal according to the target business subclass, and generate a solution based on the feedback description text using a pre-trained question-and-answer model, and send the solution to the second processing terminal.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the work order processing method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, the computer program being used to cause a processor to execute and implement the work order processing method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the work order processing method described in any embodiment of the present invention.

[0010] The technical solution of this invention involves obtaining a target feature set corresponding to a user feedback work order, and using a pre-trained business classification model to obtain the business type corresponding to the user feedback work order based on the target feature set. Based on the business type, the user feedback work order is transferred to a first processing end, and the communication voice between the first processing end and the target user is obtained. The communication voice is converted into text to obtain feedback description text, and a named entity recognition method is used to recognize the feedback description text to obtain the target business subclass. Based on the target business subclass, the user feedback work order is further transferred to a second processing end, and a solution is generated based on the feedback description text using a pre-trained question-answering model, and the solution is sent to the second processing end. By combining feature set-based business type recognition and communication voice-based business subclass recognition, a two-level transfer of user feedback work orders is achieved, enabling automatic and accurate transfer of user feedback work orders and improving processing efficiency.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a work order processing method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of a work order processing method provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of a work order processing device according to Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the work order processing method of this invention. Detailed Implementation

[0014] 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.

[0015] It should be noted that the terms "first," "second," "modification," etc., used 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 terms can be used interchangeably 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.

[0016] In addition, it should be noted that in this embodiment, the information collected is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0017] Example 1 Figure 1 This is a flowchart of a work order processing method provided in Embodiment 1 of the present invention. This embodiment is applicable to processing user feedback work orders in financial business scenarios. The method can be executed by a work order processing device, which can be implemented in hardware and / or software. Typically, the work order processing device can be configured in electronic devices, such as computer equipment, servers, etc. Figure 1 As shown, the method includes: S110. Obtain the target feature set corresponding to the user feedback work order, and obtain the business type corresponding to the user feedback work order based on the target feature set using a pre-trained business classification model.

[0018] In this embodiment, when a user's incoming feedback line is detected, the system can automatically generate a corresponding user feedback work order. The user feedback work order may include information such as the incoming line type, user identifier (e.g., user's mobile phone number), and initial request (e.g., key selection). Next, multi-dimensional feature extraction can be performed on the user feedback work order to obtain a set of features from each dimension as the target feature set. This embodiment does not specifically limit the feature extraction method. Finally, the target feature set can be input into a pre-trained business classification model to obtain the business type output by the business classification model.

[0019] The business classification model can be built based on a convolutional neural network. In this embodiment, the business classification model can be trained in advance. Specifically, firstly, a convolutional neural network model can be built based on preset model parameters as the initial business classification model; then, historical feedback work orders can be collected, and their corresponding feature sets can be extracted, and the business types corresponding to the feature sets can be labeled to generate the correspondence between feature sets and business types as training samples; next, the initial business classification model is iteratively trained based on the training samples until the preset training termination condition is met, and the trained business classification model is obtained.

[0020] S120. Based on the business type, the user feedback work order is transferred to the first processing terminal, and the communication voice between the first processing terminal and the target user is obtained.

[0021] In this embodiment, a pre-defined correspondence between business types and processing agency levels can be established. After determining the current business type, the processing agency level matching the user feedback work order can be determined by searching the pre-defined correspondence. Then, based on the principle of load balancing, the user feedback work order can be assigned to the first processing terminal under the matching processing agency level. The first processing terminal can be the system client of a business category handler. This handler primarily obtains a complete description of the feedback problem through voice communication with the target user, providing a basis for processing by professional personnel.

[0022] Then, the user's call is transferred to the assigned business category handler for processing. The business category handler communicates with the target user via voice, listening to the target user's description of the problem. During the voice communication, the system can capture the real-time audio of the communication between the business category handler and the target user.

[0023] Optionally, depending on the business type, transferring the user feedback ticket to the first processing terminal may include: Obtain the inbound dedicated line type corresponding to the user feedback work order, and obtain the processing agency level corresponding to the user feedback work order based on the service type and the inbound dedicated line type; Based on the processing mechanism hierarchy, the first processing terminal corresponding to the user feedback work order is determined, and the user feedback work order is transferred to the first processing terminal.

[0024] The incoming call leased line type can include ordinary personal business leased lines, enterprise business leased lines, etc. In this embodiment, the correspondence between business type, incoming call leased line type, and processing agency level can be pre-set through work order hierarchical processing rules. After determining the current business type and incoming call leased line type, the processing agency level corresponding to the current user feedback work order can be determined by looking up the correspondence. Then, random allocation can be performed among all processing terminals under this processing agency level to determine the first processing terminal corresponding to the current user feedback work order, and the current user feedback work order is assigned to the first processing terminal.

[0025] In this embodiment, by combining the service type and the inbound call line type for the initial transfer of user feedback work orders, the accuracy of work order transfer can be further improved, the number of work order transfers can be reduced, and the work order transfer path can be shortened.

[0026] S130. The communication voice is converted into text to obtain feedback description text, and the feedback description text is identified using a named entity recognition method to obtain the target business subclass.

[0027] In this embodiment, the communication voice can be converted into text in real time to obtain corresponding feedback description text, which is then displayed to the relevant business category processor in real time, allowing them to fully understand the details of the problem. Simultaneously, for ambiguous information provided by the target user, the business category processor can ask follow-up questions to ensure that the true needs of the target user are ultimately understood and an accurate feedback problem description is formed. After the business category processor completes communication with the target user, the system can obtain the complete feedback description text. The automatically generated feedback description text allows the business category processor to manually modify it and annotate key information.

[0028] In this process, the communication speech is converted into text. The communication speech is input into a pre-trained speech recognition model, and the feedback description text output by the speech recognition model is obtained. In this embodiment, a self-supervised speech pre-trained model can be obtained, and historical feedback speech and corresponding feedback description text can be obtained as training samples. Then, based on the training samples and the contrastive learning loss function, the self-supervised speech pre-trained model is fine-tuned to obtain a trained speech recognition model.

[0029] A speech recognition model can include a convolutional network feature extractor and an encoder. The convolutional network feature extractor encodes the preprocessed speech layer by layer, transforming continuous audio signals into a fixed-dimensional sequence of frame features to capture basic audio features. Then, a model vector quantization module maps each extracted continuous feature frame to a discrete feature vector. This discrete feature vector serves as the core objective for self-supervised training, providing a foundation for contextual semantic learning. Next, a random masking operation is performed on the discrete feature vector to simulate speech loss scenarios and improve model robustness. The masked discrete feature vector is then input into the encoder, which uses a self-attention mechanism to capture the contextual relationships between frames, generating a contextual representation containing semantic information. Finally, the encoded vector is decoded into structured text, simultaneously removing interjections and redundant phrases, and correcting speech ambiguities. By employing speech-to-text technology, user phone conversations can be automatically transcribed without manual word-by-word recording, reducing human error and improving work order processing efficiency.

[0030] In this embodiment, after obtaining the feedback description text, named entity recognition can be performed on the feedback description text to determine the business sub-category to which the user feedback issue belongs. The business sub-category can be a further subdivision of the business type. This embodiment does not specifically limit the method of named entity recognition.

[0031] S140. Based on the target business subclass, the user feedback work order is further transferred to the second processing terminal, and a solution is generated based on the feedback description text using a pre-trained question-and-answer model, and the solution is sent to the second processing terminal.

[0032] Specifically, this embodiment can pre-set the correspondence between business subclasses and processing terminals. After determining the target business subclass, the second processing terminal corresponding to the user feedback work order can be determined by searching the pre-set correspondence. The user feedback work order, the target business subclass, and the feedback description text are then forwarded to the second processing terminal. The second processing terminal can be a system client for professional personnel. These professional personnel provide expert answers to user feedback questions.

[0033] While user feedback tickets undergo a secondary processing flow, a pre-trained question-and-answer model is invoked to process the feedback description text and obtain the solution output by the model. This solution is then provided to the second processing unit to assist in answering user questions. Professional processing personnel can conduct compliance reviews of the solutions generated by the question-and-answer model based on the latest rules in the relevant business domain, and supplement personalized details to tailor a final solution for the user. This closed-loop model of "AI generation + manual verification" ensures both the accuracy and compliance of responses and the immediate processing of feedback.

[0034] In this embodiment, an open-source basic large language model can be obtained, and the correspondence between the question description text and the solution can be generated as training samples. Then, the basic large language model can be fine-tuned and trained using the training samples to obtain a trained question-answering model.

[0035] Optionally, based on the target business subclass, the user feedback ticket may be further transferred to the second processing terminal, which may include: Based on the target business subclass, obtain each candidate processing terminal, and obtain the current load and historical processing satisfaction of each candidate processing terminal; Based on the current load and the historical processing satisfaction, a second processing terminal is determined from among the candidate processing terminals, and the user feedback work order continues to flow to the second processing terminal.

[0036] In this embodiment, when performing a secondary processing of user feedback work orders, firstly, processing terminals matching the target business subclass are identified as candidate processing terminals, and the current load (number of work orders to be processed) and historical processing satisfaction rate for each candidate processing terminal are retrieved. Then, it is determined whether the current load and historical processing satisfaction rate meet preset screening conditions. If so, the current candidate processing terminal is identified as the second processing terminal. For example, the preset screening conditions could be that the current load is less than or equal to 5, and the historical processing satisfaction rate is greater than or equal to 90.

[0037] In this embodiment, by combining load and historical processing satisfaction to filter the second processing end, the accuracy of the work order secondary flow can be improved, and the work order processing efficiency can be improved.

[0038] Optionally, generating a solution based on the feedback description text using a pre-trained question-answering model may include: Obtain the supplementary rule information input from the second processing terminal, and generate target prompt words based on the supplementary rule information, the feedback description text, and the preset prompt word template; The solution is generated based on the target prompt words using a pre-trained question-answering model.

[0039] The supplementary rule information, proposed by professionals on the second processing end based on their expertise, may include the latest compliance requirements, product-specific instructions, and other information. In this embodiment, the supplementary rule information and feedback description text can be filled into the corresponding positions of a preset prompt word template to generate target prompt words. Then, the target prompt words can be input into a pre-trained question-answering model to obtain a solution output by the model. The solution may include problem characterization, processing basis, and specific operational steps.

[0040] In this embodiment, by combining supplementary rule information and feedback description text to generate target prompt words, the completeness and accuracy of the solution can be improved.

[0041] The technical solution of this invention involves obtaining a target feature set corresponding to a user feedback work order, and using a pre-trained business classification model to obtain the business type corresponding to the user feedback work order based on the target feature set. Based on the business type, the user feedback work order is transferred to a first processing end, and the communication voice between the first processing end and the target user is obtained. The communication voice is converted into text to obtain feedback description text, and a named entity recognition method is used to recognize the feedback description text to obtain the target business subclass. Based on the target business subclass, the user feedback work order is further transferred to a second processing end, and a solution is generated based on the feedback description text using a pre-trained question-answering model, and the solution is sent to the second processing end. By combining feature set-based business type recognition and communication voice-based business subclass recognition, a two-level transfer of user feedback work orders is achieved, enabling automatic and accurate transfer of user feedback work orders and improving processing efficiency.

[0042] Example 2 Figure 2 This is a flowchart of a work order processing method provided in Embodiment 2 of the present invention. This embodiment is a further refinement of the above technical solution, and the technical solution in this embodiment can be combined with one or more of the above implementation methods. Figure 2 As shown, the method includes: S210. Obtain the target feature set corresponding to the user feedback work order, and obtain the business type corresponding to the user feedback work order based on the target feature set using a pre-trained business classification model.

[0043] S220. Based on the business type, the user feedback work order is transferred to the first processing terminal, and the communication voice between the first processing terminal and the target user is obtained.

[0044] S230. The communication voice is converted into text to obtain feedback description text, and the feedback description text is segmented and vectorized to obtain word vectors. The state label corresponding to each word vector is obtained through a pre-trained annotation model.

[0045] Specifically, in named entity recognition, the feedback description text is first preprocessed with word vectors, including word segmentation, stop word removal, and vectorization to obtain individual word vectors. Simultaneously, the business type can be converted into a one-hot vector, which is also used as a word vector. Then, the word vectors are input into a pre-trained annotation model, and the state label corresponding to each word vector output by the annotation model is obtained.

[0046] The annotation model can include an input layer, a bidirectional hidden layer, and an output layer. It uses word vectors corresponding to historical feedback description text as training data, and each word vector is labeled with a corresponding state label. The state labels can include three types: business subcategory prefix, business subcategory middle, and non-business subcategory.

[0047] S240. Generate initial label sequences based on the state labels corresponding to each word vector, and use the Viterbi algorithm to determine the target label sequence from the initial label sequences.

[0048] Then, all state labels are randomly combined to generate initial label sequences. The Viterbi algorithm is used to traverse all initial label sequences, calculate and extract the initial label sequence with the highest probability, and use it as the target label sequence.

[0049] S250. Obtain the target business subclass based on the target tag sequence.

[0050] Finally, based on the correspondence between word vectors and status labels, the vector sequence corresponding to the target label sequence can be obtained. Then, based on the pre-defined correspondence between vector sequences and business subclasses, the target business subclass matching the current vector sequence can be found. Alternatively, the correspondence between label sequences and business subclasses can be pre-defined. After determining the target label sequence, the corresponding target business subclass can be determined by searching the pre-defined correspondence.

[0051] S260. Based on the target business subclass, the user feedback work order is further transferred to the second processing terminal, and a solution is generated based on the feedback description text using a pre-trained question-and-answer model, and the solution is sent to the second processing terminal.

[0052] Optionally, the technical solution of this embodiment of the invention may further include: Obtain the business data and behavior data corresponding to the current user, and preprocess the business data and behavior data to obtain structured data; The structured data is classified and labeled to obtain the static and dynamic tags corresponding to the current user, and the current feature set corresponding to the current user is generated based on the static and dynamic tags. Obtain the current user identifier corresponding to the current user, and generate a correspondence between the current user identifier and the current feature set, storing it in the target database.

[0053] In this embodiment, the feature set corresponding to the user feedback work order may further include the features of the user associated with the work order. Specifically, historical user feedback data can be collected, including historical feedback work orders and historical feedback description text, and business data and behavioral data can be extracted from this historical feedback data. Next, the business data and behavioral data can be preprocessed, including cleaning, deduplication, and standardization, to obtain preprocessed business data and preprocessed behavioral data as structured data. Further, a tagging engine is invoked to classify and label the preprocessed business data to obtain static tags, and simultaneously, the preprocessed behavioral data is classified and labeled to obtain dynamic tags. This embodiment does not specifically limit the tagging engine.

[0054] Finally, a set of static and dynamic tags is generated as the current feature set, which is then stored in the target database, associated with the current user identifier, and supported for real-time querying. Static tags may include basic information, product holding information, risk level information, etc. Dynamic tags may include business processing records, historical feedback records, etc.

[0055] It should be noted that users have been fully informed and have been fully authorized regarding the collection, processing, and application of historical feedback data, and that the relevant data will not be used for other purposes.

[0056] In this embodiment, by pre-labeling users with static and dynamic tags based on user data to generate corresponding feature sets, the accuracy of business type identification can be improved, and the accuracy of the first-time flow of work orders can be further improved.

[0057] Optionally, obtaining the target feature set corresponding to the user feedback ticket may include: Obtain the target user identifier corresponding to the user feedback work order, and based on the target user identifier and the correspondence between user identifiers and feature sets in the target database, find the target feature set corresponding to the user feedback work order in the target database.

[0058] Correspondingly, when obtaining the target feature set, we can first obtain the target user identifier corresponding to the user feedback ticket, and then retrieve the target feature set corresponding to the user feedback ticket from the target database based on the target user identifier. The target user identifier can be a string that uniquely identifies the user.

[0059] In this embodiment, by directly searching the target feature set corresponding to the user feedback work order from the target database based on the user identifier, the efficiency of feature set acquisition can be improved, and the efficiency of work order processing can be improved.

[0060] The technical solution of this invention involves segmenting and vectorizing the feedback description text to obtain word vectors, and using a pre-trained annotation model to obtain state labels corresponding to each word vector. Based on the state labels corresponding to each word vector, initial label sequences are generated, and a Viterbi algorithm is used to determine the target label sequence from these initial label sequences. The target business subclass is then obtained based on the target label sequence. By combining the annotation model and the Viterbi algorithm to identify the target business subclass from the feedback description text, the accuracy of business subclass identification can be improved, thus enhancing the precision of work order secondary processing.

[0061] This invention proposes a work order processing method that integrates feature set construction, intelligent business type identification, and professional solution provision, building upon existing work order workflow processing. During the work order workflow, preliminary business category identification is achieved by constructing a feature set. Based on automatically generated user feedback problem descriptions, sub-categories of businesses are accurately identified, enabling intelligent workflow and assignment of specialized personnel. This increases the accuracy of work order workflow and reduces manual input of feedback problems by agents. When responding to customer feedback, professional personnel, combining professional business rules and a comprehensive model, can provide complete, accurate, and compliant solutions, improving feedback processing efficiency, reducing the volume of user feedback work orders, and enhancing user satisfaction.

[0062] Example 3 Figure 3 This is a schematic diagram of a work order processing device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a business type acquisition module 310, a communication voice acquisition module 320, a business subclass acquisition module 330, and a solution generation module 340; wherein, The business type acquisition module 310 is used to acquire the target feature set corresponding to the user feedback work order, and to acquire the business type corresponding to the user feedback work order based on the target feature set through a pre-trained business classification model. The communication voice acquisition module 320 is used to transfer the user feedback work order to the first processing terminal according to the business type, and acquire the communication voice between the first processing terminal and the target user; The business subclass acquisition module 330 is used to convert the communication voice into text to obtain feedback description text, and to use a named entity recognition method to recognize the feedback description text to obtain the target business subclass. The solution generation module 340 is used to continue the flow of the user feedback work order to the second processing terminal according to the target business subclass, and generate a solution based on the feedback description text through a pre-trained question-and-answer model, and send the solution to the second processing terminal.

[0063] The technical solution of this invention involves obtaining a target feature set corresponding to a user feedback work order, and using a pre-trained business classification model to obtain the business type corresponding to the user feedback work order based on the target feature set. Based on the business type, the user feedback work order is transferred to a first processing end, and the communication voice between the first processing end and the target user is obtained. The communication voice is converted into text to obtain feedback description text, and a named entity recognition method is used to recognize the feedback description text to obtain the target business subclass. Based on the target business subclass, the user feedback work order is further transferred to a second processing end, and a solution is generated based on the feedback description text using a pre-trained question-answering model, and the solution is sent to the second processing end. By combining feature set-based business type recognition and communication voice-based business subclass recognition, a two-level transfer of user feedback work orders is achieved, enabling automatic and accurate transfer of user feedback work orders and improving processing efficiency.

[0064] Optionally, the work order processing device also includes: The structured data acquisition module is used to acquire the business data and behavioral data corresponding to the current user, and to preprocess the business data and behavioral data to obtain structured data. The feature set generation module is used to classify and label the structured data to obtain the static and dynamic labels corresponding to the current user, and generate the current feature set corresponding to the current user based on the static and dynamic labels. The correspondence generation module is used to obtain the current user identifier corresponding to the current user, and generate the correspondence between the current user identifier and the current feature set and store it in the target database.

[0065] Optionally, the business type acquisition module 310 is specifically used to acquire the target user identifier corresponding to the user feedback work order, and to find the target feature set corresponding to the user feedback work order from the target database based on the target user identifier and the correspondence between user identifiers and feature sets in the target database.

[0066] Optionally, the communication voice acquisition module 320 is specifically used to acquire the incoming call line type corresponding to the user feedback work order, and to acquire the processing agency level corresponding to the user feedback work order based on the service type and the incoming call line type. Based on the processing mechanism hierarchy, the first processing terminal corresponding to the user feedback work order is determined, and the user feedback work order is transferred to the first processing terminal.

[0067] Optionally, the business subclass acquisition module 330 is specifically used to perform word segmentation and vectorization processing on the feedback description text to obtain each word vector, and to obtain the state label corresponding to each word vector through a pre-trained annotation model; Based on the state labels corresponding to each word vector, generate each initial label sequence, and use the Viterbi algorithm to determine the target label sequence from each initial label sequence; Based on the target tag sequence, obtain the target business subclass.

[0068] Optionally, the solution generation module 340 is specifically used to obtain each candidate processing terminal according to the target business subclass, and obtain the current load and historical processing satisfaction corresponding to each candidate processing terminal; Based on the current load and the historical processing satisfaction, a second processing terminal is determined from among the candidate processing terminals, and the user feedback work order continues to flow to the second processing terminal.

[0069] Optionally, the solution generation module 340 is further configured to obtain supplementary rule information input from the second processing terminal, and generate target prompt words based on the supplementary rule information, the feedback description text, and the preset prompt word template; The solution is generated based on the target prompt words using a pre-trained question-answering model.

[0070] The work order processing device provided in the embodiments of the present invention can execute the work order processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0071] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0072] Example 4 Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device 40 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 40 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0073] like Figure 4 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 42 or loaded from the storage unit 48 into the random access memory 43. The RAM 43 can also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0074] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0075] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as work order processing methods.

[0076] In some embodiments, the work order processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the work order processing method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the work order processing method by any other suitable means (e.g., by means of firmware).

[0077] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), system-on-a-chip (SoCs), complex programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0078] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0079] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0080] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 40, which includes: a display device (e.g., a cathode ray tube or liquid crystal display) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device 40. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0081] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0082] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact via a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server.

[0083] This embodiment may also include a computer program product, which includes a computer program that, when executed by a processor, implements the work order processing method provided in any embodiment of the present invention.

[0084] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0085] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A work order processing method, characterized in that, include: Obtain the target feature set corresponding to the user feedback work order, and obtain the business type corresponding to the user feedback work order based on the target feature set using a pre-trained business classification model; Based on the business type, the user feedback work order is transferred to the first processing terminal, and the communication voice between the first processing terminal and the target user is obtained; The communication voice is converted into text to obtain feedback description text, and the feedback description text is identified using a named entity recognition method to obtain the target business subclass. Based on the target business subclass, the user feedback work order is further transferred to the second processing end, and a solution is generated based on the feedback description text using a pre-trained question-and-answer model, and the solution is sent to the second processing end.

2. The method according to claim 1, characterized in that, Also includes: Obtain the business data and behavior data corresponding to the current user, and preprocess the business data and behavior data to obtain structured data; The structured data is classified and labeled to obtain the static and dynamic tags corresponding to the current user, and the current feature set corresponding to the current user is generated based on the static and dynamic tags. Obtain the current user identifier corresponding to the current user, and generate a correspondence between the current user identifier and the current feature set, storing it in the target database.

3. The method according to claim 2, characterized in that, Obtain the target feature set corresponding to the user feedback ticket, including: Obtain the target user identifier corresponding to the user feedback work order, and based on the target user identifier and the correspondence between user identifiers and feature sets in the target database, find the target feature set corresponding to the user feedback work order in the target database.

4. The method according to claim 1, characterized in that, Based on the business type, the user feedback work order is transferred to the first processing terminal, including: Obtain the inbound dedicated line type corresponding to the user feedback work order, and obtain the processing agency level corresponding to the user feedback work order based on the service type and the inbound dedicated line type; Based on the processing mechanism hierarchy, the first processing terminal corresponding to the user feedback work order is determined, and the user feedback work order is transferred to the first processing terminal.

5. The method according to claim 1, characterized in that, The feedback description text is identified using a named entity recognition method to obtain the target business subclass, including: The feedback description text is segmented and vectorized to obtain word vectors, and the state label corresponding to each word vector is obtained through a pre-trained annotation model. Based on the state labels corresponding to each word vector, generate each initial label sequence, and use the Viterbi algorithm to determine the target label sequence from each initial label sequence; Based on the target tag sequence, obtain the target business subclass.

6. The method according to claim 1, characterized in that, Based on the target business subclass, the user feedback work order is further transferred to the second processing end, including: Based on the target business subclass, obtain each candidate processing terminal, and obtain the current load and historical processing satisfaction of each candidate processing terminal; Based on the current load and the historical processing satisfaction, a second processing terminal is determined from among the candidate processing terminals, and the user feedback work order continues to flow to the second processing terminal.

7. The method according to claim 1, characterized in that, Using a pre-trained question-answering model, a solution is generated based on the feedback description text, including: Obtain the supplementary rule information input from the second processing terminal, and generate target prompt words based on the supplementary rule information, the feedback description text, and the preset prompt word template; The solution is generated based on the target prompt words using a pre-trained question-answering model.

8. An electronic device, characterized in that, The electronic device includes: At least one processor, and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the work order processing method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the work order processing method according to any one of claims 1-7.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the work order processing method according to any one of claims 1-7.

Images