Method, device, electronic device and storage medium for generating work order description information

The work order description information is generated through bidirectional neural networks and codec networks, which solves the problem of inefficient manual refining, and realizes efficient and accurate information refining, simplifying the work order description generation process.

CN111858911BInactive Publication Date: 2024-11-05BEIJING DIDI INFINITY TECH & DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010319301.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-04-21
Publication Date
2024-11-05
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the prior art, the generation of work order description information relies on manual refinement, which leads to inefficiency and error proneness, and the information refinement is high in complexity, which increases the work burden of manual.

Method used

Bidirectional neural network and codec network are used to generate work ticket description information. By characterizing the context information in the statement vector in the target work ticket document, selecting key statements and using codec network to simulate the thinking process, concise and clear description information is generated.

Benefits of technology

It improves the efficiency and accuracy of work order description information generation, reduces manual intervention, and enhances the simplicity and fluency of information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111858911B_ABST
    Figure CN111858911B_ABST
Patent Text Reader

Abstract

The present application relates to the field of summary generation, and in particular to a method, device, electronic device, and storage medium for generating work order description information. The present application can generate a sentence vector corresponding to each sentence by inputting multiple sentences contained in a target work order document into a bidirectional neural network. The sentence vector is used to characterize the context information of the sentence in the target work order document, and according to the sentence vector corresponding to each sentence, key sentences can be selected from the multiple sentences contained in the target work order document, and then, according to the key sentences and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, the efficiency of generating the work order description information can be improved by machine-generated work order description information, the key sentences can be extracted from the target work order document through the bidirectional neural network, and the human thinking process can be simulated through the encoding and decoding network, and then, the accuracy of the generated work order description information can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of summary generation, and in particular to a method, device, electronic device and storage medium for generating work order description information. Background Art

[0002] Since the work order document contains a large amount of information, it is usually necessary to refine the information in the work order document to generate the work order description information. The work order description information is required to be concise and clear, and summarize all the information in the work order document.

[0003] At present, work order description information is mostly generated by manually extracting information from work order documents. However, information extraction itself is a relatively difficult task, and a large amount of repetitive extraction work will greatly increase the complexity of manual work. Therefore, manual extraction will not only take a lot of time, but also be inefficient and prone to errors. Summary of the invention

[0004] In view of this, the embodiments of the present application at least provide a method, device, electronic device and storage medium for generating work order description information, which can improve the efficiency and accuracy of generating work order description information.

[0005] This application mainly includes the following aspects:

[0006] In a first aspect, an embodiment of the present application provides a method for generating work order description information, the generating method comprising:

[0007] Obtain a target work order document, and determine a plurality of statements contained in the target work order document;

[0008] Input the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document;

[0009] Selecting key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences;

[0010] The work order description information of the target work order document is generated according to the key sentence and the trained codec network.

[0011] In a possible implementation, the bidirectional neural network includes a forward neural network and a reverse neural network; the step of inputting the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence includes:

[0012] Input the multiple sentences into the trained forward neural network to obtain a forward vector corresponding to each sentence, and input the multiple sentences into the trained reverse neural network to obtain a reverse vector corresponding to each sentence; the forward vector is used to represent the previous information of the corresponding sentence in the target work order document; the reverse vector is used to represent the following information of the corresponding sentence in the target work order document;

[0013] The forward vector and the reverse vector corresponding to each sentence are merged to generate the sentence vector corresponding to each sentence.

[0014] In a possible implementation, selecting a key sentence from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences includes:

[0015] For each of the multiple sentences, input a sentence vector corresponding to each sentence into a trained binary classification network to determine the importance probability of each sentence in the target work order document;

[0016] According to the importance probability of each sentence in the target work order document, a key sentence is selected from the multiple sentences included in the target work order document.

[0017] In a possible implementation manner, generating the work order description information of the target work order document according to the key sentence and the trained codec network includes:

[0018] Input the words contained in the key sentence into the trained encoding and decoding network to generate a word sequence;

[0019] The work order description information of the target work order document is generated according to the word sequence.

[0020] In a possible implementation, the encoding and decoding network includes an encoding network and a decoding network; the step of inputting the words included in the key sentence into the trained encoding and decoding network to generate a word sequence includes:

[0021] Input the words contained in the key sentence into the trained encoding network to generate an encoding vector;

[0022] The encoding vector is input into the trained decoding network to generate the word sequence.

[0023] In a possible implementation manner, before obtaining the target work order document and determining a plurality of statements contained in the target work order document, the generating method further includes:

[0024] Training the initial work order description information generation model to obtain the trained work order description information generation model;

[0025] Among them, the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

[0026] In a possible implementation, the training of the initial work order description information generation model to obtain the trained work order description information generation model includes:

[0027] Obtain work order sample documents in various fields and the first sample description corresponding to each work order sample document;

[0028] According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial work order description information generation model is trained to obtain the trained work order description information generation model.

[0029] In a possible implementation, the initial work order description information generation model is trained according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model, including:

[0030] Inputting each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents;

[0031] According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0032] In a possible implementation, adjusting the parameters in the initial work order description information generation model according to the first sample description and the second sample description corresponding to each of the work order sample documents to obtain the trained work order description information generation model includes:

[0033] According to a preset quality assessment standard, the first sample description and the second sample description corresponding to each of the work order sample documents are compared to obtain a comparison result corresponding to each of the work order sample documents;

[0034] According to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0035] In a possible implementation, the initial work order description information generation model is trained according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model, including:

[0036] According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial bidirectional neural network is trained to obtain the trained bidirectional neural network;

[0037] According to the work order sample documents of each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain the trained codec network corresponding to each field.

[0038] In a second aspect, an embodiment of the present application further provides a device for generating work order description information, the device comprising:

[0039] An acquisition module, used to acquire a target work order document and determine a plurality of statements contained in the target work order document;

[0040] A first generating module is used to input the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document;

[0041] A selection module, configured to select key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences;

[0042] The second generating module is used to generate the work order description information of the target work order document according to the key sentence and the trained encoding and decoding network.

[0043] In a possible implementation, the bidirectional neural network includes a forward neural network and a reverse neural network; the first generation module includes:

[0044] A first generating unit is used to input the multiple sentences into the trained forward neural network to obtain a forward vector corresponding to each sentence, and input the multiple sentences into the trained reverse neural network to obtain a reverse vector corresponding to each sentence; the forward vector is used to represent the previous information of the corresponding sentence in the target work order document; the reverse vector is used to represent the following information of the corresponding sentence in the target work order document;

[0045] The second generating unit is used to merge the forward vector and the reverse vector corresponding to each sentence to generate the sentence vector corresponding to each sentence.

[0046] In a possible implementation, the selection module includes:

[0047] A determination unit, configured to input, for each of the plurality of sentences, a sentence vector corresponding to each sentence into a trained binary classification network, and determine an importance probability of each sentence in the target work order document;

[0048] The selection unit is used to select key sentences from the multiple sentences included in the target work order document according to the importance probability of each sentence in the target work order document.

[0049] In a possible implementation, the second generating module includes:

[0050] A third generating unit, used for inputting the words contained in the key sentence into the trained encoding and decoding network to generate a word sequence;

[0051] The fourth generating unit is used to generate the work order description information of the target work order document according to the word sequence.

[0052] In a possible implementation, the encoding and decoding network includes an encoding network and a decoding network; the third generating unit is specifically configured to generate the word sequence according to the following steps:

[0053] Input the words contained in the key sentence into the trained encoding network to generate an encoding vector;

[0054] The encoding vector is input into the trained decoding network to generate the word sequence.

[0055] In a possible implementation manner, the generating device further includes:

[0056] A third generation module is used to train the initial work order description information generation model to obtain the trained work order description information generation model;

[0057] Among them, the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

[0058] In a possible implementation, the third generation module includes:

[0059] An acquisition unit, used to acquire work order sample documents in various fields and a first sample description corresponding to each work order sample document;

[0060] The fifth generating unit is used to train the initial work order description information generation model according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model.

[0061] In a possible implementation manner, the fifth generating unit is used to obtain the trained work order description information generation model according to the following steps:

[0062] Inputting each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents;

[0063] According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0064] In a possible implementation manner, the fifth generating unit is specifically configured to obtain the trained work order description information generation model according to the following steps:

[0065] According to a preset quality assessment standard, the first sample description and the second sample description corresponding to each of the work order sample documents are compared to obtain a comparison result corresponding to each of the work order sample documents;

[0066] According to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0067] In a possible implementation manner, the fifth generating unit is further configured to obtain the trained work order description information generation model according to the following steps:

[0068] According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial bidirectional neural network is trained to obtain the trained bidirectional neural network;

[0069] According to the work order sample documents of each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain the trained codec network corresponding to each field.

[0070] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the method for generating work order description information described in the first aspect or any possible implementation manner of the first aspect.

[0071] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for generating work order description information described in the above-mentioned first aspect or any possible implementation manner of the first aspect are executed.

[0072] In an embodiment of the present application, by inputting multiple sentences contained in a target work order document into a bidirectional neural network, a sentence vector corresponding to each sentence can be generated, and the sentence vector is used to characterize the context information of the sentence in the target work order document, and according to the sentence vector corresponding to each sentence, a key sentence can be selected from the multiple sentences contained in the target work order document, and then, according to the key sentence and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, by machine-generated work order description information, the efficiency of generating work order description information can be improved, and through the bidirectional neural network, key sentences can be extracted from the target work order document, and through the encoding and decoding network, the human thinking process can be simulated, and then, the accuracy of the generated work order description information can be improved.

[0073] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 A flowchart of a method for generating work order description information provided in an embodiment of the present application is shown;

[0076] Figure 2 A flowchart showing another method for generating work order description information provided in an embodiment of the present application is shown;

[0077] Figure 3One of the functional module diagrams of a device for generating work order description information provided in an embodiment of the present application is shown;

[0078] Figure 4 Shows Figure 3 The functional module diagram of the first generation module in FIG.

[0079] Figure 5 Shows Figure 3 Functional module diagram of the selected module;

[0080] Figure 6 Shows Figure 3 Functional module diagram of the second generation module in FIG.

[0081] Figure 7 A second functional module diagram of a device for generating work order description information provided in an embodiment of the present application is shown;

[0082] Figure 8 Shows Figure 7 Functional module diagram of the third generation module;

[0083] Fig. 9 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown.

[0084] Description of main component symbols:

[0085] In the figure: 300-a device for generating work order description information; 310-an acquisition module; 320-a first generation module; 321-a first generation unit; 322-a second generation unit; 330-a selection module; 331-a determination unit; 332-a selection unit; 340-a second generation module; 341-a third generation unit; 342-a fourth generation unit; 350-a third generation module; 351-an acquisition unit; 352-a fifth generation unit; 900-an electronic device; 910-a processor; 920-a memory; 930-a bus. DETAILED DESCRIPTION

[0086] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0087] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0088] In order to enable those skilled in the art to use the contents of this application, the following implementation method is given in combination with the specific application scenario of "generating work order description information". For those skilled in the art, the general principles defined here can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application.

[0089] The following methods, devices, electronic devices or computer-readable storage media of the embodiments of the present application can be applied to any scenario where work order description information needs to be generated. The embodiments of the present application are not limited to specific application scenarios. Any scheme using the method and device for generating work order description information provided by the embodiments of the present application is within the scope of protection of the present application.

[0090] It is worth noting that before the present application was submitted, work order description information was mostly generated by manually extracting information from work order documents. However, information extraction itself is a relatively difficult task, and a large amount of repetitive extraction work will greatly increase the complexity of manual work. Therefore, manual extraction will not only take a lot of time, but is also inefficient and prone to errors.

[0091] In response to the above problems, in an embodiment of the present application, a sentence vector corresponding to each sentence can be generated by inputting multiple sentences contained in a target work order document into a bidirectional neural network. The sentence vector is used to characterize the context information of the sentence in the target work order document, and based on the sentence vectors corresponding to each sentence, key sentences can be selected from the multiple sentences contained in the target work order document, and then, based on the key sentences and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, by machine-generated work order description information, the efficiency of generating work order description information can be improved, and through the bidirectional neural network, key sentences can be extracted from the target work order document, and through the encoding and decoding network, the human thinking process can be simulated, and then the accuracy of the generated work order description information can be improved.

[0092] To facilitate the understanding of the present application, the technical solution provided by the present application is described in detail below in conjunction with specific embodiments.

[0093] Figure 1 This is a flow chart of a method for generating work order description information provided in an embodiment of the present application. Figure 1 As shown, the method for generating work order description information provided in the embodiment of the present application includes the following steps:

[0094] S101: Acquire a target work order document, and determine a plurality of statements contained in the target work order document.

[0095] In a specific implementation, a target work order document for work order description information to be generated is obtained, and a plurality of statements contained in the target work order document is determined.

[0096] Here, extracting work order description information from the work order document is suitable for multiple application scenarios, including but not limited to:

[0097] Application scenario 1: When a user interacts with customer service, the manual customer service needs to refine the information fed back by the user and generate work order description information.

[0098] Application scenario 2: In the driver and passenger conflict detection scenario, it is necessary not only to determine whether there is a conflict in the audio, but also to describe the content of the conflict to facilitate subsequent statistics and analysis. This task can be regarded as equivalent to generating work order description information.

[0099] Application scenario 3: In user comments, it is necessary not only to know the user's emotional tendencies, but also the user's general demands, so as to facilitate adjustments and improvements to the product. This task can be regarded as equivalent to generating work order description information.

[0100] S102: Input the multiple sentences into a trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent context information of the corresponding sentence in the target work order document.

[0101] In a specific implementation, multiple sentences contained in the target work order document are input into a trained bidirectional neural network to obtain sentence vectors corresponding to each sentence, wherein the sentence vector of each sentence can represent the context information of the sentence in the target work order document.

[0102] Here, a bidirectional neural network (Bi-directional Long Short-Term Memory, BiLSTM) is a type of recurrent neural network (RNN), which is used to extract context information in documents in natural language processing tasks. Therefore, by inputting multiple sentences contained in the target work order document into the trained bidirectional neural network together, a sentence vector corresponding to each sentence that can represent the context information of the sentence in the target work order document can be obtained.

[0103] Furthermore, the bidirectional neural network includes a forward neural network and a reverse neural network; here, the process of generating a sentence vector for each sentence through the bidirectional neural network is described, and the step S102 of inputting the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence includes the following steps:

[0104] Step a1: input the multiple sentences into the trained forward neural network to obtain a forward vector corresponding to each sentence, and input the multiple sentences into the trained reverse neural network to obtain a reverse vector corresponding to each sentence; the forward vector is used to represent the context information of the corresponding sentence in the target work order document; the reverse vector is used to represent the context information of the corresponding sentence in the target work order document.

[0105] In a specific implementation, multiple sentences in the target work order document are input into a trained forward neural network together, and the forward vector corresponding to each sentence can be obtained. Also, multiple sentences in the target work order document are input into a trained reverse neural network together, and the reverse vector corresponding to each sentence can be obtained. Here, the forward vector of each sentence can represent the context information of the sentence in the target work order document, that is, the relationship information between the sentence and the sentence before the sentence in the target work order document; the reverse vector of each sentence can represent the context information of the sentence in the target work order document, that is, the relationship information between the sentence and the sentence after the sentence in the target work order document. Among them, the forward vector and reverse vector corresponding to each sentence can be vectors of the same dimension, and the number of elements contained in each vector is usually the same.

[0106] Step a2: Merge the forward vector and the reverse vector corresponding to each sentence to generate the sentence vector corresponding to each sentence.

[0107] In a specific implementation, for each of the multiple statements contained in the target work order document, the forward vector and the reverse vector of each statement are merged, that is, the forward vector and the reverse vector of each statement are spliced ​​together to obtain the statement vector corresponding to each statement. For example, the forward vector corresponding to statement A is c(c1, c2, c3), and the reverse vector corresponding to statement A is d(d1, d2, d3), then the statement vector corresponding to statement A is (c1, c2, c3, d1, d2, d3).

[0108] S103: Select key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the respective sentences.

[0109] In a specific implementation, after obtaining the sentence vectors corresponding to the multiple sentences contained in the target work order document, since the sentence vector of each sentence can represent the context information of the sentence in the target work order document, the key sentences that play an important role in the target work order document can be selected from the multiple sentences contained in the target work order document according to the sentence vectors corresponding to each sentence.

[0110] Further, the process of selecting key sentences is described, that is, selecting key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to each sentence in step S103 includes the following steps:

[0111] Step b1: for each of the multiple sentences, input the sentence vector corresponding to each sentence into the trained binary classification network to determine the importance probability of each sentence in the target work order document.

[0112] In a specific implementation, for each of the multiple sentences contained in the target work order document, the sentence vector corresponding to each sentence is input into a trained binary classification network, and the importance probability of each sentence in the target work order document is calculated through the binary classification network.

[0113] Here, the binary classification network can be a multi-classification network based on multi-layer perception (MLP-Softmax, MultilayerPerceptron-Softmax). The working principle is that, assuming there are K different categories, for a given input x, the probability value of x belonging to each category is estimated, and then the input x is classified according to the probability value.

[0114] Step b2: Select key sentences from the multiple sentences included in the target work order document according to the importance probability of each sentence in the target work order document.

[0115] In a specific implementation, from multiple sentences contained in the target work order document, sentences with an importance probability higher than a preset threshold are selected as key sentences. Specifically, after each sentence vector is input into the binary classification network, it will output 0 or 1, 0 means that the sentence is not selected, that is, the sentence is not a key sentence; 1 means that the sentence is selected, that is, the sentence is a key sentence.

[0116] It should be noted that by using a bidirectional neural network, key sentences can be extracted from the target work order document. Therefore, the generated work order description information can be more closely linked to the target work order document, and the meaning of the target work order document can be better preserved.

[0117] S104: Generate work order description information of the target work order document according to the key sentence and the trained codec network.

[0118] In a specific implementation, after extracting key sentences from the target work order document, the work order description information of the target work order document can be obtained according to the extracted key sentences and the trained codec network, wherein there are multiple key sentences. Through the codec network, the human thinking process can be simulated, thereby improving the content fluency of the work order description information and the accuracy of the generated work order description information.

[0119] Here, the encoder-decoder network (ED) is a very common model framework in deep learning. The most notable feature of the encoder-decoder framework is that it is an end-to-end learning algorithm. Encoding is to convert the input sequence into a vector of fixed length, and decoding is to convert the previously generated fixed vector into an output sequence. If we think from a human perspective, we first see the source sequence, read it once, and then remember the source sequence in our brain, and store it in a certain position in the brain to form our own memory. Then we think about it and turn the things in the brain into output and write it down. Then the process of reading in our brain is called encoding, which is to turn the input into our own memory and put it in the brain. This memory can be called context, and then we convert it into the answer based on this context and write it down. This writing process is called decoding, which is the process of encoding-storage-decoding.

[0120] Further, the work order description information of the target work order document is generated according to the key sentence and the trained codec network. That is, the work order description information of the target work order document is generated according to the key sentence and the trained codec network in step S104, including the following steps:

[0121] Step c1: input the words contained in the key sentence into the trained encoding and decoding network to generate a word sequence.

[0122] In a specific implementation, the words contained in the key sentences extracted from the target work order document are input into the trained codec network together to generate a word sequence corresponding to the target work order document. That is, in the work of the codec network, one network in the codec network is used to simulate the brain's reading action, a feature vector of a specific length is used to simulate the brain's memory, and another network in the codec network is used to simulate the brain's action of thinking and getting an answer, thereby obtaining a word sequence.

[0123] Furthermore, the codec network includes an encoding network and a decoding network; the generation of a word sequence is described, that is, the words contained in the key sentence are input into the trained codec network in step c1 to generate a word sequence, including the following steps:

[0124] The words contained in the key sentence are input into the trained encoding network to generate an encoding vector; the encoding vector is input into the trained decoding network to generate the word sequence.

[0125] In the specific implementation, in the work of the encoding and decoding network, the encoding network is used to simulate the brain's reading action, a feature vector of a specific length is used to simulate the brain's memory, and the decoding network is used to simulate the brain's action of thinking to get the answer, that is, to obtain the word sequence corresponding to the target work order document. In this way, through the encoding and decoding network, the human thinking process can be simulated, and then the accuracy of the generated work order description information can be improved.

[0126] Step c2: Generate the work order description information of the target work order document according to the word sequence.

[0127] In a specific implementation, the generated word sequence is used as the work order description information of the target work order document.

[0128] Here, the Token Generation / Pointer mechanism can be used to process words so that they can be correctly generated into word sequences.

[0129] It should be noted that, by combining a bidirectional neural network and an encoding-decoding network, the present application can extract relatively important sentences, namely key sentences, and identify the semantics of the sentences. In this way, the generated work order description can retain the meaning of the original text while ensuring the fluency of the content.

[0130] In an embodiment of the present application, by inputting multiple sentences contained in the target work order document into a bidirectional neural network, a sentence vector corresponding to each sentence can be generated, and the sentence vector is used to characterize the context information of the sentence in the target work order document, and according to the sentence vector corresponding to each sentence, a key sentence can be selected from the multiple sentences contained in the target work order document, and then, according to the key sentence and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, by machine-generated work order description information, the efficiency of generating work order description information can be improved, and through the bidirectional neural network, key sentences can be extracted from the target work order document, and through the encoding and decoding network, the human thinking process can be simulated, and then, the accuracy of the generated work order description information can be improved.

[0131] Figure 2 This is a flowchart of another method for generating work order description information provided by an embodiment of the present application. Figure 2 As shown, the method for generating work order description information provided in the embodiment of the present application includes the following steps:

[0132] S201: Train the initial work order description information generation model to obtain the trained work order description information generation model; wherein the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

[0133] In the specific implementation, the work order description information generation model is a multi-task learning model, which includes a bidirectional neural network and multiple codec networks, where different codec networks are used to process work order documents in different fields. In the process of training the initial work order description information generation model, the bidirectional neural network and multiple codec networks can be trained simultaneously.

[0134] Here, the multi-task learning model is a joint learning model, where multiple tasks are learned in parallel. The results of multiple tasks affect each other. In machine learning, for example, to optimize a specific indicator, multiple model sets are trained to complete the specified task in order to achieve this goal. Then, the model is improved by fine-tuning the parameters until the performance no longer improves. Although this can achieve an acceptable performance for a task, it may ignore some information that helps to do better in the indicator. Specifically, this information is the supervision data of related tasks. By sharing representation information between related tasks, the model generalizes better on the original task. This method is called multi-task learning.

[0135] Here, work order documents in different fields include but are not limited to customer service work order documents, comment work order documents, and driver-passenger conversation documents. The generation of work order description information for work order documents in various fields of this application shares a bidirectional neural network. Since the data between various fields have a certain relationship, the same bidirectional neural network is used to extract key sentences. Since each field has its own characteristics, different encoding and decoding networks are used to generate work order description information corresponding to each field.

[0136] Further, the training process of the work order description information generation model is described; that is, the initial work order description information generation model is trained in step S201 to obtain the trained work order description information generation model, including the following steps:

[0137] Step d1: Obtain sample work order documents in various fields and the first sample description corresponding to each sample work order document.

[0138] In a specific implementation, work order sample documents belonging to various fields and a first sample description corresponding to each work order sample document are obtained. The first sample description is a standard work order description of the corresponding work order sample document. Usually, the first sample description is a manual description of the corresponding work order sample document.

[0139] Step d2: According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial work order description information generation model is trained to obtain the trained work order description information generation model.

[0140] In a specific implementation, the initial work order description information generation model is trained through work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain a trained work order description information generation model.

[0141] Furthermore, in step d2, the initial work order description information generation model is trained according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model, including the following steps:

[0142] According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial bidirectional neural network is trained to obtain the trained bidirectional neural network; according to the work order sample documents in each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain the trained codec network corresponding to each field.

[0143] In a specific implementation, the work order description information generation model includes a bidirectional neural network and multiple codec networks. Different codec networks are used to process work order documents in different fields. Therefore, the initial bidirectional neural network can be trained together according to the work order sample documents in each field and the first sample description corresponding to each work order sample document to obtain a trained bidirectional neural network; according to the work order sample documents in each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain a trained codec network corresponding to each field.

[0144] Furthermore, in step d2, the initial work order description information generation model is trained according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model, including the following steps:

[0145] Step d21: input each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents.

[0146] In a specific implementation, each work order sample document is input into the initial work order description information generation model, and the second sample description of each work order sample document can be obtained, that is, the sample description of each work order sample document generated by the machine is obtained.

[0147] Step d22: According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain a trained work order description information generation model.

[0148] In a specific implementation, according to the first sample description and the second sample description corresponding to each work order sample document, the parameters in the initial work order description information generation model can be adjusted to obtain a trained work order description information generation model.

[0149] Furthermore, in step d22, according to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model, including the following steps:

[0150] According to the preset quality assessment standard, the first sample description and the second sample description corresponding to each of the work order sample documents are compared to obtain the comparison results corresponding to each of the work order sample documents; according to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0151] In a specific implementation, the first sample description and the second sample description corresponding to each work order sample document are compared according to a preset quality assessment standard to obtain a comparison result corresponding to each work order sample document. Here, the quality assessment standard can be any one of the ROUGE evaluation method, the pyramid evaluation method, and the BE evaluation method. Then, according to the comparison results corresponding to each work order sample document, the parameters in the initial work order description information generation model are adjusted to obtain a trained work order description information generation model.

[0152] S202: Acquire a target work order document, and determine a plurality of statements contained in the target work order document.

[0153] S203: Input the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document.

[0154] S204: Select key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the respective sentences.

[0155] S205: Generate work order description information of the target work order document according to the key sentence and the trained codec network.

[0156] The description of steps S202-S205 may be included in the description of steps S101-S104, and can achieve the same technical effect, so it will not be elaborated on again.

[0157] In an embodiment of the present application, by inputting multiple sentences contained in the target work order document into a bidirectional neural network, a sentence vector corresponding to each sentence can be generated, and the sentence vector is used to characterize the context information of the sentence in the target work order document, and according to the sentence vector corresponding to each sentence, a key sentence can be selected from the multiple sentences contained in the target work order document, and then, according to the key sentence and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, by machine-generated work order description information, the efficiency of generating work order description information can be improved, and through the bidirectional neural network, key sentences can be extracted from the target work order document, and through the encoding and decoding network, the human thinking process can be simulated, and then, the accuracy of the generated work order description information can be improved.

[0158] Based on the same application concept, the embodiment of the present application also provides a device for generating work order description information corresponding to the method for generating work order description information provided in the above embodiment. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the method for generating work order description information in the above embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0159] like Figures 3 to 8 As shown, Figure 3 One of the functional module diagrams of a device 300 for generating work order description information provided in an embodiment of the present application is shown; Figure 4 Shows Figure 3 Functional module diagram of the first generation module 320; Figure 5 Shows Figure 3 Select module 330 functional module diagram; Figure 6 Shows Figure 3 The second generation module 340 module diagram; Figure 7 A second module diagram of a device 300 for generating work order description information provided in an embodiment of the present application is shown; Figure 8 Shows Figure 7 Module diagram of the third generation module 350.

[0160] like Figure 3 As shown, the work order description information generating device 300 includes:

[0161] The acquisition module 310 is used to acquire a target work order document and determine a plurality of statements contained in the target work order document;

[0162] A first generating module 320 is used to input the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document;

[0163] A selection module 330, configured to select key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences;

[0164] The second generating module 340 is used to generate the work order description information of the target work order document according to the key sentence and the trained codec network.

[0165] In a possible implementation, Figure 4 As shown, the bidirectional neural network includes a forward neural network and a reverse neural network; the first generation module 320 includes:

[0166] The first generating unit 321 is used to input the multiple sentences into the trained forward neural network to obtain a forward vector corresponding to each sentence, and input the multiple sentences into the trained reverse neural network to obtain a reverse vector corresponding to each sentence; the forward vector is used to represent the previous information of the corresponding sentence in the target work order document; the reverse vector is used to represent the following information of the corresponding sentence in the target work order document;

[0167] The second generating unit 322 is used to merge the forward vector and the reverse vector corresponding to each sentence to generate the sentence vector corresponding to each sentence.

[0168] In a possible implementation, Figure 5 As shown, the selection module 330 includes:

[0169] A determination unit 331 is used for inputting a sentence vector corresponding to each sentence in the plurality of sentences into a trained binary classification network to determine an importance probability of each sentence in the target work order document;

[0170] The selection unit 332 is configured to select key sentences from the multiple sentences included in the target work order document according to the importance probability of each sentence in the target work order document.

[0171] In a possible implementation, Figure 6 As shown, the second generating module 340 includes:

[0172] The third generating unit 341 is used to input the words contained in the key sentence into the trained encoding and decoding network to generate a word sequence;

[0173] The fourth generating unit 342 is used to generate the work order description information of the target work order document according to the word sequence.

[0174] In a possible implementation, Figure 6 As shown, the encoding and decoding network includes an encoding network and a decoding network; the third generating unit 341 is specifically used to generate the word sequence according to the following steps:

[0175] Input the words contained in the key sentence into the trained encoding network to generate an encoding vector;

[0176] The encoding vector is input into the trained decoding network to generate the word sequence.

[0177] In a possible implementation, Figure 7 As shown, the work order description information generating device 300 also includes:

[0178] The third generation module 350 is used to train the initial work order description information generation model to obtain the trained work order description information generation model;

[0179] Among them, the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

[0180] In a possible implementation, Figure 8 As shown, the third generation module 350 includes:

[0181] An acquisition unit 351 is used to acquire work order sample documents in various fields and a first sample description corresponding to each work order sample document;

[0182] The fifth generating unit 352 is used to train the initial work order description information generation model according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model.

[0183] In a possible implementation, Figure 8 As shown, the fifth generating unit 352 is used to obtain the trained work order description information generation model according to the following steps:

[0184] Inputting each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents;

[0185] According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0186] In a possible implementation, Figure 8 As shown, the fifth generating unit 352 is specifically used to obtain the trained work order description information generation model according to the following steps:

[0187] According to a preset quality assessment standard, the first sample description and the second sample description corresponding to each of the work order sample documents are compared to obtain a comparison result corresponding to each of the work order sample documents;

[0188] According to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

[0189] In a possible implementation, Figure 8 As shown, the fifth generating unit 352 is further used to obtain the trained work order description information generation model according to the following steps:

[0190] According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial bidirectional neural network is trained to obtain the trained bidirectional neural network;

[0191] According to the work order sample documents of each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain the trained codec network corresponding to each field.

[0192] In an embodiment of the present application, by inputting multiple sentences contained in the target work order document into a bidirectional neural network, a sentence vector corresponding to each sentence can be generated, and the sentence vector is used to characterize the context information of the sentence in the target work order document, and according to the sentence vector corresponding to each sentence, a key sentence can be selected from the multiple sentences contained in the target work order document, and then, according to the key sentence and the encoding and decoding network, the work order description information of the target work order document can be generated. Based on the above method, by machine-generated work order description information, the efficiency of generating work order description information can be improved, and through the bidirectional neural network, key sentences can be extracted from the target work order document, and through the encoding and decoding network, the human thinking process can be simulated, and then, the accuracy of the generated work order description information can be improved.

[0193] Based on the same application concept, see Fig. 9 As shown, it is a structural diagram of an electronic device 900 provided in an embodiment of the present application, including: a processor 910, a memory 920 and a bus 930, the memory 920 stores machine-readable instructions executable by the processor 910, when the electronic device 900 is running, the processor 910 communicates with the memory 920 through the bus 930, and the machine-readable instructions are executed by the processor 910 when running to perform the steps of the method for generating work order description information as described in any of the above embodiments.

[0194] Specifically, when the machine-readable instruction is executed by the processor 910, the following processing may be performed:

[0195] Obtain a target work order document, and determine a plurality of statements contained in the target work order document;

[0196] Input the multiple sentences into the trained bidirectional neural network to generate a sentence vector corresponding to each sentence; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document;

[0197] Selecting key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences;

[0198] The work order description information of the target work order document is generated according to the key sentence and the trained codec network.

[0199] Based on the same application concept, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for generating work order description information provided in the above embodiment are executed.

[0200] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, the above-mentioned method for generating work order description information can be executed. By generating the work order description information by a machine, the efficiency of generating the work order description information can be improved. Through a bidirectional neural network, key sentences can be extracted from the target work order document. Through a coding and decoding network, a human thinking process can be simulated, thereby improving the accuracy of the generated work order description information.

[0201] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0202] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0203] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0204] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0205] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for generating work order description information, characterized in that: The generation method comprises: Obtain a target work order document, and determine a plurality of statements contained in the target work order document; Input the multiple sentences into the forward neural network of the trained bidirectional neural network to obtain the forward vector corresponding to each sentence, and input the multiple sentences into the reverse neural network of the trained bidirectional neural network to obtain the reverse vector corresponding to each sentence, merge the forward vector and the reverse vector corresponding to each sentence to generate a sentence vector corresponding to each sentence; the forward vector is used to represent the previous information of the corresponding sentence in the target work order document; the reverse vector is used to represent the following information of the corresponding sentence in the target work order document; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document; Selecting key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences; Input the words contained in the key sentence into a trained codec network to generate a word sequence, and generate work order description information of the target work order document according to the word sequence; The step of merging the forward vector and the reverse vector corresponding to each sentence to generate a sentence vector corresponding to each sentence includes: Concatenate the forward vector and the reverse vector corresponding to each sentence to generate a sentence vector corresponding to each sentence; The encoding and decoding network includes an encoding network and a decoding network; the step of inputting the words contained in the key sentence into the trained encoding and decoding network to generate a word sequence includes: Input the words contained in the key sentence into the trained encoding network to generate an encoding vector; Input the encoding vector into the trained decoding network to generate the word sequence; The step of selecting key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences includes: For each of the multiple sentences, input a sentence vector corresponding to each sentence into a trained binary classification network to determine the importance probability of each sentence in the target work order document; Selecting key sentences from the multiple sentences included in the target work order document according to the importance probability of each sentence in the target work order document; The generating method further comprises: Obtain work order sample documents in various fields and the first sample description corresponding to each work order sample document; According to the work order sample documents in various fields and the first sample description corresponding to each work order sample document, the initial bidirectional neural network is trained to obtain the trained bidirectional neural network; According to the work order sample documents of each field and the first sample description corresponding to each work order sample document, the initial codec network corresponding to each field is trained to obtain the trained codec network corresponding to each field.

2. The generation method according to claim 1, characterized in that: Before obtaining the target work order document and determining the multiple statements contained in the target work order document, the generating method further includes: Training the initial work order description information generation model to obtain the trained work order description information generation model; Among them, the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

3. The generation method according to claim 1, characterized in that: The initial work order description information generation model is trained according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained work order description information generation model, including: Inputting each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents; According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

4. The generation method according to claim 3, characterized in that: The adjusting the parameters in the initial work order description information generation model according to the first sample description and the second sample description corresponding to each of the work order sample documents to obtain the trained work order description information generation model includes: According to a preset quality assessment standard, the first sample description and the second sample description corresponding to each of the work order sample documents are compared to obtain a comparison result corresponding to each of the work order sample documents; According to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

5. A device for generating work order description information, characterized in that: The generating device comprises: An acquisition module, used to acquire a target work order document and determine a plurality of statements contained in the target work order document; The first generation module is used to input the multiple sentences into the forward neural network of the trained bidirectional neural network to obtain the forward vector corresponding to each sentence, and input the multiple sentences into the reverse neural network of the trained bidirectional neural network to obtain the reverse vector corresponding to each sentence, and merge the forward vector and the reverse vector corresponding to each sentence to generate a sentence vector corresponding to each sentence; the forward vector is used to represent the previous information of the corresponding sentence in the target work order document; the reverse vector is used to represent the following information of the corresponding sentence in the target work order document; each sentence vector is used to represent the context information of the corresponding sentence in the target work order document; A selection module, configured to select key sentences from the multiple sentences included in the target work order document according to the sentence vectors corresponding to the sentences; A second generating module is used to input the words contained in the key sentence into a trained codec network to generate a word sequence, and generate work order description information of the target work order document according to the word sequence; The first generating module is specifically used to concatenate the forward vector and the reverse vector corresponding to each sentence to generate a sentence vector corresponding to each sentence; The second generation module includes a third generation unit; the encoding and decoding network includes an encoding network and a decoding network; the third generation unit is specifically used to input the words contained in the key sentence into the trained encoding network to generate an encoding vector; and input the encoding vector into the trained decoding network to generate the word sequence; The selection module includes a determination unit and a selection unit; The determining unit is used to input the sentence vector corresponding to each sentence in the multiple sentences into the trained binary classification network to determine the importance probability of each sentence in the target work order document; The selection unit is used to select key sentences from the multiple sentences included in the target work order document according to the importance probability of each sentence in the target work order document; The device further includes a third generating module, wherein the third generating module includes an acquiring unit and a fifth generating unit; The acquisition unit is used to acquire work order sample documents in various fields and a first sample description corresponding to each work order sample document; The fifth generating unit is used to train the initial bidirectional neural network according to the work order sample documents in various fields and the first sample description corresponding to each work order sample document to obtain the trained bidirectional neural network; according to the work order sample documents in each field and the first sample description corresponding to each work order sample document, train the initial codec network corresponding to each field to obtain the trained codec network corresponding to each field.

6. The generating device according to claim 5, characterized in that The third generation module is used to train the initial work order description information generation model to obtain the trained work order description information generation model; Among them, the work order description information generation model includes one bidirectional neural network and multiple codec networks; different codec networks are used to process work order documents in different fields.

7. The generating device according to claim 5, characterized in that The fifth generating unit is further configured to obtain the trained work order description information generation model according to the following steps: Inputting each of the work order sample documents into the initial work order description information generation model to generate a second sample description corresponding to each of the work order sample documents; According to the first sample description and the second sample description corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

8. The generating device according to claim 7, characterized in that The fifth generation unit is specifically used to obtain the trained work order description information generation model according to the following steps: comparing the first sample description and the second sample description corresponding to each of the work order sample documents according to a preset quality assessment standard to obtain a comparison result corresponding to each of the work order sample documents; According to the comparison results corresponding to each of the work order sample documents, the parameters in the initial work order description information generation model are adjusted to obtain the trained work order description information generation model.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the method for generating work order description information as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for generating work order description information as claimed in any one of claims 1 to 4 are executed.