Text data processing method and device

Abstract

The present invention provides a text data processing method and device, the method comprising: obtaining a description text of the reason for return order; determining the type of the description text of the reason for return order; the type is the first type or the second type, the first type indicates that the description text of the reason for return order is an open text filled in by the user, and the second type indicates that the description text of the reason for return order is a text selected by the user in a set of return order options; if the type of the description text of the reason for return order is the first type, semantic recognition is performed on the description text of the reason for return order to obtain a semantic recognition result of the description text of the reason for return order; if it is determined that the semantic recognition result meets the pre-set text supplement condition, the semantic recognition result is added to the set of return order options. That is, semantic recognition can be performed on the description text of the reason for return order filled in by the user to obtain a generalized semantic recognition result, and supplement it to the set of return order options, which can optimize the subsequent business process so that the user can accurately describe the reason for return order in the subsequent process.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a text data processing method and device. Background Art

[0002] During the process of reviewing business, for business requests that do not meet the requirements, users will select the corresponding return reason from the set reason option menu to provide feedback.

[0003] However, the options in the menu are limited. If the business conditions are complex or the conditions have changed, some abnormal return orders may not fit the options in the menu, resulting in the user being unable to accurately describe the reason for the return order. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a text data processing method so that users can accurately describe the reason for the return order.

[0005] The present invention also provides a text data processing device to ensure the implementation and application of the above method in practice.

[0006] A text data processing method, comprising:

[0007] Get the description text of the return order reason;

[0008] Determine the type of the chargeback reason description text; the type is one of a first type and a second type, the first type indicates that the chargeback reason description text is an open text filled in by the user, and the second type indicates that the chargeback reason description text is a text selected by the user from a preset chargeback option set;

[0009] When the type of the return order reason description text is the first type, performing semantic recognition on the return order reason description text to obtain a semantic recognition result of the return order reason description text;

[0010] If it is determined that the semantic recognition result meets the preset text supplement condition, the semantic recognition result is added to the return order option set.

[0011] The above method may optionally further include:

[0012] When the type of the return order reason description text is the second type, determining the classification dimension to which the return order reason description text belongs;

[0013] After receiving the user's filtering instruction for the classification dimension, a summary analysis report of the classification dimension is generated based on the return order reason description text.

[0014] The above method, optionally, the process of determining whether the semantic recognition result satisfies a preset text supplementation condition includes:

[0015] Determine the current cumulative number of occurrences of the semantic recognition result;

[0016] If the cumulative number of occurrences of the semantic recognition result is greater than a preset number of occurrences threshold, it is determined that the semantic recognition result meets the text supplementation condition.

[0017] In the above method, optionally, the process of determining whether the semantic recognition result satisfies a preset text supplementation condition includes:

[0018] Determining the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each pre-stored candidate semantic recognition result;

[0019] Sort the semantic recognition result and each of the candidate semantic recognition results in descending order according to the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results;

[0020] If the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold, it is determined that the semantic recognition result meets the text supplementation condition.

[0021] The above method, optionally, after adding the semantic recognition result to the return order option set, further comprises:

[0022] In response to the user's instruction to select a return order reason, the return order option set to which the semantic recognition result has been added is displayed.

[0023] A text data processing device, comprising:

[0024] The acquisition unit is used to obtain the description text of the return order reason;

[0025] a first determining unit, configured to determine a type of the chargeback reason description text; the type is one of a first type and a second type, the first type indicating that the chargeback reason description text is an open text filled in by the user, and the second type indicating that the chargeback reason description text is a text selected by the user from a preset chargeback option set;

[0026] A recognition unit, configured to, when the type of the return order reason description text is the first type, perform semantic recognition on the return order reason description text to obtain a semantic recognition result of the return order reason description text;

[0027] An execution unit is used to add the semantic recognition result to the return order option set if it is determined that the semantic recognition result meets a preset text supplement condition.

[0028] The above device may optionally further include:

[0029] A second determining unit, configured to determine, when the type of the return order reason description text is the second type, the classification dimension to which the return order reason description text belongs;

[0030] A generating unit is used to generate a summary analysis report of the classification dimension based on the description text of the return order reason after receiving a filtering instruction from a user for the classification dimension.

[0031] In the above device, optionally, the execution unit includes:

[0032] A first determining subunit, used to determine the current cumulative number of occurrences of the semantic recognition result;

[0033] The second determining subunit is used to determine that the semantic recognition result meets the text supplementation condition if the cumulative occurrence times of the semantic recognition result is greater than a preset occurrence times threshold.

[0034] In the above device, optionally, the execution unit includes:

[0035] A third determination subunit is used to determine the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each pre-stored candidate semantic recognition result;

[0036] A sorting subunit, used to sort the semantic recognition result and each of the candidate semantic recognition results in descending order according to the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results;

[0037] The fourth determining subunit is configured to determine that the semantic recognition result satisfies the text supplementation condition if the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold.

[0038] The above device may optionally further include:

[0039] A display unit is used to display the return order option set to which the semantic recognition result has been added in response to the user's return order reason selection instruction.

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] The present invention provides a text data processing method and device, which can first obtain a description text of the reason for return order; determine the type of the description text of the reason for return order; the type is one of the first type and the second type, the first type indicates that the description text of the reason for return order is an open text filled in by the user, and the second type indicates that the description text of the reason for return order is a text selected by the user in a preset set of return order options; when the type of the description text of the reason for return order is the first type, semantic recognition is performed on the description text of the reason for return order to obtain a semantic recognition result of the description text of the reason for return order; if it is determined that the semantic recognition result meets the preset text supplement condition, the semantic recognition result is added to the return order option set. That is to say, by applying the scheme provided by the present invention, the user can fill in an open text or select a text from the return order option set as the description text of the reason for return order, so that when the options of the return order option set are limited, the user can still accurately describe the reason for return order, and the description text of the reason for return order filled in by the user can be semantically recognized to obtain a generalized semantic recognition result, and supplement it to the return order option set, which can optimize the subsequent business process, so that the user can accurately describe the reason for return order in the subsequent process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0043] Figure 1 A method flow chart of a text data processing method provided by the present invention;

[0044] Figure 2 A flowchart of a process for determining that a semantic recognition result meets a text supplementation condition provided by the present invention;

[0045] Figure 3 A flowchart of another process of determining that a semantic recognition result satisfies a text supplementation condition provided by the present invention;

[0046] Figure 4 A flowchart of a chargeback reason processing process provided by the present invention;

[0047] Figure 5 A structural schematic diagram of a text data processing device provided by the present invention;

[0048] Figure 6 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0051] The embodiment of the present invention provides a text data processing method, which can be applied to electronic devices. The method flow chart of the method is as follows: Figure 1 As shown, specifically including:

[0052] S101: Obtain the description text of the return order reason.

[0053] In the embodiment of the present invention, the return order reason description text is a text fed back by the user regarding a service application that does not meet the conditions, and is used to describe the return order reason of the service request.

[0054] S102: Determine the type of the return reason description text; the type is one of a first type and a second type, the first type indicates that the return reason description text is an open text filled in by the user, and the second type indicates that the return reason description text is a text selected by the user from a pre-set return option set.

[0055] In the embodiment of the present invention, the chargeback reason description text may be the text edited by the user in the chargeback reason text box; or it may be an option selected by the user in the chargeback option set, where the option is text.

[0056] The return order option set may be in the form of a list menu, and may be displayed after the user clicks a preset option control.

[0057] S103: When the type of the return order reason description text is the first type, semantic recognition is performed on the return order reason description text to obtain a semantic recognition result of the return order reason description text.

[0058] In the embodiment of the present invention, semantic recognition can be performed on the return order reason description text by using a semantic recognition model with a predetermined semantic recognition function to obtain a semantic recognition result of the return order reason description text.

[0059] S104: If it is determined that the semantic recognition result meets the preset text supplement condition, the semantic recognition result is added to the return order option set.

[0060] In an embodiment of the present invention, the text supplementation condition may be that the cumulative number of occurrences of the semantic recognition result is greater than a preset number threshold, or the frequency of occurrence within a preset time period is greater than a preset frequency threshold, etc.

[0061] Among them, the semantic recognition result is added to the return order option set, so that the semantic recognition result can be used as a new return order reason option text.

[0062] By applying the method provided in the embodiment of the present invention, the user can fill in open text or select text from the return option set as the return reason description text, so that when the options in the return option set are limited, the user can still accurately describe the return reason, and the return reason description text filled in by the user can be semantically recognized to obtain a generalized semantic recognition result and supplement it to the return option set, which can optimize subsequent business processes so that users can accurately describe the return reason in subsequent processes.

[0063] In the embodiment provided by the present invention, based on the above implementation process, specifically, the text data processing further includes:

[0064] When the type of the return order reason description text is the second type, determining the classification dimension to which the return order reason description text belongs;

[0065] After receiving the user's filtering instruction for the classification dimension, a summary analysis report of the classification dimension is generated based on the return order reason description text.

[0066] Among them, the classification dimension is pre-set, and the number of classification dimensions to which the return order reason description text belongs can be one or more, for example, it can be dimensions such as branch, position, cycle, etc.

[0067] After receiving the user's filtering instructions for the classification dimension, the summary analysis report of the classification dimension is improved based on the description text of each return order reason under the classification dimension, thereby improving readability.

[0068] In the embodiment provided by the present invention, based on the above implementation process, specifically, a feasible way to determine that the semantic recognition result meets the preset text supplementation condition is as follows: Figure 2 As shown, including:

[0069] S201: Determine the current cumulative number of occurrences of the semantic recognition result.

[0070] Wherein, each time the semantic recognition result is recognized, the cumulative number of occurrences of the semantic recognition result may be increased by one to obtain the current cumulative number of occurrences of the semantic recognition result.

[0071] S202: If the cumulative number of occurrences of the semantic recognition result is greater than a preset number of occurrences threshold, it is determined that the semantic recognition result meets the text supplementation condition.

[0072] In the embodiment of the present invention, the occurrence number threshold can be set according to actual needs.

[0073] Optionally, if the cumulative number of occurrences of the semantic recognition result is not greater than the occurrence threshold, it can be determined that the semantic recognition result does not meet the text supplementation condition.

[0074] In the embodiment provided by the present invention, based on the above implementation process, specifically, another feasible way to determine whether the semantic recognition result satisfies the preset text supplementation condition is as follows: Figure 3 As shown, including:

[0075] S301: Determine the current cumulative occurrence count of the semantic recognition result and the cumulative occurrence count of each pre-stored candidate semantic recognition result.

[0076] The candidate semantic recognition result may be a previously appeared historical semantic recognition result.

[0077] S302: Sort the semantic recognition result and each of the candidate semantic recognition results in descending order of the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results.

[0078] The semantic recognition results and each candidate semantic recognition result are sorted to obtain the sorting values ​​of the semantic recognition results and each candidate semantic recognition result. The sorting value of the semantic recognition result is the sequence number of the arrangement order of the semantic recognition results.

[0079] S303: If the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold, it is determined that the semantic recognition result meets the text supplementation condition.

[0080] The size of the sorting threshold can be set by a technician according to actual needs, for example, it can be set to 1, 2, 5 or 10, etc.

[0081] Optionally, if the ranking value of the semantic recognition result is greater than the ranking threshold, it is determined that the semantic recognition result does not meet the text supplementation condition.

[0082] In an embodiment of the present invention, the semantic recognition result is used as a new option text and added to the return option set. If the number of option texts in the return option set to which the semantic recognition result has been added is greater than or equal to a preset quantity threshold, the N option texts with the smallest number of occurrences in the return option set can be removed, so that the number of option texts in the return option set is less than the quantity threshold.

[0083] In the embodiment provided by the present invention, based on the above implementation process, specifically, after adding the semantic recognition result to the return order option set, it also includes:

[0084] In response to the user's instruction to select a return order reason, the return order option set to which the semantic recognition result has been added is displayed.

[0085] The return order reason selection instruction may be an instruction triggered by a user clicking a preset control when a new business exception return order occurs.

[0086] The user can select an option text from the displayed return order option set as the return order reason description text.

[0087] The text data processing method provided by the embodiment of the present invention can be applied to various business fields. For example, it can be used in the parameter change process of a bank. When submitting a parameter change application, the applicant needs to correctly fill in the application form, select the correct circulation link, submit reasonable and standardized attachments, etc. If the parameter change application does not meet the parameter change conditions, it will cause a circuitous process, affect the timeliness of the parameter change, and then affect the business development.

[0088] The abnormal business return in this scenario can be an abnormal parameter return, which means that a rollback occurs at any stage in the parameter change process.

[0089] In the related technology, there is a lack of methods for collecting and classifying the reasons for order returns due to abnormal parameters, and it is impossible to propose effective methods to reduce the order return rate for different order return situations. Repeated order returns not only fail to meet the high-timeliness parameter operation and maintenance requirements, but also greatly reduce the user experience and increase the redundant pressure of the system.

[0090] Therefore, in the embodiment of the present invention, the reasons for abnormal parameter return orders are collected and classified, which can effectively distinguish the return order situations. Different optimization strategies are adopted for different return order situations to effectively reduce the return order rate, improve the efficiency of parameter application circulation and user experience, and improve the efficiency of parameter operation and maintenance. For the specific implementation process, please refer to Figure 4, is a flow chart of a chargeback reason processing process provided by an embodiment of the present invention, and the specific implementation process is as follows:

[0091] When an abnormal return occurs during the process, the user can select the abnormal return reason parameter from the drop-down list; or fill in the text content to implement an open description.

[0092] After selecting the drop-down list, the selection result will enter the parameter abnormality return order reason classification statistics module for statistics. Finally, based on the statistical results, the user can filter by conditions of different dimensions to finally form a summary analysis report on the parameter abnormality return order reasons.

[0093] After filling in the open text parameter abnormal return reason, the text analysis and extraction module analyzes the text entered by the user, recompiles it, forms a brief summary of the return reason, and transmits it to the abnormal return reason list iteration module. The abnormal return reason list iteration module sorts the stored data by high and low frequency, and pushes the high-frequency abnormal return reason above the threshold to the abnormal return reason list. When the user encounters an abnormal parameter return scenario in the future, he can select more abnormal return reasons through the interface. The entire process uses the text entered by the user to realize the system's automatic iteration and update of the return reason list, and finally forms a feedback process driven by user behavior habits.

[0094] By applying the method provided in the embodiment of the present invention, based on the text input by the user, the drop-down selection list of abnormal return reasons is iteratively updated, which helps to improve the user experience of the platform. The summary and analysis report of the reasons for abnormal return of parameters helps the parameter proposal department to effectively manage the parameter process and improve the efficiency of parameter circulation. In the process of parameter circulation, repeated approvals caused by return orders are reduced, and the work pressure of the approval department is reduced.

[0095] and Figure 1 Corresponding to the method described above, the embodiment of the present invention further provides a text data processing device for Figure 1 In order to implement the method in the text data processing apparatus provided in the embodiment of the present invention, the text data processing apparatus can be applied to an electronic device, and its structural diagram is as follows: Figure 5 As shown, specifically including:

[0096] An acquisition unit 501 is used to acquire a description text of a return order reason;

[0097] A first determining unit 502 is used to determine the type of the chargeback reason description text; the type is one of a first type and a second type, the first type indicates that the chargeback reason description text is an open text filled in by the user, and the second type indicates that the chargeback reason description text is a text selected by the user from a preset chargeback option set;

[0098] The recognition unit 503 is used to perform semantic recognition on the return order reason description text when the type of the return order reason description text is the first type, and obtain a semantic recognition result of the return order reason description text;

[0099] The execution unit 504 is configured to add the semantic recognition result to the return order option set if it is determined that the semantic recognition result meets a preset text supplement condition.

[0100] In an embodiment provided by the present invention, based on the above implementation process, specifically, it also includes:

[0101] A second determining unit, configured to determine, when the type of the return order reason description text is the second type, the classification dimension to which the return order reason description text belongs;

[0102] A generating unit is used to generate a summary analysis report of the classification dimension based on the description text of the return order reason after receiving a filtering instruction from a user for the classification dimension.

[0103] In an embodiment provided by the present invention, based on the above implementation process, specifically, the execution unit 504 includes:

[0104] A first determining subunit, used to determine the current cumulative number of occurrences of the semantic recognition result;

[0105] The second determining subunit is used to determine that the semantic recognition result meets the text supplementation condition if the cumulative occurrence times of the semantic recognition result is greater than a preset occurrence times threshold.

[0106] In an embodiment provided by the present invention, based on the above implementation process, specifically, the execution unit 504 includes:

[0107] A third determination subunit is used to determine the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each pre-stored candidate semantic recognition result;

[0108] A sorting subunit, used to sort the semantic recognition result and each of the candidate semantic recognition results in descending order according to the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results;

[0109] The fourth determining subunit is configured to determine that the semantic recognition result satisfies the text supplementation condition if the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold.

[0110] In an embodiment provided by the present invention, based on the above implementation process, specifically, it also includes:

[0111] A display unit is used to display the return order option set to which the semantic recognition result has been added in response to the user's return order reason selection instruction.

[0112] The specific principles and execution processes of each unit and module in the text data processing device disclosed in the above embodiment of the present invention are the same as those of the text data processing method disclosed in the above embodiment of the present invention. Please refer to the corresponding parts of the text data processing method provided in the above embodiment of the present invention, and will not be repeated here.

[0113] An embodiment of the present invention further provides a storage medium, which includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the above-mentioned text data processing method.

[0114] The embodiment of the present invention further provides an electronic device, the structural diagram of which is shown in FIG. Figure 6 As shown, it specifically includes a memory 601 and one or more instructions 602, wherein the one or more instructions 602 are stored in the memory 601 and are configured to be executed by one or more processors 603 to perform the following operations:

[0115] Get the description text of the return order reason;

[0116] Determine the type of the chargeback reason description text; the type is one of a first type and a second type, the first type indicates that the chargeback reason description text is an open text filled in by the user, and the second type indicates that the chargeback reason description text is a text selected by the user from a preset chargeback option set;

[0117] When the type of the return order reason description text is the first type, performing semantic recognition on the return order reason description text to obtain a semantic recognition result of the return order reason description text;

[0118] If it is determined that the semantic recognition result meets the preset text supplement condition, the semantic recognition result is added to the return order option set.

[0119] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0120] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0121] For the convenience of description, the above device is described as being divided into various units according to their functions. Of course, when implementing the present invention, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0122] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0123] The above is a detailed introduction to a text data processing method provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A text data processing method, characterized in that: include: Get the description text of the return order reason; Determine the type of the chargeback reason description text; the type is one of a first type and a second type, the first type indicates that the chargeback reason description text is an open text filled in by the user, and the second type indicates that the chargeback reason description text is a text selected by the user from a preset chargeback option set; When the type of the return order reason description text is the first type, performing semantic recognition on the return order reason description text to obtain a semantic recognition result of the return order reason description text; If it is determined that the semantic recognition result meets the preset text supplement condition, the semantic recognition result is added as a new option text to the return order option set; The process of determining whether the semantic recognition result meets the preset text supplementation condition includes: Determine the current cumulative number of occurrences of the semantic recognition result; If the cumulative number of occurrences of the semantic recognition result is greater than a preset number of occurrences threshold, it is determined that the semantic recognition result meets the text supplementation condition.

2. The method according to claim 1, characterized in that Also includes: When the type of the return order reason description text is the second type, determining the classification dimension to which the return order reason description text belongs; After receiving the user's filtering instruction for the classification dimension, a summary analysis report of the classification dimension is generated based on the return order reason description text.

3. The method according to claim 1, characterized in that The process of determining whether the semantic recognition result meets the preset text supplementation condition includes: Determining the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each pre-stored candidate semantic recognition result; Sort the semantic recognition result and each of the candidate semantic recognition results in descending order according to the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results; If the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold, it is determined that the semantic recognition result meets the text supplementation condition.

4. The method according to claim 1, characterized in that: After adding the semantic recognition result to the return order option set, the method further includes: In response to the user's instruction to select a return order reason, the return order option set to which the semantic recognition result has been added is displayed.

5. A text data processing device, characterized in that: include: The acquisition unit is used to obtain the description text of the return order reason; A first determining unit is used to determine the type of the return order reason description text; the type is one of a first type and a second type, the first type indicates that the return order reason description text is an open text filled in by the user, and the second type indicates that the return order reason description text is a text selected by the user from a preset return order option set; A recognition unit, configured to, when the type of the return order reason description text is the first type, perform semantic recognition on the return order reason description text to obtain a semantic recognition result of the return order reason description text; An execution unit, configured to add the semantic recognition result as a new option text to the return order option set if it is determined that the semantic recognition result meets a preset text supplement condition; Wherein, the execution unit includes: A first determining subunit, used to determine the current cumulative number of occurrences of the semantic recognition result; The second determining subunit is used to determine that the semantic recognition result meets the text supplementation condition if the cumulative occurrence times of the semantic recognition result is greater than a preset occurrence times threshold.

6. The device according to claim 5, characterized in that Also includes: A second determining unit, configured to determine, when the type of the return order reason description text is the second type, the classification dimension to which the return order reason description text belongs; A generating unit is used to generate a summary analysis report of the classification dimension based on the description text of the return order reason after receiving a filtering instruction from a user for the classification dimension.

7. The device according to claim 5, characterized in that The execution unit comprises: A third determination subunit is used to determine the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each pre-stored candidate semantic recognition result; A sorting subunit, used to sort the semantic recognition result and each of the candidate semantic recognition results in descending order according to the current cumulative number of occurrences of the semantic recognition result and the cumulative number of occurrences of each of the candidate semantic recognition results; The fourth determining subunit is configured to determine that the semantic recognition result satisfies the text supplementation condition if the ranking value of the semantic recognition result is less than or equal to a preset ranking threshold.

8. The device according to claim 5, characterized in that Also includes: A display unit is used to display the return order option set to which the semantic recognition result has been added in response to the user's return order reason selection instruction.

Images