Method and device for analyzing user experience of bank counter service product

Abstract

The application provides a user experience analysis method and device for bank counter service products, and relates to the technical field of natural language processing, which comprises the following steps: forming a service product word library and a service product function word library respectively; performing word segmentation on the sentiment tendency text of the bank counter service to be determined; matching the word segmentation result with the key words in the service product word library and the service product function word library, and determining the bank counter service product or function pointed by the text according to the matched key words; performing sentiment tendency classification on the sentiment tendency text by using a text sentiment classification model to obtain the belonging sentiment category; and analyzing the user experience of the counter service product or function based on the sentiment category. The application can accurately locate the specific product or function corresponding to the text, and then perform sentiment analysis on the text, so that the analysis result can depict the user's local experience in a more fine-grained manner, which is beneficial to the continuous improvement of the product and function, and can save the manual cost of traditional evaluation collection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of natural language processing, and particularly relates to a user experience analysis method and device for a bank counter service product. BACKGROUND

[0002] The bank counter service is complex, and the details are different in actual situations under the same service, so different processing procedures are often needed to solve the service problems. In the current products for counter service, the evaluation of product functions is generally collected by a special demand operation team for a certain number of grassroots colleagues and is comprehensively judged in combination with the feedback channels specially provided in the product. Although the related technology of using artificial intelligence algorithm to analyze user experience has appeared in the prior art, the data sources to be analyzed are all pre-determined to belong to a certain product. Therefore, the user experience analysis of the bank counter service product has the following technical defects:

[0003] The complexity of the bank counter service leads to the fact that the artificial intelligence algorithm cannot accurately locate the product and the specific function to which the data source to be analyzed is directed, and the manual collection of evaluation opinions consumes human and resource costs and cannot be expanded for continuous collection. SUMMARY

[0004] Therefore, the present application provides a user experience analysis method and device for a bank counter service product to solve at least one of the above problems.

[0005] In order to achieve the above purpose, the present application adopts the following scheme:

[0006] According to a first aspect of the present application, a user experience analysis method for a bank counter service product is provided, which comprises: forming a service product word library and a service product function word library based on bank counter service product information and service product function information respectively; performing word segmentation on the emotional tendency text of the bank counter service to be determined; matching the word segmentation result with the key words in the service product word library and the service product function word library to obtain matching key words, and determining the bank counter service product or function to which the emotional tendency text of the bank counter service to be determined points according to the matching key words; performing emotional tendency classification on the emotional tendency text of the bank counter service to be determined by using a text emotion classification model to obtain the emotional category to which the emotional tendency text of the bank counter service to be determined belongs; and analyzing the user experience of the corresponding bank counter service product or function based on the emotional category.

[0007] According to a second aspect of the present application, there is provided a device for analyzing user experience of a bank counter service product, comprising: a word library generating unit configured to generate a service product word library and a service product function word library based on bank counter service product information and service product function information respectively; a word segmentation unit configured to perform word segmentation on a text of emotional tendency of a bank counter service to be judged; a matching unit configured to match a word segmentation result with keywords in the service product word library and the service product function word library to obtain matching keywords, and determine a bank counter service product or function to which the text of emotional tendency of the bank counter service to be judged points according to the matching keywords; an emotional classification unit configured to perform emotional tendency classification on the text of emotional tendency of the bank counter service to be judged by using a text emotional classification model to obtain an emotional category to which the text of emotional tendency of the bank counter service to be judged belongs; and an experience analysis unit configured to analyze user experience of a corresponding bank counter service product or function based on the emotional category.

[0008] According to a third aspect of the present application, there is provided an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0009] According to a fourth aspect of the present application, there is provided a computer readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0010] According to a fifth aspect of the present application, there is provided a computer program product comprising computer programs / instructions, wherein the computer programs / instructions implement the steps of the above method when executed by a processor.

[0011] According to the above technical solution, the method and device for analyzing user experience of a bank counter service product provided by the present application first generate a bank counter service product word library and a service product function word library, then perform word segmentation on a text to be judged, and match the service product word library and the service product function word library to obtain keywords related to products or functions, so that when facing complex counter service products, the text to be judged can be accurately located to a specific product or function, and then emotional analysis is performed thereon, and the analysis result can depict user experience of a local bank counter service product in a more fine-grained manner, which is beneficial to continuous improvement of products and functions. In addition, the present application can save labor cost of traditional evaluation collection. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor. In the drawings:

[0013] Figure 1 is a flowchart of a user experience analysis method for a bank counter service product provided by an embodiment of the present application;

[0014] Figure 2 is a training flowchart of a text sentiment classification model provided by an embodiment of the present application;

[0015] Figure 3 is a flowchart of a user experience analysis method for a bank counter service product provided by another embodiment of the present application;

[0016] Figure 4 is a training flowchart of an entity recognition model provided by an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of a user experience analysis method for a bank counter service product provided by an embodiment of the present application;

[0018] Figure 6 is a structural schematic diagram of a user experience analysis device for a bank counter service product provided by an embodiment of the present application;

[0019] Figure 7 is a schematic block diagram of a system structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will further describe the embodiments of the present application in combination with the drawings. Herein, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but not as a limitation of the present application.

[0021] In order to solve the problems in the prior art that the evaluation of the bank counter service product and its function cannot be continuously collected, and due to the limited number of front-line staff in the evaluation collection process, the comprehensive evaluation of the product function cannot be obtained, a user experience analysis method for a bank counter service product is provided to effectively analyze the preference tendency of the user in the process of using the bank counter service product, and to provide better data support for the product function requirement design stage.

[0022] As Figure 1As shown is a flowchart of a user experience analysis method of a bank counter service product provided by the embodiment of the application. It should be noted that the analysis data source of the application is the short text question, answer, comment and reply data exchanged between employees in the bank internal grassroots service support platform. The method comprises the following steps:

[0023] Step S101: Forming a service product word library and a service product function word library based on bank counter service product information and service product function information.

[0024] In the embodiment, the service product information and the service product function information can respectively refer to the service product name and the function name of the service product. Therefore, the related names of the bank counter service product used in the bank can be sorted and summarized to form the service product word library, and the related function names of the bank counter service product used in the bank can be sorted and summarized to form the service product function word library.

[0025] Step S102: Segmenting the emotional tendency text of the bank counter service to be judged.

[0026] Here, the emotional tendency text of the bank counter service to be judged is the short text question, answer, comment and reply information exchanged between employees collected from the bank internal grassroots service support platform in a specified period. For the emotional tendency text, the natural language processing technology (NLP) can be used in the embodiment to segment it, such as using the jieba tool to segment and extract the product or function. Specifically, the service product word library and the service product function word library can be used as the custom dictionary of the jieba segmentation tool, and then the emotional tendency text of the bank counter service to be judged is subjected to jieba segmentation processing, and the specific operation and processing process are as follows:

[0027] Segm text =jieba(text)

[0028] Wherein, Segm text represents the result of text segmentation; text represents the emotional tendency text of the bank counter service to be judged.

[0029] Step S103: Matching the segmentation result with the key words in the service product word library and the service product function word library to obtain the matching key words, and determining the bank counter service product or function pointed by the emotional tendency text of the bank counter service to be judged according to the matching key words. Since the product name and the function name are stored in the service product word library and the service product function word library, the product or function pointed by the text can be determined by matching the key words.

[0030] This step can be represented as follows:

[0031] res segm = Match (Segm text , Lib)

[0032] Wherein res segm represents the specific business product or function of the bank counter recognized by the word segmentation, and Lib represents the business product library and the business product function library.

[0033] Step S104: using a text sentiment classification model to classify the sentiment of the bank counter business to be judged, and obtaining the sentiment category to which the sentiment text of the bank counter business to be judged belongs. It should be pointed out that this step can be carried out at the same time as steps S102 and S103, and the results of this step and step S103 are used as the basis for analysis in step S105.

[0034] Preferably, as shown in Figure 2 , the text sentiment classification model is obtained by the following method:

[0035] Step S1041: obtaining relevant information of different products or functions. The relevant information here refers to historical questions, answers, comments and replies data for various products and functions in the bank. It should be pointed out that the historical data collected here can be historical data in the bank internal grassroots service support platform, or various historical data left by users or bank staff replies on the Internet.

[0036] Step S1042: labeling the sentiment tendency label for the relevant information to obtain the sentiment classification data set.

[0037] The sentiment tendency label here can be, for example, "good experience", "normal experience", "bad experience", etc. Of course, more other sentiment labels can also be labeled in the present application, and the embodiments of the present application do not limit this.

[0038] Step S1043: training the text sentiment classification model using the sentiment classification data set. The text sentiment classification model can be trained using existing deep learning methods, so that it can process the input text and output the user sentiment category.

[0039] Step S105: analyzing the user experience of the corresponding bank counter business product or function based on the sentiment category.

[0040] From the above technical solutions, the user experience analysis method of the bank counter service product provided by the application can generate a bank counter service product vocabulary and a service product function vocabulary, then perform word segmentation on the text to be judged, and match the service product vocabulary and the service product function vocabulary to obtain product or function related keywords, so that the specific product or function corresponding to the text to be judged can be accurately located when facing complex counter service products, and then the text to be judged is subjected to sentiment analysis, and the analysis result can more finely depict the user's experience of the local bank counter service product, which is beneficial to the continuous improvement of the product and function. In addition, the application can also save the labor cost of traditional evaluation collection.

[0041] As Figure 3 The flowchart of the user experience analysis method of the bank counter service product provided by another embodiment of the application is shown, and the method comprises the following steps:

[0042] Step S301: Forming a service product vocabulary and a service product function vocabulary based on bank counter service product information and service product function information.

[0043] Step S302: Performing word segmentation on the sentiment tendency text of the bank counter service to be judged.

[0044] Step S303: Matching the word segmentation result with the keywords in the service product vocabulary and the service product function vocabulary to obtain matching keywords.

[0045] Step S304: Using an entity recognition model to perform text entity recognition on the sentiment tendency text of the bank counter service to be judged, and outputting the recognition keywords corresponding to the sentiment tendency text of the bank counter service to be judged.

[0046] If only the word segmentation matching method is used to locate the service product or function, the context semantic information of the related content in the entire text content will be ignored, which may lead to incorrect analysis of part of the text content. To solve this problem, the embodiment collects a large amount of text information related to the product or function and the product or function corresponding to the text information, constructs a large amount of entity data set, and trains an entity recognition model for entity recognition.

[0047] Preferably, as Figure 4 The entity recognition model is trained by the following method:

[0048] Step S3041: Acquiring text information related to the bank counter service product or function.

[0049] In the embodiment, a large amount of evaluation information of user's product or function can be obtained from the Internet or the internal platform of the bank, and the text content of the product or function specified by the user is screened out, and nearly one million pieces of such comment text information are collected.

[0050] Step S3042: The text information is labeled with entity tags, and the specific product or function pointed by the text information is labeled to form an entity recognition data set.

[0051] Step S3043: The entity recognition model is trained using the entity recognition data set to obtain an entity recognition model. The entity recognition model used in the embodiment can be a Mengzi network model, and a pre-trained Mengzi model can be used as the initial network parameter. The learning rate and the number of iterations in the training process are set. Preferably, the learning rate in the embodiment can be set to 1x10e -5 , and the number of iterations can be set to 5 times. The model with the best classification result in the validation set in the training process is saved as the entity recognition model.

[0052] The trained entity recognition model Mengzi is used to determine the sentiment tendency text of the bank counter business product or function, and the product or function information that may exist in the text is obtained. The process of text entity recognition can be represented as follows:

[0053] res ner =Mengzi(text)

[0054] Where res ner represents the specific business product or function of the bank counter recognized by the entity recognition model.

[0055] Step S305: The matching keywords and the recognized keywords are fused and analyzed by using the voting method to determine the bank counter business product or function pointed by the sentiment tendency text of the bank counter business to be determined.

[0056] This step can be expressed by the following formula:

[0057] res cont =Vote(res segm ,res ner )

[0058] res cont represents the bank counter business product or function result obtained by the voting method.

[0059] The voting method here refers to the determination of a business product or function when both methods in steps S303 and S304 match or identify the corresponding business product or function words. In other words, when the matching keyword and the identified keyword are the same, the bank counter business product or function pointed to by the sentiment text of the bank counter business to be judged is determined based on the same keyword.

[0060] As can be seen, this application introduces an entity recognition model to perform entity recognition, enabling keywords to match the contextual semantic relationships of the entire text. The results are then fused and analyzed with the results of word segmentation and recognition, resulting in more accurate identification of products or functions.

[0061] Step S306: Use a text sentiment classification model to classify the sentiment tendency text of the bank counter service to be judged, and obtain the sentiment category to which the sentiment tendency text of the bank counter service to be judged belongs.

[0062] Step S307: Analyze the user experience of the corresponding bank counter service products or functions based on the emotional category.

[0063] The principles behind each of the above steps can be understood through... Figure 5 To indicate, such as Figure 5 As shown, the sentimental text of the bank counter transactions to be judged (i.e. Figure 5 The text describing the user's experience with the bank's counter services products undergoes two processes: "product or function description word localization" and "sentiment classification." The results of these two processes are then fused and analyzed to obtain a sentiment analysis result regarding the user's perception of the product or function. "Product or function description word localization" comprises two parts: text content entity recognition and text content word segmentation and matching. The results from both parts are then combined to locate the product or function description in the text. "Sentiment classification" utilizes a text sentiment classification model to output sentiment analysis results. Finally, the combined results yield an accurate sentiment analysis result for the text to be judged.

[0064] According to the technical scheme, the user experience analysis method of the bank counter service product provided by the application first generates a bank counter service product vocabulary and a service product function vocabulary, then performs word segmentation on the text to be judged, matches the service product vocabulary and the service product function vocabulary, obtains product or function related keywords, so that the specific product or function corresponding to the text to be judged can be accurately positioned when facing complex counter service products, and then the text to be judged is subjected to sentiment analysis, and the analysis result can depict the user's local experience of the bank counter service product in a more fine-grained manner, which is beneficial to the continuous improvement of the product and function, and can save the labor cost of traditional evaluation collection. In addition, the application not only uses the word segmentation matching method to position the service product or function, but also considers the context semantic information of the related content in the entire text content, uses an entity recognition model to perform entity recognition on the text to be judged, effectively reduces the error analysis of the text content to be judged, and makes the positioning of the product and function to which the text to be judged belongs more accurate.

[0065] As Figure 6 The structure of the user experience analysis device of the bank counter service product provided by the embodiment of the application is shown in the figure, and the device comprises a vocabulary generation unit 610, a word segmentation unit 620, a matching unit 630, a sentiment classification unit 640 and an experience analysis unit 650. The vocabulary generation unit 610 is connected to the word segmentation unit 620 and the matching unit 630 respectively, the matching unit 630 is further connected to the word segmentation unit 620 and the experience analysis unit 650 respectively, and the sentiment classification unit 640 is connected to the experience analysis unit 650.

[0066] The vocabulary generation unit 610 is used to form a service product vocabulary and a service product function vocabulary based on bank counter service product information and service product function information respectively.

[0067] The word segmentation unit 620 is used to perform word segmentation on the sentiment tendency text of the bank counter service to be judged.

[0068] The matching unit 630 is used to match the word segmentation result with the keywords in the service product vocabulary and the service product function vocabulary to obtain matching keywords, and determine the bank counter service product or function to which the sentiment tendency text of the bank counter service to be judged points according to the matching keywords.

[0069] The sentiment classification unit 640 is used to perform sentiment tendency classification on the sentiment tendency text of the bank counter service to be judged by using a text sentiment classification model, and obtain the sentiment category to which the sentiment tendency text of the bank counter service to be judged belongs.

[0070] The experience analysis unit 650 is used to analyze the user experience of the corresponding bank counter service product or function based on the sentiment category.

[0071] Preferably, the device of the embodiment further comprises an entity recognition unit configured to perform text entity recognition on the to-be-judged sentiment tendency text of the bank counter service by using an entity recognition model, and output an identified keyword corresponding to the to-be-judged sentiment tendency text of the bank counter service. The matching unit 630 determines the bank counter service product or function to which the to-be-judged sentiment tendency text of the bank counter service points according to the matching keyword, and further comprises: performing fusion analysis on the matching keyword and the identified keyword by using a voting method to determine the bank counter service product or function to which the to-be-judged sentiment tendency text of the bank counter service points.

[0072] Preferably, the matching unit 630 of the embodiment performs fusion analysis on the matching keyword and the identified keyword by using a voting method to determine the bank counter service product or function to which the to-be-judged sentiment tendency text of the bank counter service points comprises: performing fusion analysis on the matching keyword and the identified keyword by using a voting method, and when the matching keyword and the identified keyword are the same, determining the bank counter service product or function to which the to-be-judged sentiment tendency text of the bank counter service points based on the same keyword.

[0073] Preferably, the entity recognition model is trained in the following manner: obtaining text information related to bank counter service products or functions; performing entity label annotation on the text information to annotate specific products or functions to which the text information points, forming an entity recognition data set; and training the entity recognition model by using the entity recognition data set to obtain the entity recognition model.

[0074] Preferably, the entity recognition model is a Mengzi network model, and the training of the entity recognition model by using the entity recognition data set comprises: using a pre-trained Mengzi model as initial network parameters, setting a learning rate and an iteration number in the training process, preferably, the learning rate in the embodiment can be set to 1x10e -5 , and the iteration number can be set to 5 times, and the model with the best classification result in the validation set in the training process is saved as the entity recognition model.

[0075] Preferably, the text sentiment classification model is trained in the following manner: obtaining related information of different products or functions; annotating sentiment tendency labels for the related information to obtain a sentiment classification data set; and training a text sentiment classification model by using the sentiment classification data set.

[0076] From the above technical solutions, the user experience analysis device of the bank counter service product provided by the application generates a bank counter service product vocabulary and a service product function vocabulary first, then carries out word segmentation on the text to be judged, and matches the service product vocabulary and the service product function vocabulary to obtain product or function related keywords, so that when facing complex counter service products, the specific product or function corresponding to the text to be judged can be accurately located, and then the text to be judged is subjected to sentiment analysis, and the analysis result can more finely depict the user's experience of the local bank counter service product, which is beneficial to the continuous improvement of the product and function. In addition, the application can also save the labor cost of traditional evaluation collection.

[0077] The embodiment of the application further provides an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the above method when executing the program.

[0078] The embodiment of the application further provides a computer program product, including computer programs / instructions, and the computer programs / instructions implement the steps of the above method when executed by the processor.

[0079] The embodiment of the application further provides a computer readable storage medium, which stores a computer program for executing the above method.

[0080] As shown in Figure 7 , the electronic device 600 can further include a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It should be noted that the electronic device 600 does not necessarily include all the components shown in Figure 7 ; in addition, the electronic device 600 can further include components not shown in Figure 7 , which can refer to prior art.

[0081] As shown in Figure 7 , the central processor 100, also known as a controller or operation control, can include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of each component of the electronic device 600.

[0082] The memory 140, for example, can be one or more of a cache, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory or other suitable device. The above information related to failure can be stored, and programs for executing the information can also be stored. The central processor 100 can execute the programs stored in the memory 140 to realize information storage or processing, etc.

[0083] The input unit 120 provides input to the central processing unit 100. The input unit 120 is, for example, a key or touch input device. The power supply 170 is used to supply power to the electronic device 600. The display 160 is used to display display objects such as images and text. The display can be, for example, an LCD display, but is not limited thereto.

[0084] The memory 140 can be a solid state memory such as a read only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROM, etc. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 can include an application / function storage section 142 for storing application programs and function programs or a flow for executing the operation of the electronic device 600 by the central processing unit 100.

[0085] The memory 140 can also include a data storage section 143 for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. A driver storage section 144 of the memory 140 can include various drivers of the electronic device for a communication function and / or for performing other functions of the electronic device such as a messaging application, an address book application, etc.

[0086] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, as in the case of a conventional mobile communication terminal.

[0087] Based on different communication technologies, a plurality of communication modules 110 can be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling a conventional telecommunication function. The audio processor 130 can include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 130 is also coupled to the central processing unit 100, thereby enabling recording on the local device via the microphone 132 and playing stored sounds on the local device via the speaker 131.

[0088] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the methods can be tangibly embodied in a machine-readable storage medium having stored thereon instructions that can be used to program a computer to perform any of the operations described herein. The software implementation can be for example, in the form of a computer program product. The software implementation can be implemented in a centralized fashion in one computer system or network, or be distributed over a network such computer systems, some of which can implement the

[0089] The present application is described in relation to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It is understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing each of the functions specified in the flow diagram and / or block diagram block or blocks.

[0090] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing each of the functions specified in the flow diagram and / or block diagram block or blocks.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing each of the functions specified in the flow diagram and / or block diagram block or blocks.

[0092] The principles and implementation of the present application are described in the detailed description of specific embodiments. The above description of the embodiments is only for the purpose of understanding the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific embodiments and the scope of application will be changed; in view of the above, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method of analyzing a user experience of a bank teller product, characterized by, The method comprises: forming a service product vocabulary and a service product function vocabulary based on bank counter service product information and service product function information; performing word segmentation on the emotional tendency text of the bank counter service to be judged; wherein the emotional tendency text of the bank counter service to be judged is a short text of employee communication collected periodically from a bank internal grassroots service support platform; matching the word segmentation result with key words in the service product vocabulary and the service product function vocabulary to obtain matching key words; performing text entity recognition on the emotional tendency text of the bank counter service to be judged by using an entity recognition model, and outputting the recognition key words corresponding to the emotional tendency text of the bank counter service to be judged; performing fusion analysis on the matching key words and the recognition key words by using a voting method, and determining the bank counter service product or function to which the emotional tendency text of the bank counter service to be judged points based on the same key words when the matching key words and the recognition key words are the same; performing emotional tendency classification on the emotional tendency text of the bank counter service to be judged by using a text emotion classification model, and obtaining the emotional category to which the emotional tendency text of the bank counter service to be judged belongs; analyzing the user experience of the corresponding bank counter service product or function based on the emotional category analysis.

2. The bank teller product user experience analysis method according to claim 1, wherein, The entity recognition model is obtained by training in the following manner: obtaining text information related to bank counter service products or functions; performing entity label annotation on the text information to annotate the specific product or function pointed by the text information, and forming an entity recognition data set; training the entity recognition model by using the entity recognition data set to obtain the entity recognition model.

3. The bank teller product user experience analysis method according to claim 2, wherein, The entity recognition model is a Mengzi network model, and the training of the entity recognition model by using the entity recognition data set comprises: using a pre-trained Mengzi model as the initial network parameters, setting the learning rate and the number of iterations in the training process, and saving the model with the best classification result in the validation set in the training process as the entity recognition model.

4. The bank teller product user experience analysis method according to claim 1, wherein, The word segmentation on the emotional tendency text of the bank counter service to be judged comprises: using the service product vocabulary and the service product function vocabulary as the self-defined dictionary of the Jieba word segmentation tool, and then performing Jieba word segmentation processing on the emotional tendency text of the bank counter service to be judged.

5. The method of claim 1, wherein the user experience analysis of the bank teller product is performed by a server computer. The text emotion classification model is obtained by training in the following manner: obtaining related information of different products or functions; annotating the emotional tendency labels for the related information to obtain an emotional classification data set; training the text emotion classification model by using the emotional classification data set.

6. A bank counter service product user experience analysis device characterized by comprising: The device comprises: a vocabulary generation unit configured to form a service product vocabulary and a service product function vocabulary based on bank counter service product information and service product function information; a word segmentation unit configured to perform word segmentation on the emotional tendency text of the bank counter service to be judged; wherein the emotional tendency text of the bank counter service to be judged is a short text of employee communication collected periodically from a bank internal grassroots service support platform; The matching unit is configured to match the word segmentation result with the key words in the business product word library and the business product function word library to obtain matched key words; the entity recognition model is used to perform text entity recognition on the sentiment tendency text of the bank counter service to be judged, and output the recognized key words corresponding to the sentiment tendency text of the bank counter service to be judged; the voting method is used to perform fusion analysis on the matched key words and the recognized key words, and when the matched key words and the recognized key words are the same, the same key words are used to determine the bank counter service product or function to which the sentiment tendency text of the bank counter service to be judged is directed. The sentiment classification unit is configured to use a text sentiment classification model to perform sentiment tendency classification on the sentiment tendency text of the bank counter service to be judged, and obtain the sentiment category to which the sentiment tendency text of the bank counter service to be judged belongs. The experience analysis unit is configured to analyze the user experience of the corresponding bank counter service product or function based on the sentiment category.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 5.

Images