Data processing method and device, readable storage medium, and computer program product

Abstract

Provided are a data processing method and device, a readable storage medium, and a computer program product. The method comprises: acquiring first data, the first data comprising a text record and / or a call recording in a customer journey; on the basis of text data transcribed from the text record and / or the call recording and by means of a first model, analyzing a first index of the customer journey, the first index representing the probability of a drop-off point in the customer journey; when the first index is less than a first threshold, determining that no drop-off point exists in the customer journey.

Description

Data processing methods and equipment, readable storage media, computer program products

[0001] This application claims priority to Chinese Patent Application No. 202411656427.7, filed on November 18, 2024, with the China National Intellectual Property Administration, entitled “Data Processing Method and Apparatus, Readable Storage Medium, Computer Program Product”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of data processing, and more particularly to a data processing method and apparatus, a readable storage medium, and a computer program product. Background Technology

[0003] In the process of serving customers, a company's customer service center may experience customer journey interruptions due to various reasons, as the core needs of customers (or users) may not be met in a timely manner.

[0004] Operations managers of enterprise customer service centers urgently need to understand what customer service gaps exist during the process of enterprise customers requesting customer service, and take timely remedial measures to restore customer satisfaction and enhance the value of enterprise customer service centers. Summary of the Invention

[0005] This application provides a data processing method and apparatus, a readable storage medium, and a computer program product that can perform customer journey breakpoint analysis for each customer, generate remedial measures based on the analysis results, and promptly remedy customer service to restore customer satisfaction and experience.

[0006] Firstly, a data processing method is provided. This method can be executed by a data processing device (i.e., a computer device, such as a server) or by a module (e.g., a processor, chip, or chip system) applied to the data processing device. It can also be implemented by a logical node, logical module, or software capable of implementing all or part of the functions of the data processing device. In this data processing method, the data processing device first acquires first data, which may include text records and / or call recordings from a customer journey. Then, the data processing device analyzes a first index of the customer journey based on the text data transcribed from the text records and / or call recordings using a first model. The first index represents the probability of a breakpoint existing in the customer journey. If the first index is less than a first threshold, it can be determined that there is no breakpoint in the customer journey.

[0007] As can be seen from the above embodiments, the data processing device analyzes the customer journey using data stored in the customer service center. This solves the problem of the large workload caused by the need for invasive modifications to contact devices to collect relevant data for each customer journey in the prior art. At the same time, the data processing device analyzes each customer journey separately and uses artificial intelligence technology to automatically discover potential customer journey breakpoints, accurately match each customer, and promptly identify each customer's problems, facilitating targeted solutions to problems and improving customer satisfaction and customer experience. When the probability of a breakpoint in a customer journey is low, it can be determined that there is no breakpoint in the customer journey, and no further analysis or processing is required for that customer journey, thus improving the efficiency of the data processing device.

[0008] In one possible implementation, the first threshold is 0.5. If the first index is less than 0.5, that is, less than the first threshold, it can be determined that there is no breakpoint in the customer journey.

[0009] As can be seen from the above embodiments, the data processing device analyzes the customer journey through the data stored in the customer service center, which solves the problem of the large workload caused by the need for invasive modification of the contact point device to collect relevant data of each customer journey in the prior art. At the same time, the data processing device analyzes each customer journey separately and accurately calculates the probability of a breakpoint in each customer journey. When the probability is less than 0.5, that is, when the probability of a breakpoint in the customer journey is very small, it can be determined that there is no breakpoint in the customer journey, that is, no further operation is required for the customer journey, thus reducing the workload of the data processing device.

[0010] In one possible implementation, if the first index is not less than a first threshold, the data processing device can analyze the customer's sentiment index in the customer journey based on text data transcribed from text records and / or call records using a second model. The sentiment index represents the degree of positive sentiment of the customer in the customer journey. If the sentiment index is less than a second threshold, the data processing device can determine that there is a breakpoint in the customer journey.

[0011] As can be seen, in the above embodiments, the data processing device performs sentiment analysis on the text data transcribed from text records and / or call records, and uses the analysis of the text data transcribed from text records and / or call records to corroborate the results of whether there are customer journey breakpoints, thereby increasing the accuracy of the results; when there is a high probability that there are breakpoints in the customer journey and the customer's sentiment index is low, it is determined that there are breakpoints in the customer journey, which facilitates further analysis of customer journey breakpoints.

[0012] In one possible implementation, the second threshold is equal to 0; the data processing device can analyze the customer's sentiment index in the customer journey based on the text data transcribed from text records and / or call records using the second model. If the second model identifies the customer as having a negative sentiment, the sentiment index is -1; if the second model identifies the customer as having a neutral sentiment, the sentiment index is 0; if the second model identifies the customer as having a positive sentiment, the sentiment index is 1.

[0013] As can be seen, in the above embodiments, the data processing device performs sentiment analysis on the text data transcribed from text records and / or call records, classifying sentiment into negative, neutral, and positive sentiments. These three sentiments correspond to sentiment indices of -1, 0, and 1, respectively. The sentiment index in the customer journey is compared with a second threshold of 0 to corroborate the results obtained from analyzing the text data transcribed from text records and / or call records to determine whether there is a customer journey breakpoint. If there is a high probability that there is a customer journey breakpoint and the customer's sentiment index is -1, that is, if there is a high probability that there is a breakpoint in the customer journey and the customer has a negative sentiment in the customer journey, it is determined that there is a breakpoint in the customer journey, increasing the accuracy of the results and facilitating further analysis of customer journey breakpoints.

[0014] In one possible implementation, the first index is not less than a first threshold, and the first data also includes customer satisfaction; the data processing device obtains a satisfaction score based on the customer's satisfaction rating, the satisfaction score representing the customer's level of satisfaction with the service provided; if the satisfaction score is less than a third threshold, it is determined that there is a breakpoint in the customer journey.

[0015] As can be seen, in the above embodiments, the data processing device obtains a satisfaction score through satisfaction evaluation, and analyzes the text data transcribed from text records and / or call records to corroborate the results of whether there are customer journey breakpoints, thereby increasing the accuracy of the results; when there is a high probability that there are breakpoints in the customer journey and the satisfaction score is low, it is determined that there are breakpoints in the customer journey, which facilitates further analysis of customer journey breakpoints.

[0016] In one possible implementation, the third threshold is equal to 3; the data processing device obtains a satisfaction score based on the customer satisfaction evaluation, and the satisfaction score is 5 when the data processing device identifies that the customer satisfaction evaluation indicates very high satisfaction; 4 when the data processing device identifies that the customer satisfaction evaluation indicates satisfaction; 3 when the data processing device identifies that the customer satisfaction evaluation indicates neutral satisfaction; 2 when the data processing device identifies that the customer satisfaction evaluation indicates dissatisfaction; and 1 when the data processing device identifies that the customer satisfaction evaluation indicates very low satisfaction.

[0017] As can be seen from the above embodiments, the data processing device obtains a satisfaction score through a satisfaction evaluation, and divides the satisfaction evaluation into five levels: very satisfied, satisfied, neutral, dissatisfied, and very dissatisfied. These five levels of satisfaction evaluation correspond to satisfaction scores of 5, 4, 3, 2, and 1, respectively. The satisfaction score is compared with a third threshold 3, and the results of analyzing the text data obtained through text records and / or call records are used to verify whether there is a customer journey breakpoint. If there is a high probability that there is a customer journey breakpoint and the satisfaction score is 1 or 2, that is, if there is a high probability that there is a breakpoint in the customer journey and the customer expresses dissatisfaction or very dissatisfaction during the customer journey, it is determined that there is a breakpoint in the customer journey, which increases the accuracy of the results and facilitates further analysis of customer journey breakpoints.

[0018] In one possible implementation, the first index is not less than the first threshold, and the first data also includes the problem resolution rate; if the problem resolution rate indicates that the problem in the customer journey has not been resolved, it can be confirmed that there is a breakpoint in the customer journey.

[0019] As can be seen, in the above embodiments, the data processing device obtains the problem resolution rate and analyzes the text data transcribed from text records and / or call records to corroborate the results of whether there are customer journey breakpoints, thereby increasing the accuracy of the results; when there is a high probability that there are breakpoints in the customer journey and the problem resolution rate indicates that the problem has not been resolved, it is determined that there are breakpoints in the customer journey, which facilitates further analysis of customer journey breakpoints.

[0020] In one possible embodiment, the first index is not less than a first threshold, and the first data also includes a customer satisfaction score and a problem resolution rate. The data processing device analyzes the customer's emotional index in the customer journey based on text data transcribed from text records and / or call records using a second model. The emotional index represents the customer's positive emotional level in the customer journey. The data processing device obtains the satisfaction score based on the customer satisfaction evaluation. The satisfaction score represents the customer's satisfaction with the customer service. If the emotional index is greater than or equal to a second threshold, the satisfaction score is greater than or equal to a third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, a second index is calculated based on the emotional index, the satisfaction score, and the problem resolution rate. The second index represents the probability that there is a breakpoint in the customer journey.

[0021] As can be seen, in the above embodiments, when the first index is not less than the first threshold, the sentiment index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, the probability of a breakpoint in the customer journey is calculated by comprehensively considering the satisfaction score, the problem resolution rate, and the sentiment index. This allows for further analysis of whether a breakpoint exists in the customer journey, thereby increasing the accuracy of the analysis results.

[0022] In one possible embodiment, the second threshold is equal to 0, and the third threshold is equal to 3. The data processing device can analyze the customer's sentiment index during the customer journey using a second model based on text data transcribed from text records and / or call logs. If the second model identifies the customer as having negative emotions, the sentiment index is -1; if the second model identifies the customer as having neutral emotions, the sentiment index is 0; and if the second model identifies the customer as having positive emotions, the sentiment index is 1. The data processing device obtains a satisfaction score based on customer satisfaction ratings. If the data processing device identifies the customer satisfaction rating as very satisfied, the satisfaction score is 5; if the data processing device identifies the customer satisfaction rating as satisfied, the satisfaction score is 4; if the data processing device identifies the customer satisfaction rating as average, the satisfaction score is 3; if the data processing device identifies the customer satisfaction rating as dissatisfied, the satisfaction score is 2; and if the data processing device identifies the customer satisfaction rating as very dissatisfied, the satisfaction score is 1. When the sentiment index is greater than or equal to 0, the satisfaction score is greater than or equal to 3, and the problem resolution rate indicates that the customer's problem has been resolved, the data processing device calculates a second index based on the sentiment index, satisfaction score, and problem resolution rate. The second index represents the probability that there is a breakpoint in the customer's journey.

[0023] As can be seen, in the above embodiments, when the first index is not less than the first threshold, the emotional index is greater than or equal to the second threshold 0, the satisfaction score is greater than or equal to the third threshold 3, and the problem resolution rate indicates that the customer's problem has been resolved, the probability of a breakpoint in the customer journey is calculated by comprehensively considering the satisfaction score, the problem resolution rate, and the emotional index. This allows for further analysis of whether a breakpoint exists in the customer journey, thereby increasing the accuracy of the analysis results.

[0024] In one possible implementation, if the emotional index is greater than or equal to a second threshold, the satisfaction score is greater than or equal to a third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, the formula is used. Calculate the second index. Where I is the second index, Q is the problem resolution rate (Q equals 1 if the customer's problem has been resolved), E is the sentiment index, a is the parameter weight of the sentiment index, used to control the influence of the sentiment index on the second index, C is the satisfaction score, and b is the parameter weight of the satisfaction score, used to control the influence of the satisfaction score on the second index.

[0025] As can be seen, in the above embodiments, when the first index is not less than the first threshold, the sentiment index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, the satisfaction score, problem resolution rate, and sentiment index are substituted into the formula to calculate the probability that there is a breakpoint in the customer journey. Further analysis is then conducted on whether there is a breakpoint in the customer journey, thereby increasing the accuracy of the result.

[0026] In one possible embodiment, the data processing device may also output a second index, and the data processing device receives breakpoint indication information indicating whether a breakpoint exists in the customer journey.

[0027] As can be seen, in the above embodiments, after the data processing device calculates the second index, it can wait for the breakpoint indication information to determine whether there is a breakpoint in the customer journey. That is, other devices can be introduced to determine whether there is a breakpoint in the customer journey or a manual judgment can be made, which increases the accuracy of the result of whether the customer journey is a breakpoint and facilitates further analysis of the customer journey breakpoint.

[0028] In one possible embodiment, the data processing device may further perform intent recognition based on the first data using a third model, the intent including the customer's business needs; based on the recognized intent, the data processing device generates remedial measures in conjunction with a remedial measures library, the remedial measures library including methods for implementing the intent.

[0029] As can be seen from the above embodiments, after the data processing device identifies the breakpoints in the customer journey, it identifies the reasons for the breakpoints, such as unmet business needs of customers, and generates remedial measures by combining the large model and the remedial measures library of operations and management personnel to improve customer satisfaction and experience.

[0030] In one possible embodiment, the data processing device may also proactively send remedial measures, or send remedial measures after receiving a query signal, the query signal including querying whether there is a breakpoint in the customer journey and the corresponding remedial measures.

[0031] As can be seen from the above embodiments, the data processing device actively or passively sends remedial measures to promptly remedy and repair customer journeys with interruptions through the self-service or manual service of the customer service center, thereby improving customer experience and customer satisfaction.

[0032] In a second aspect, a data processing apparatus includes at least one processor; wherein the at least one processor is configured to perform the methods in any possible implementation of the first aspect. The at least one processor can execute computer programs or instructions stored in memory to cause the described methods to be performed. The memory may be included in the data processing apparatus or located externally to the data processing apparatus. Furthermore, the data processing apparatus may also include an interface.

[0033] Thirdly, a communication system is provided, comprising a data processing device. The data processing device is used to perform the method as described in any one of the first aspects.

[0034] Fourthly, a computer-readable storage medium is provided that stores computer instructions, which, when executed, cause a computer to perform a method as described in any possible implementation of the first aspect.

[0035] Fifthly, a computer program product is provided, comprising: computer program code, which, when executed by a computer, causes the computer to perform a method as described in any possible implementation of the first aspect.

[0036] In a sixth aspect, a chip is provided, the chip including at least one processor and an interface, the processor being configured to read and execute instructions stored in a memory, which, when executed, cause the chip to perform a method as described in any possible implementation of the first aspect.

[0037] For the technical effects of the various possible implementations of the second to sixth aspects, please refer to the introduction of the technical effects of the first aspect or the various possible implementations of the first aspect. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0039] Figure 1 is a schematic diagram of a customer service process provided in an embodiment of this application;

[0040] Figure 2 is a schematic diagram of a system architecture or scenario provided in an embodiment of this application;

[0041] Figure 3 is a schematic diagram of the workflow of a customer journey breakpoint analysis tool provided in an embodiment of this application;

[0042] Figure 4 is a schematic diagram of a data processing method provided in an embodiment of this application;

[0043] Figure 5 is a schematic diagram of the structure of a data processing device provided in an embodiment of this application. Detailed Implementation

[0044] The technical solutions in the embodiments of this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B; the word "and / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, and A and B existing simultaneously. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more, and "multiple types" refers to two or more.

[0045] It should be understood that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0046] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application can be combined with other embodiments.

[0047] In serving customers, a company's customer service center (call center) can divide customer service into self-service and human service. In this embodiment, the customer journey refers to the entire process from a customer's initial awareness of a product or service to their final purchase or use of it, including subsequent interactions and feedback. This represents the interaction path between the customer and the brand or company at each touchpoint. As shown in Figure 1, various reasons may cause customer core needs to be unmet in a timely manner, resulting in customer service breakpoints (i.e., customer journey breakpoints), meaning customer service is interrupted.

[0048] For example, during self-service, customers' service requests may not be met due to malfunctions in the customer service center support system or a lack of relevant service content; similarly, during human assistance, customers' service requests may not be met due to the lack of knowledge of customer service personnel.

[0049] Therefore, the operations managers of enterprise customer service centers urgently need to understand what service gaps exist in the customer's request process and take timely remedial measures to restore customer satisfaction and enhance the value of the enterprise customer service center.

[0050] In the process of serving customers in a company's customer service center, the analysis of customer journey breakpoints typically falls into the following two categories:

[0051] 1. Touch Point Data Collection and Analysis: Enterprises modify touch point applications, adding preset data points to record customer data at each touch point. The collected data is then statistically analyzed to identify customer breakpoints. Touch point applications are those that allow customers to connect with the enterprise's customer service center, such as the customer service hotline or the customer service channel on the enterprise's official WeChat account.

[0052] 2. Report Statistics: When it's impossible to modify the contact point application, customer satisfaction and the first-response rate of customer needs are statistically analyzed in the contact statistics reports to indirectly assess customer dissatisfaction with the service. The first-response rate refers to the percentage of customers whose problems are resolved on their first contact with the customer service center when using human assistance.

[0053] The above analysis of customer journey breakpoints has two main drawbacks:

[0054] 1. The contact touch data collection solution based on the modification of contact touch applications requires intrusive development of contact touch applications to collect business acceptance data from contact touch points. In addition, enterprise call centers usually have multiple contact touch applications at the same time. This solution requires modification of all contact touch applications, which involves a large amount of modification work. Furthermore, the different channels through which enterprise customers and customer service centers establish connections may be from different vendors. The solution for modifying contact touch applications requires collaboration with multiple vendors, making it difficult to implement.

[0055] 2. Report analysis solutions based on contact statistics, which statistically analyze customer satisfaction and the first-response rate of customer needs, can only indirectly identify the number of dissatisfied customers during the service process. They cannot accurately match each customer, meaning they lack in-depth analysis of the raw data from contact point applications and cannot conduct targeted analysis and solutions for the problems of each user.

[0056] To address the aforementioned issues, this application provides a data processing method. In this method, a data processing device analyzes customer journey-related data from a customer service center, identifies whether there are breakpoints in the customer journey, generates corresponding remedial measures for customer journeys with breakpoints, and then remediates the customer journey according to these measures. Correspondingly, this application provides a data processing device that may consist of a newly added "customer journey breakpoint analysis tool" and enhanced customer service center components. The enhanced customer service center components may include "self-service" and "human service" components.

[0057] This application provides a customer journey breakpoint analysis tool that supports the analysis and identification of service breakpoints in the customer service process without altering the customer service center's contact channels (i.e., without modifying contact point applications or adding preset tracking points). It utilizes contact records, text records, and call records from the customer service center's voice, video, and social media contact channels, leveraging contact statistics reports and combining voice and text analysis tools. This tool identifies interaction breakpoints (i.e., customer journey breakpoints) and, by combining a large model with the remedial experience library (remedial measures library) of operations management personnel, automatically provides self-service or human remedial measures. This customer journey breakpoint analysis tool eliminates the intrusive modifications to the system required for collecting data through contact point tracking. Instead, it utilizes customer journey-related data from the enterprise's customer service center, combined with artificial intelligence technology, to automatically uncover potential customer journey breakpoints and provide remedial solutions, thereby improving customer satisfaction and customer experience. Text records include conversations between customers and the customer service center through various channels. Customer satisfaction can refer to subjective ratings from customer satisfaction surveys of the customer service center's services, a key metric reflecting whether customer requests have been handled satisfactorily.

[0058] The enhanced customer service center component provided in this application can restore the service context in a timely manner when a customer comes in, based on the results of historical customer journey breakpoint analysis and remedial measures, that is, to understand the historical progress and subsequent steps of the customer's requested service.

[0059] Figure 2 illustrates the system architecture or scenario to which this application can be applied.

[0060] The following description uses the system architecture 200 shown in Figure 2 as an example. It should be understood that system architecture 200 may have more or fewer components than shown in the figure. The various components shown in the figure can be implemented in combinations of one or more devices. System architecture 200 includes contact channels, a customer service center, and a customer journey breakpoint analysis tool. The contact channels can be contact touchpoint applications, and the customer journey breakpoint analysis tool is used to analyze whether customer journey breakpoints exist and the corresponding remedial measures in the event of such breakpoints.

[0061] Customer service centers include self-service (such as interactive voice response (IVR) or voice and text chatbots) and human service. Customer service centers are customer-facing and provide customer service functions through multiple channels such as voice, video, and text.

[0062] It should be noted that the customer journey breakpoint analysis tool can be part of the customer service center and located on the same data processing device, or it can be separate from the customer service center and located on a separate device. The two devices together constitute the data processing device in this embodiment, and this application does not limit it in this way.

[0063] In the system architecture, the customer journey breakpoint analysis tool is responsible for identifying customer journey breakpoints based on relevant content from historical and real-time contact records, call recordings, and text logs of customer service self-service and human services, combined with artificial intelligence analysis. It then generates remedial measures using a remedial measures library, enabling the customer service center to restore customer service and improve customer experience and satisfaction. The remedial measures library includes historical customer journey breakpoints and corresponding remedial measures.

[0064] The workflow of the customer journey breakpoint analysis tool provided in this application can be explained with reference to the flowchart shown in Figure 3: Customer journey data includes contact records, call recordings, and text records. Contact records undergo contact statistical analysis, and the analysis results are input into a statistical indicator library for further analysis. Based on the indicators in the statistical indicator library, relevant data corresponding to the customer journey are obtained, i.e., the contact record analysis results are obtained. For example, if the statistical indicator library contains two indicators, satisfaction and problem resolution rate, the satisfaction and problem resolution rate of the customer journey can be analyzed based on the data in the contact records, forming structured data. The customer journey breakpoint analysis tool performs text analysis on the text data transcribed from text records and call recordings to initially determine whether there are customer journey breakpoints. Simultaneously, it can also perform sentiment analysis and intent analysis on the text data transcribed from text records and call recordings. Based on the results of sentiment analysis and contact record analysis, the initial judgment on whether there are customer journey breakpoints is corroborated, further confirming the existence of customer journey breakpoints. In the case of customer journey breakpoints, intent analysis is used to identify the customer's intent, for example, identifying the customer's needs in this customer journey. Intent analysis and sentiment analysis are optional steps, and users can choose whether to perform them based on actual circumstances. This application does not impose any restrictions on them. The customer journey breakpoint analysis tool further analyzes customer journeys with breakpoints, that is, it generates remedial measures based on the remedial measure library and the breakpoint analysis results (i.e., the results of intent analysis), and applies the remedial measures to the customer service remediation process.

[0065] To identify customer journey breakpoints and provide timely remedial solutions to improve customer satisfaction and experience, this application provides a data processing method. This method uses a customer journey breakpoint analysis tool to analyze contact records, text records, call recordings, etc., to confirm the existence of customer journey breakpoints. For customer journeys with breakpoints, further analysis is conducted, namely, remedial measures are generated based on the customer journey breakpoints and the remedial database. These measures are then implemented through the customer service center in real time, proactively, or passively, thereby improving customer experience and satisfaction.

[0066] In one embodiment, the data processing device acquires first data, which includes text records and / or call recordings from a customer journey. The data processing device analyzes a first index of the customer journey based on the text data transcribed from the text records and / or call recordings using a first model. The first index represents the probability of a breakpoint existing in the customer journey. If the first index is less than a first threshold, it is determined that there is no breakpoint in the customer journey. If the first index is not less than the first threshold, the existence of a breakpoint in the customer journey can be further determined.

[0067] The above method embodiments include many possible implementation schemes. Some of these implementation schemes are illustrated below with reference to Figure 4.

[0068] In this application, the embodiment shown in FIG4 can be used as a separate embodiment, and some steps in the embodiment shown in FIG4 can also be used as separate embodiments.

[0069] Next, please refer to Figure 4, which illustrates a data processing method provided by an embodiment of this application. In this embodiment, the data processing device is divided into a customer journey breakpoint analysis tool and a customer service center. The method includes, but is not limited to, the following steps:

[0070] S401: The customer requests enterprise customer service from the customer service center.

[0071] Customers can request customer service from the customer service center through multiple channels, such as the company's customer service hotline and the customer service section of the company's official WeChat account. During the process of a customer requesting customer service, there may be service interruptions (i.e., customer journey breakpoints). For example, malfunctions in the customer service center's support system or the lack of relevant service content may prevent the customer's service request (i.e., the customer's business needs) from being met, or the lack of relevant knowledge among human service personnel may also lead to unmet customer service requests. These situations can all result in service interruptions.

[0072] Service interruptions caused by various reasons can reduce customer satisfaction and customer experience. Therefore, in this embodiment, a customer journey interruption analysis tool is used to identify customer journey interruptions based on relevant content in the customer service center's historical and / or real-time contact records, call records, and text records, combined with artificial intelligence analysis. Remedial measures are then generated using a remedial measures library. The core processing procedure for customer journey interruption analysis is shown in steps S402-S408. It should be noted that the order of steps S405-S408 is not limited except that step S406 must precede step S407. That is, it can also be performed in the order of S408, S406, S407, and S405. During the execution of steps S405-S408, steps S406 and S407 can also be performed sequentially as a single step, synchronously with S405 and S408. Furthermore, steps S405 and S408 are optional steps, meaning that whether or not to perform these steps can be decided based on the actual situation. This embodiment does not impose any limitations on this.

[0073] S402: Customer journey breakpoint analysis tool obtains real-time and / or historical contact records, call recordings, and text records from customer service centers.

[0074] Customer journey breakpoint analysis tools acquire primary data, which may include text records and / or call recordings from the customer journey, and may also include customer satisfaction ratings and / or problem resolution rates.

[0075] For example, text records may include transcripts of conversations between customers and the customer service center through various channels, and contact records may include structured data such as call time, call duration, customer satisfaction rating, and customer problem resolution rate.

[0076] It should be noted that the acquisition of contact records, call recordings, and text records is only one embodiment. In other embodiments, it may include only one of call recordings and text records, or only one of call recordings and text records, and contact records. For example, when the first data only includes call recordings, steps S406 and S407 do not need to be executed; as another example, when the first data only includes text records, steps S403, S406, and S407 do not need to be executed; as another example, when the first data includes call recordings and text records, steps S406 and S407 do not need to be executed; as another example, when the first data includes text records and contact records, steps S406 and S407 do not need to be executed; as yet another example, when the first data includes call recordings and contact records, each step described in this embodiment needs to be executed completely. In summary, the specific steps to be executed can be adjusted based on the first data obtained by the customer journey breakpoint analysis tool.

[0077] S403: The Customer Journey Breakpoint Analysis Tool converts call recordings into text.

[0078] The customer journey breakpoint analysis tool acquires customer journey-related data stored in the customer service center, i.e., the first data. The first data may include contact records, call recordings, and text records. Specifically, if the first data includes audio files such as call recordings, step S403 can be executed to transcribe the audio files into text data to facilitate subsequent analysis by the text analysis tool. If the first data acquired by the customer journey breakpoint analysis tool does not include audio files or audio data, step S403 can be skipped, and step S404 can be executed directly.

[0079] It should be noted that the method of transcribing audio files into text is not limited in the embodiments of this application.

[0080] S404: The Customer Journey Breakpoint Analysis Tool identifies customer journey breakpoints through text analysis.

[0081] The customer journey breakpoint analysis tool analyzes a first index of the customer journey based on text data transcribed from text records and / or call records using a first model. The first index represents the probability of a breakpoint in the customer journey. If the first index is less than a first threshold, it is determined that there is no breakpoint in the customer journey. If the first index is not less than the first threshold, subsequent steps are used to further determine whether there is a breakpoint in the customer journey.

[0082] The first model can use text analysis technology to analyze the text and / or text records transcribed from call recordings, identify customer journey breakpoints in the service process, and calculate the probability of breakpoints in the customer journey.

[0083] For example, a customer journey breakpoint analysis tool can use parameter P to represent the probability of a customer journey breakpoint during a customer service process. Parameter P can be obtained by analyzing the text and / or text records transcribed from the call recordings during this customer journey. The value range of parameter P is [0, 1] and parameter P is a real number.

[0084] In some embodiments, the first threshold is equal to 0.5. If the value of parameter P corresponding to the customer journey is greater than or equal to 0.5, the customer journey breakpoint analysis tool can determine that there may be a customer journey breakpoint in the customer journey; if the value of parameter P corresponding to the customer journey is less than 0.5, the customer journey breakpoint analysis tool can determine that there is no customer journey breakpoint in the customer journey.

[0085] In some embodiments, the process of analyzing customer journey breakpoints using text and / or text records transcribed from call recordings during the customer journey can be performed by the intelligent processing module in the customer journey breakpoint analysis tool. The intelligent processing module contains a breakpoint identification model (i.e., the first model), which can be trained using historical text data transcribed from call recordings and / or historical text records, along with the probability of breakpoints in the corresponding customer journey. For example, the breakpoint identification model can be trained using data consisting of 5000 sets of historical text data transcribed from call recordings and the probabilities of corresponding customer journey breakpoints.

[0086] S405: The Customer Journey Breakpoint Analysis Tool uses sentiment analysis to help corroborate customer journey breakpoints.

[0087] The customer journey breakpoint analysis tool uses a second model to analyze the customer's sentiment index based on text data transcribed from text records and / or call logs. The sentiment index represents the degree of positive emotion the customer experiences during the customer journey. If the sentiment index is less than a second threshold, provided the first index is not less than a first threshold, a breakpoint in the customer journey can be identified. If the sentiment index is greater than or equal to the second threshold, further judgment is required in conjunction with other steps.

[0088] For example, a customer journey breakpoint analysis tool can use parameter E to represent the customer’s sentiment index during a customer service process. The sentiment index E can be obtained by analyzing the text data and / or text records transcribed from the call recordings during this customer journey.

[0089] In some embodiments, the customer sentiment index E, obtained through analysis of text data and / or text records transcribed from call recordings during customer service, can be processed by the intelligent processing module in the customer journey breakpoint analysis tool. The intelligent processing module contains a sentiment recognition model (i.e., a second model), which can be trained using text data transcribed from historical call recordings and / or historical text records, along with the corresponding customer sentiment index E. For example, the sentiment recognition model can be trained using data consisting of 5000 sets of text data transcribed from historical call recordings and the corresponding customer sentiment index E.

[0090] In some embodiments, the first threshold is equal to 0.5. The customer journey breakpoint analysis tool quantifies the customer's sentiment index E based on the text and / or text records transcribed from the call recordings during customer service. It can determine that the sentiment index E is -1 if the customer expresses negative emotions; 0 if the customer expresses neutral emotions; and 1 if the customer expresses positive emotions, in which case the second threshold is equal to 0. If the value of the parameter P corresponding to customer service is greater than or equal to 0.5 (i.e., the first index is not less than the first threshold) and the corresponding customer's sentiment index E is -1 (i.e., the customer's sentiment index is less than the second threshold), the customer journey breakpoint analysis tool determines that a customer journey breakpoint exists. If the value of the parameter P corresponding to customer service is greater than or equal to 0.5 (i.e., the first index is not less than the first threshold) and the corresponding customer's sentiment index E is 0 or 1 (i.e., the customer's sentiment index is not less than the second index), the customer journey breakpoint analysis tool needs to combine other steps for further judgment to determine whether a breakpoint exists in the customer journey.

[0091] S406: Customer journey breakpoint analysis tool statistically analyzes historical contact record metrics.

[0092] Contact record metrics may include call time, call duration, customer satisfaction rating, and customer problem resolution rate. For example, customer satisfaction ratings can be obtained by asking the customer at the end of a customer service session. For instance, after completing a customer service session, a satisfaction rating is sent to the customer, who can choose between 0-5 stars and / or between "problem resolved" and "problem unresolved." The customer's choice is then quantified, i.e., converted into a numerical form. For example, in the problem resolution rate, 0 indicates the problem is unresolved, and 1 indicates the problem is resolved.

[0093] It should be noted that the step of compiling historical contact record metrics can be achieved by a customer journey breakpoint analysis tool using metric statistical techniques to compile historical customer service contact record metrics and obtain structured data, or it can be obtained directly from the customer service center. This application does not limit the scope of this step.

[0094] S407: Use historical contact record metrics to corroborate customer journey breakpoints.

[0095] Contact record metrics may include satisfaction ratings and / or problem resolution rates.

[0096] The customer journey breakpoint analysis tool obtains a satisfaction score, which represents the customer's level of satisfaction with the service provided. If the satisfaction score is less than a third threshold, provided the first index is not less than a first threshold, a breakpoint in the customer journey can be identified. If the satisfaction score is not less than the third threshold, further judgment is needed by combining other steps and / or indicators.

[0097] The customer journey breakpoint analysis tool obtains the problem resolution rate. If the problem resolution rate indicates that the customer's problem has not been resolved, it can also be determined that there is a breakpoint in the customer journey, provided that the first index is not less than the first threshold. If the problem resolution rate indicates that the customer's problem has been resolved, further judgment needs to be made in combination with other steps and / or indicators.

[0098] In some embodiments, the contact record indicator includes a customer satisfaction rating C. The satisfaction rating C can range from 1 to 5, where 1 represents very dissatisfied, 2 represents dissatisfied, 3 represents neutral, 4 represents satisfied, and 5 represents very satisfied. When the value of parameter P corresponding to the customer journey is greater than or equal to a first threshold, indicating customer dissatisfaction with the customer journey (i.e., a satisfaction rating C of 1 or 2), the customer journey breakpoint analysis tool can determine the existence of a customer journey breakpoint. When the customer expresses neutrality or satisfaction with the customer journey (i.e., a satisfaction rating C of 3, 4, or 5), the customer journey breakpoint analysis tool can combine other indicators and / or steps to calculate the probability of a breakpoint occurring in the customer journey.

[0099] In some embodiments, contact record metrics include the customer problem resolution rate Q. The problem resolution rate Q indicates whether the customer's problem has been effectively resolved. A problem resolution rate Q of 0 indicates that the problem is unresolved, and a problem resolution rate Q of 1 indicates that the problem has been resolved. The problem resolution rate Q can be collected in the satisfaction survey process at the end of a call or session, or it can be obtained through post-event analysis of repeated requests from the same customer. For example, if a customer repeatedly contacts the customer service center within 2 hours regarding the same or similar problem, it indicates that the problem is unresolved.

[0100] It should be noted that steps S406 and S407 can be analyzed by a customer journey breakpoint analysis tool, or the analysis can be performed by the customer service center, and the results (such as satisfaction scores and / or problem resolution rates) can be sent to the customer journey breakpoint analysis tool to comprehensively determine whether a breakpoint exists in the customer journey, as shown in Figure 3. This application embodiment does not limit this. In addition, steps S406 and S407 are optional steps and can be executed as needed. The customer journey breakpoint analysis tool can decide whether to execute this step based on whether relevant data exists in the contact records, or the enterprise can decide whether to execute this step. This application does not limit this.

[0101] In some embodiments, the customer journey breakpoint analysis tool selects to execute steps S405 and S406. The first data includes text records and / or call recordings from the customer journey, customer satisfaction ratings, and problem resolution rates. The first index is not less than a first threshold. The customer journey breakpoint analysis tool analyzes the customer's sentiment index in the customer journey based on text data transcribed from text records and / or call recordings using a second model, and obtains a satisfaction score based on the customer satisfaction rating. If the sentiment index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, the tool calculates a second index based on the sentiment index, satisfaction score, and problem resolution rate. The second index represents the probability that there is a breakpoint in the customer journey.

[0102] In some embodiments, the customer journey breakpoint analysis tool obtains the parameters P, sentiment index E, customer satisfaction score C, and customer problem resolution rate Q corresponding to customer service. When the parameter P corresponding to customer service is greater than or equal to the first threshold, the sentiment index E is greater than or equal to the second threshold, the customer satisfaction score C is greater than or equal to the third threshold, and the customer problem resolution rate Q is equal to 1 (i.e., the customer's problem has been resolved), the customer journey breakpoint index I, i.e., the second index, is calculated based on formula (1).

[0103] The Customer Journey Breakpoint Index I represents the probability of a service breakpoint. The value of the Customer Journey Breakpoint Index I is (0, 1]. a and b are parameter weights. a controls the influence of the Emotion Index E on the Customer Journey Breakpoint Index I, and b controls the influence of the customer satisfaction score C on the Customer Journey Breakpoint Index I.

[0104] For example, the customer's emotional index E is divided into -1, 0, and 1 according to the customer's emotions. E equals -1, which indicates negative emotions; E equals 0, which indicates neutral emotions; and E equals 1, which indicates positive emotions. The customer's satisfaction score C is divided into 1-5 points according to the customer's satisfaction level, which respectively represent very dissatisfied, dissatisfied, average, satisfied, and very satisfied. The customer's problem resolution rate is equal to 1 when the problem has been resolved and equal to 0 when the problem has not been resolved. The first threshold is equal to 0.5, the second threshold is equal to 0, and the third threshold is equal to 3. When the parameter P corresponding to customer service is greater than or equal to 0.5 (i.e., there may be a customer journey breakpoint in the customer service process in step S404), the emotional index E is greater than or equal to 0 (i.e., the customer expresses neutral or positive emotions in step S405), the customer's satisfaction score C is equal to 3, 4, or 5 (i.e., the customer is average, satisfied, or very satisfied with the customer service), and the customer's problem resolution rate Q is equal to 1 (i.e., the problem has been resolved), the customer journey breakpoint index I, i.e., the second index, can be calculated based on formula (1).

[0105] In other embodiments, after calculating the second index, the customer journey breakpoint analysis tool outputs the second index and waits to receive breakpoint indication information, which indicates whether a breakpoint exists in the customer journey. For example, if the results of text analysis (i.e., the results of the customer journey breakpoint analysis tool's analysis of whether a customer journey breakpoint exists) are inconsistent with the customer's actual feedback on satisfaction (such as satisfaction ratings in contact records) and the results of whether the problem was resolved (such as the customer's problem resolution rate in contact records), the existence of the customer journey breakpoint can be determined manually, i.e., by receiving manually input breakpoint indication information.

[0106] S408: Generate remedial measures based on intent analysis and a remedial measure library.

[0107] The customer journey breakpoint analysis tool identifies intents based on first data using a third model. These intents can be customer business needs. Based on the customer's intent, the tool combines a remedial measures library to generate remedial measures for timely restoration of customer service, improving customer service experience and customer satisfaction. The remedial measures library includes methods for achieving customer intents.

[0108] Based on the remedial measures generated in step S408, the customer journey breakpoint analysis tool can proactively send remedial measures to the customer service center, or it can send remedial measures after receiving a query signal from the customer service center. The query signal includes querying whether there are breakpoints in the customer journey and the corresponding remedial measures when breakpoints exist.

[0109] After receiving remedial measures, the customer service center can provide real-time (if the current contact is not yet over) or post-contact (if the current contact has ended) remediation for customer service. Post-contact remediation is divided into proactive (the customer service center actively contacts the customer) remediation and reactive (the customer service center responds to the customer during the next contact) remediation, reducing the customer's repetitive input. For example, when a customer accesses self-service, based on the customer's journey breakpoint and remedial measures, the customer is prompted whether to continue the previous customer service, thereby improving customer experience and satisfaction.

[0110] Remedial methods can be divided into real-time remediation, passive remediation, and active remediation. The detailed steps are as follows.

[0111] Steps S409 and S410 are real-time remedies.

[0112] S409: Push remedial measures.

[0113] The customer journey breakpoint analysis tool pushes remedial measures to the customer service center for repairing the corresponding customer journey.

[0114] S410: Perform remediation in the current customer service.

[0115] During the service process, the customer service center can restore the ongoing customer request (i.e., customer service) based on the remedial measures pushed by the customer journey breakpoint analysis tool.

[0116] For example, when a customer is transferred from self-service to human assistance, the customer service center automatically prompts the customer journey breakpoints during the self-service phase and the corresponding remedial measures.

[0117] Steps S411 to S414 are passive remedies.

[0118] S411: The customer requests enterprise customer service again.

[0119] If a customer's needs are not met during the first customer service session, the customer will request customer service again from the company's customer service center.

[0120] S412: Check if there are any customer journey breakpoints and remedial measures.

[0121] The customer service center sends a query to the customer service breakpoint analysis tool. This query includes whether there was a breakpoint in the customer journey during the previous customer service interaction, and if so, the corresponding remedial measures. Based on the query, the customer service breakpoint analysis tool sends a query indicating whether there was a breakpoint in the current customer journey, and if so, the corresponding remedial measures.

[0122] S413: Implement service recovery and context restoration based on the remediation measures for customer journey interruptions.

[0123] The customer service center uses the output of the customer journey breakpoint analysis tool to repair customer requests.

[0124] For example, when a customer accesses self-service, based on the breakpoints in the customer's journey and the corresponding remedial measures when breakpoints exist, the customer is prompted whether to continue the previous customer service.

[0125] S414: Remedial action is crucial for improving customer satisfaction.

[0126] The customer service center provides timely remedial services to improve customer satisfaction and experience.

[0127] Steps S415 to S416 are active remedies.

[0128] S415: Proactively push out remedial measures.

[0129] After obtaining remedial measures for a specific customer journey breakpoint, the customer journey breakpoint analysis tool proactively pushes these measures to the customer service center.

[0130] S416: Proactively contact customers to remedy any disruptions in their customer journey.

[0131] After receiving the remedial measures pushed by the customer journey breakpoint analysis tool, the customer service center proactively contacts the customer to remedy the customer journey breakpoint.

[0132] For example, after the remediation library is updated, new remediation measures are generated for service issues that were originally unsolvable. The customer journey breakpoint analysis tool sends the customer journey with the breakpoint and the corresponding remediation measures to the customer service center. The customer service center then proactively calls the customer to ask whether they need to continue the customer journey.

[0133] In this embodiment, without altering the customer service center's contact channels, the system analyzes contact records, text records, and call records stored by the customer service center from channels such as voice, video, and social media. This eliminates the need to modify contact point applications, making the method easier to implement. The customer journey breakpoint analysis tool utilizes data from contact records to statistically analyze relevant indicators. Combined with voice and text analysis tools, it analyzes and identifies customer journey breakpoints during customer service, identifying interaction breakpoints and automatically providing self-service and human-assisted remedial measures based on a large model and the remedial database of operations management personnel. This embodiment intelligently analyzes and mines unstructured call recordings and text records during customer service, enabling enterprise customer service centers to better understand the underlying reasons for customer journey breakpoints behind contact record indicators. It allows for targeted analysis of potential problems in each customer service process and automatic generation of remedial measures based on detailed information about customer journey breakpoints, thereby improving customer satisfaction and user experience and enhancing the role of the customer service center.

[0134] It is understood that, in order to achieve the aforementioned functions, the device includes corresponding hardware structures and / or software modules for performing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is implemented in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0135] This application embodiment can divide the data processing device into functional modules according to the above method example. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0136] Referring to Figure 5, which is a schematic diagram of the structure of a data processing device provided in an embodiment of this application, the data processing device 5 can be applied to the method shown in the method embodiment of Figure 4 above. As shown in Figure 5, the data processing device 5 includes a processing module 50 and a transceiver module 51. The processing module 50 may be one or more processors, and the transceiver module 51 may be a transceiver or a communication interface. The data processing device can be used to implement the steps involved in the data processing device execution in the above method embodiments. Optionally, the data processing device 5 may further include a storage module 52 for storing the program code and data of the data processing device 5.

[0137] In one example, the data processing device functions as a customer journey breakpoint analysis tool or is a chip applied within a customer journey breakpoint analysis tool, and executes the steps performed by the customer journey breakpoint analysis tool in the above method embodiments. The transceiver module 51 is used to specifically execute the sending and / or receiving actions performed by the customer journey breakpoint analysis tool in the embodiments shown in FIG4; for example, the customer journey breakpoint analysis tool performs other processes of the technology described herein. The processing module 50 can be used to support the data processing device 5 in performing the processing actions in the above method embodiments; for example, it can support the customer journey breakpoint analysis tool in performing other processes of the technology described herein.

[0138] For example, the transceiver module 51 is used to receive first data from the customer service center; to send remedial measures to the customer service center; to output a second index; and to receive breakpoint indication information.

[0139] Processing module 50 is configured to: analyze a first index of the customer journey based on text data transcribed from text records and / or call recordings using a first model; determine that there are no breakpoints in the customer journey if the first index is less than a first threshold; analyze a customer sentiment index in the customer journey based on text data transcribed from text records and / or call recordings using a second model; determine that there are breakpoints in the customer journey if the first index is not less than a first threshold and the sentiment index is less than a second threshold; obtain a satisfaction score based on customer satisfaction evaluation; determine that there are breakpoints in the customer journey if the first index is not less than a first threshold and the satisfaction score is less than a third threshold; and determine that the problem resolution rate is a key indicator for the customer journey if the first index is not less than a first threshold. If a customer's problem is not resolved, it is determined that there is a breakpoint in the customer journey; it is used to calculate a second index based on the emotional index, satisfaction score, and problem resolution rate when the first index is not less than the first threshold, the emotional index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved; it is used to calculate a second index by formula (1) when the first index is not less than the first threshold, the emotional index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved; it is used to perform intent recognition based on the first data through the third model when there is a breakpoint in the customer journey; it is used to generate remedial measures based on the customer's intent and combined with the remedial measure library.

[0140] In one possible implementation, the transceiver module 51 acquires first data, which includes text records and / or call recordings from the customer journey; the processing module 50 analyzes a first index of the customer journey based on the text data transcribed from the text records and / or call recordings using a first model, whereby the first index represents the probability of a breakpoint in the customer journey; if the first index is less than a first threshold, the processing module 50 determines that there is no breakpoint in the customer journey.

[0141] In one possible implementation, the first index is not less than the first threshold. The processing module 50 analyzes the customer's emotional index in the customer journey based on the text data transcribed from text records and / or call recordings using the second model. The emotional index represents the degree of positive emotions of the customer in the customer journey. If the emotional index is less than the second threshold, the processing module 50 determines that there is a breakpoint in the customer journey.

[0142] In one possible implementation, the first index is not less than a first threshold, and the first data also includes customer satisfaction evaluation; the processing module 50 obtains a satisfaction score based on the customer satisfaction evaluation, and the satisfaction score represents the customer's satisfaction with the customer journey; if the satisfaction score is less than a third threshold, the processing module 50 determines that there is a breakpoint in the customer journey.

[0143] In one possible implementation, the first index is not less than a first threshold, and the first data also includes a problem resolution rate; the processing module 50 determines that there is a breakpoint in the customer journey when the problem resolution rate indicates that the customer's problem has not been resolved.

[0144] In one possible implementation, the first index is not less than a first threshold, and the first data also includes customer satisfaction rating and problem resolution rate; the processing module 50 analyzes the customer's emotional index in the customer journey based on text data transcribed from text records and / or call recordings using a second model, the emotional index representing the degree of positive emotions the customer experiences in the customer journey; the processing module 50 obtains a satisfaction score based on the customer satisfaction rating, the satisfaction score representing the customer's satisfaction with the customer journey; when the emotional index is greater than or equal to the second threshold, the satisfaction score is greater than or equal to the third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, the processing module 50 calculates a second index based on the emotional index, satisfaction score, and problem resolution rate, the second index representing the probability of a breakpoint in the customer journey.

[0145] In one possible implementation, the first index is not less than a first threshold, and the first data also includes customer satisfaction ratings and problem resolution rates; the processing module 50 uses a formula... Calculate the second exponent;

[0146] Wherein, I is the second index, Q is the problem resolution rate, which indicates that Q equals 1 when the customer's problem has been resolved; E is the sentiment index, a is the parameter weight of the sentiment index, used to control the degree of influence of the sentiment index on the second index; C is the satisfaction score, b is the parameter weight of the satisfaction score, used to control the degree of influence of the satisfaction score on the second index.

[0147] In one possible implementation, the transceiver module 51 outputs a second index and receives breakpoint indication information, which indicates whether a breakpoint exists in the customer's journey.

[0148] In one possible implementation, when there is a break in the customer journey, the processing module 50 performs intent recognition based on the first data through a third model. The intent includes the customer's business needs. Based on the intent, the processing module 50 generates remedial measures in conjunction with a remedial measures library, which includes methods for implementing the intent.

[0149] In one possible implementation, the transceiver module 51 actively sends remedial measures, or after receiving a query signal, the transceiver module 51 sends remedial measures. The query signal includes querying whether there is a breakpoint in the customer journey and the corresponding remedial measures.

[0150] In one possible implementation, when the data processing device 5 is a chip for a customer journey breakpoint analysis tool, the transceiver module 51 can be a communication interface, pins, or circuits. The communication interface can be used to input data to be processed to the processor and can output the processor's processing results. Specifically, the communication interface can be a general purpose input / output (GPIO) interface, which can connect to multiple peripheral devices (such as LCD displays, cameras, radio frequency (RF) modules, antennas, etc.). The communication interface is connected to the processor via a bus.

[0151] The processing module 50 may be a processor, which can execute computer execution instructions stored in the storage module to cause the chip to execute the method involved in the embodiment shown in FIG4. Further, the processor may include a controller, an arithmetic logic unit (ALU), and registers. For example, the controller is mainly responsible for instruction decoding and issuing control signals for the operations corresponding to the instructions. The ALU is mainly responsible for performing fixed-point or floating-point arithmetic operations, shift operations, and logical operations, and can also perform address operations and conversions. The registers are mainly responsible for storing register operands and intermediate operation results temporarily stored during instruction execution. In specific implementations, the processor's hardware architecture may be an ASIC architecture, a microprocessor without interlocked piped stages architecture (MIPS) architecture, an advanced reduced instruction set machine (RISC) machine (ARM) architecture, or a network processor (NP) architecture, etc. The processor may be single-core or multi-core. The storage module may be an internal storage module of the chip, such as a register or cache. The storage module may also be an external storage module, such as ROM or other types of static storage devices that can store static information and instructions, RAM, etc.

[0152] It should be noted that the functions of the processor and interface can be implemented through hardware design, software design, or a combination of both; no restrictions are imposed here.

[0153] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0154] In summary, the above description is merely an embodiment of the technical solution of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made based on the disclosure of this application should be included within the scope of protection of this application.

Claims

1. A data processing method, characterized by, include: Acquire first data, which includes text records and / or call recordings from the customer journey; A first index of the customer journey is analyzed using a first model based on the text data transcribed from the text records and / or the call recordings. The first index represents the probability of a breakpoint in the customer journey. If the first index is less than the first threshold, it is determined that there are no breakpoints in the customer journey.

2. The method of claim 1, wherein, The first index is not less than the first threshold; the method further includes: The second model analyzes the customer's emotional index in the customer journey based on the text data transcribed from the text records and / or the call recordings. The emotional index represents the degree of positive emotions of the customer in the customer journey. If the sentiment index is less than the second threshold, it is determined that there is a breakpoint in the customer journey.

3. The method according to claim 1 or 2, characterized in that, The first index is not less than the first threshold, and the first data also includes customer satisfaction evaluation; the method further includes: The satisfaction score is obtained based on the customer satisfaction evaluation, and the satisfaction score represents the customer's level of satisfaction with the customer journey. If the satisfaction score is less than the third threshold, it is determined that there is a breakpoint in the customer journey.

4. The method according to any one of claims 1 to 3, characterized in that, The first index is not less than the first threshold, and the first data also includes the problem resolution rate; the method further includes: If the problem resolution rate indicates that the customer's problem has not been resolved, it is determined that there is a breakpoint in the customer journey.

5. The method according to any one of claims 1 to 4, characterized in that, The first index is not less than the first threshold, and the first data also includes customer satisfaction ratings and problem resolution rates; the method further includes: The second model analyzes the customer's emotional index in the customer journey based on the text data transcribed from the text records and / or the call recordings. The emotional index represents the degree of positive emotions of the customer in the customer journey. The satisfaction score is obtained based on the customer satisfaction evaluation, and the satisfaction score represents the customer's level of satisfaction with the customer journey. When the sentiment index is greater than or equal to a second threshold, the satisfaction score is greater than or equal to a third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, a second index is calculated based on the sentiment index, the satisfaction score, and the problem resolution rate. The second index represents the probability that there is a breakpoint in the customer journey.

6. The method of claim 5, wherein, When the sentiment index is greater than or equal to a second threshold, the satisfaction score is greater than or equal to a third threshold, and the problem resolution rate indicates that the customer's problem has been resolved, calculating the second index based on the sentiment index, the satisfaction score, and the problem resolution rate includes: By the formula Calculate the second exponent; Wherein, Q is the problem resolution rate, which indicates that Q equals 1 when the customer's problem has been resolved; E is the sentiment index, a is the parameter weight of the sentiment index, used to control the degree of influence of the sentiment index on the second index; C is the satisfaction score, b is the parameter weight of the satisfaction score, used to control the degree of influence of the satisfaction score on the second index.

7. The method according to claim 5 or 6, characterized in that, The method further includes: Output the second exponent; Receive breakpoint indication information, which indicates whether a breakpoint exists in the customer journey.

8. The method according to any one of claims 1 to 7, characterized in that, In the event of a breakpoint in the customer journey, the method further includes: The third model performs intent recognition based on the first data, and the intent includes the customer's business needs. Based on the stated intent, and in conjunction with a remedial measures library, remedial measures are generated, the remedial measures library including methods for implementing the stated intent.

9. The method of claim 8, wherein, The method further includes any one of the following: Actively send the remedial measures; Upon receiving a query signal, the remedial measures are sent. The query signal includes a query to determine if there is a breakpoint in the customer journey and the corresponding remedial measures.

10. A data processing device, characterized by Includes units or modules for implementing the method as described in any one of claims 1 to 9.

11. A computer readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed, cause the computer to perform the method as described in any one of claims 1 to 9.

12. A computer program product, characterised in that, The computer program product includes: computer program code, which, when executed by a computer, causes the computer to perform the method as described in any one of claims 1 to 9.

13. A chip, characterized by The chip includes at least one processor and an interface, the processor being configured to read and execute instructions stored in a memory, which, when executed, cause the chip to perform the method as described in any one of claims 1 to 9.

Images