Method for acquiring integrity of customer service session under large-scale seats of e-commerce platform

Abstract

The invention discloses a method for obtaining integrity of customer service sessions under large-scale seats of an e-commerce platform, and particularly relates to the technical field of business customer service session VOC / e-commerce customer service session quality inspection, and the method comprises the following steps: firstly, employing an encrypted account password to simulate and log in a customer service background, and keeping a login state; according to the difference between the monitoring information comparison library and the webpage data, dynamically obtaining customer service seat and customer list information at high frequency; calling an acquisition method at different frequencies according to the monitoring information, and obtaining complete dialogue information based on a customer list; and finally structuring dialogue information for subsequent analysis and quality inspection. According to the method, the problem of incomplete data acquisition is effectively solved, the completeness and timeliness of customer service session data acquisition are remarkably improved, the method has important application value in the fields of e-commerce customer service session VOC and quality inspection, and the improvement of the service quality of enterprise customer service and the optimization of service decisions are powerfully promoted.

Description

Technical Field

[0001] The present invention relates to the field of VOC / e-commerce customer service conversation quality inspection technology, and more specifically, to a method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform. Background Art

[0002] In the customer service industry, ensuring data integrity is key to ensuring service quality and customer satisfaction. However, existing collection solutions do not fully consider protecting customer privacy and data security while ensuring data integrity. Due to the limitations of websites (such as Taobao), the collected page data is not always complete. It usually only shows the most recent 500 sessions, while for busy stores, the number of consultations per day may be far more than this, especially during peak periods such as Double Eleven, when the number of consultations may surge to tens of thousands. Therefore, traditional collection tools often find it difficult to capture all session data.

[0003] We need a new type of collection solution that can collect as much session data as possible while respecting customer privacy and data security, so that companies can conduct comprehensive service quality analysis and promptly identify and resolve potential service issues. Ideally, this solution should be able to overcome website limitations and capture more session data, thereby providing more comprehensive and detailed service quality insights and improving customer satisfaction. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for obtaining the integrity of customer service conversations under large-scale seats on an e-commerce platform to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above purpose, the present invention provides the following technical solution: a method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform, comprising the following steps: Steps for simulating logging into the customer service backend: Use the encrypted backup account password and the encrypted account password to simulate a real person logging into the e-commerce platform customer service backend system, and obtain and keep the login status code; Information acquisition steps: Based on the monitoring information, compare the data differences in the database and on the web page, dynamically and frequently acquire customer service agent information, and frequently acquire customer list information based on the customers served by the agents; Conversation information collection steps: Based on the monitoring information, use different frequencies to call the collection method on the same day to obtain complete customer conversation information based on the customer list information; Information structuring step: Structure the customer's conversation information to provide backup data for subsequent data analysis and data quality inspection.

[0006] Preferably, in the step of simulating logging into the customer service background, the account used for logging in needs to be bound to a mobile phone, and operations are performed on a specific store on the deployed platform. The store has a corresponding store ID and name, and the login process needs to ensure that the data association is accurate.

[0007] Preferably, in the information acquisition step, data difference comparison and information acquisition are achieved through analysis of overview information, such as analysis of information including time, store-related data, etc., wherein the time information is accurate to the specific date and time, and the store-related data includes the store's identification on different platforms and related statistical data.

[0008] Preferably, in the conversation information collection step, the data of the day is collected first, and the collection process involves the inspection operation of cookies, and the mapping operation of obtaining the conversation information is performed by the thread pool executor according to the set maximum number of working threads, and the number of sessions that should be collected and actually collected is counted at the same time.

[0009] Preferably, in the information structuring step, the conversation information is structured according to a specific key-value pair format, wherein the key-value pair includes fields such as Objectld, shop_name, username, chat_time, and each field corresponds to corresponding information content, which is used to accurately describe relevant attributes of the customer conversation.

[0010] Technical effects and advantages of the present invention: The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform provided by the present invention first uses an encrypted account and password to simulate logging into the customer service backend and keep the login status; then dynamically and frequently obtains customer service seats and customer list information based on the difference between the monitoring information comparison library and the web page data; then calls the collection method at different frequencies based on the monitoring information to obtain complete conversation information based on the customer list; finally, the structured conversation information is used for subsequent analysis and quality inspection. It can effectively solve the problem of incomplete data collection and significantly improve the integrity and timeliness of customer service conversation data acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 The figure is a flow chart of the method of the present invention.

[0012] Figure 2 This is a display diagram of the simulated login customer service backend ID interface of the present invention.

[0013] Figure 3 It is a schematic diagram of list information of the present invention.

[0014] Figure 4 This is a schematic diagram of the customer dialogue information acquisition code of the present invention.

[0015] Figure 5The structured customer conversation information of the present invention is processed into a data schematic diagram. DETAILED DESCRIPTION

[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0017] As attached Figure 1-Figure 5 The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform is shown, and is specifically implemented through the following steps: As attached Figure 2 As shown, simulate logging into the customer service backend and obtaining and retaining the login status code: Use the encrypted backup account password to simulate real-person operation to log in to the customer service backend system.

[0018] Use the encrypted account password to simulate real-person operation to log in to the customer service backend system.

[0019] For example, for an official flagship store on the Taobao e-commerce platform (the tenant name has been deployed, the platform is [specific platform], the store ID is lw, the backend store ID is also lw, the account is the flagship store Yunting, the password is [specific password], and the account-bound mobile phone is 166), this method ensures that you can successfully log in to the customer service backend and maintain the login status, providing a basis for subsequent data collection.

[0020] As attached Figure 3 As shown, based on the difference between the data in the monitoring information comparison database and the data on the web page, the customer service agent information is obtained dynamically and frequently, and the customer list information is obtained frequently based on the customers served by the agents: Based on the overview data in the system (such as the information shown in the example including time, store-related data on different platforms, etc.), make a detailed comparison of the data in the database and on the web page to find out the differences.

[0021] Through dynamic and high-frequency methods, customer service agent information can be quickly and accurately obtained, and customer list information can be further obtained based on the customers served by the agents, ensuring the comprehensiveness and timeliness of information acquisition.

[0022] As attached Figure 4 As shown, based on the monitoring information, different frequencies are used to call the collection method on the same day to obtain complete customer conversation information based on the customer list information: First, check the cookie (as shown in the code self.cookie = self.check_cookie()) to ensure the stability of data collection.

[0023] For each date ori_day in the set date list self.day_lst, first count the number of conversations to be collected (ChatCountSpider(self.shop_info,day=ori_day)), and then collect the data for the day (after converting the date format, run it through TmalLChatCustomer(ori_day,self.shop_info).run()).

[0024] Get the customer list of the day customers=self.get_customer(day) and construct the parameter args=[(cus,day)forcusincustomers].

[0025] Use the ThreadPooLExecutor(max_workers=self.worker) thread pool executor, according to the set maximum number of working threads, through the executor.map(self.get_chat_wrap,*iterables:*zip(*args)) operation to obtain complete customer conversation information, and count the actual collection conversation volume (chat_count(self.shop_id,"chat",ori_day)).

[0026] As attached Figure 5 As shown, the structured customer conversation information is used for subsequent data analysis and data quality inspection: Structuring customer conversation information according to a specific key-value pair format (such as the fields including Objectld, shop_name, username, chat_time, etc. shown in the example).

[0027] For example, for a conversation message, its Objectld is "5de8d090edfb220d5ecfc18b", shop_name is [specific store name], username is [customer or customer service username], chat_time is "2019-12-050 [specific time]", etc. This information is accurately organized into structured data to facilitate subsequent efficient data analysis and strict data quality inspection.

[0028] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform, characterized in that: The following steps are involved: Steps for simulating logging into the customer service backend: Use the encrypted backup account password and the encrypted account password to simulate a real person logging into the e-commerce platform customer service backend system, and obtain and keep the login status code; Information acquisition steps: Based on the monitoring information, compare the data differences in the database and on the web page, dynamically and frequently acquire customer service agent information, and frequently acquire customer list information based on the customers served by the agents; Conversation information collection steps: Based on the monitoring information, use different frequencies to call the collection method on the same day to obtain complete customer conversation information based on the customer list information; Information structuring step: Structure the customer's conversation information to provide backup data for subsequent data analysis and data quality inspection.

2. The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform according to claim 1 is characterized in that: In the simulated login to the customer service background step, the account used for login needs to be bound to the mobile phone, and the operation is performed on a specific store on the deployed platform. The store has a corresponding store ID and name, and the login process needs to ensure that the data association is accurate.

3. The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform according to claim 1 is characterized in that: In the information acquisition step, data difference comparison and information acquisition are achieved through analysis of overview information, such as analysis of information including time, store-related data, etc., wherein the time information is accurate to the specific date and time, and the store-related data includes the store's identification on different platforms and related statistical data.

4. The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform according to claim 1 is characterized in that: In the conversation information collection step, the data of the day is collected first. The collection process involves checking the cookie, and the mapping operation of obtaining the conversation information is performed by the thread pool executor according to the set maximum number of working threads. At the same time, the number of sessions that should be collected and actually collected is counted.

5. The method for obtaining integrity of customer service conversations under large-scale seats on an e-commerce platform according to claim 1 is characterized in that: In the information structuring step, the conversation information is structured according to a specific key-value pair format, wherein the key-value pair includes fields such as Objectld, shop_name, username, and chat_time, and each field corresponds to corresponding information content, which is used to accurately describe the relevant attributes of the customer conversation.

Images