WO2020062672A1 - 一种评论信息的回复方法、存储介质、服务器和装置 - Google Patents

一种评论信息的回复方法、存储介质、服务器和装置 Download PDF

Info

Publication number
WO2020062672A1
WO2020062672A1 PCT/CN2018/123788 CN2018123788W WO2020062672A1 WO 2020062672 A1 WO2020062672 A1 WO 2020062672A1 CN 2018123788 W CN2018123788 W CN 2018123788W WO 2020062672 A1 WO2020062672 A1 WO 2020062672A1
Authority
WO
WIPO (PCT)
Prior art keywords
review
negative
information
customer service
service system
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/123788
Other languages
English (en)
French (fr)
Inventor
陈百门
徐志成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
OneConnect Smart Technology Co Ltd
Original Assignee
OneConnect Smart Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by OneConnect Smart Technology Co Ltd filed Critical OneConnect Smart Technology Co Ltd
Publication of WO2020062672A1 publication Critical patent/WO2020062672A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/01Customer relationship services
    • G06Q30/015Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
    • G06Q30/016After-sales

Definitions

  • the present application relates to the technical field of network information processing, and in particular, to a method, a storage medium, a server, and a device for replying to comment information.
  • APP STORE After a user downloads an APP from a mobile app store (APP STORE) and uses it, they usually comment on the app, and these reviews include some negative reviews. After receiving negative comments from users, if these negative comments are not processed, it will affect other users ’willingness to download the APP.
  • the customer service logs in to the system background of the application store and finds the negative feedback from the user; responds and appeases the users who feedback the negative feedback, and provides a solution to eliminate the negative feedback.
  • the customer service often needs to go to the system background to check whether there are new negative reviews, and the response efficiency is low.
  • the embodiments of the present application provide a method, a storage medium, a server, and a device for replying to review information, which can improve the reply response efficiency.
  • the first aspect of the embodiments of the present application provides a method for replying to comment information, including:
  • the number of threads and the timing interval used by the web crawler task are adjusted by the following steps:
  • a computer-readable storage medium stores computer-readable instructions.
  • the computer-readable instructions are executed by a processor, the computer-readable instructions are implemented as in the embodiments of the present application. Steps of the method for replying to the comment information presented in the first aspect.
  • a server including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, and the processor executes the computer
  • the following steps are implemented when the instruction is readable:
  • the number of threads and the timing interval used by the web crawler task are adjusted by the following steps:
  • a device for replying to comment information including:
  • Comment information grabbing module used to regularly crawl app store back-end system comment information using pre-built web crawler tasks
  • Negative evaluation screening module which is used to filter the captured comment information to obtain the negative evaluations therein;
  • a negative evaluation sending module configured to send the negative evaluation to a customer service system
  • a comment reply module configured to synchronize the reply information to the back-end system of the application store after obtaining the reply information of the customer service system;
  • the comment information grabbing module includes:
  • a parameter obtaining unit configured to obtain the release time and download times of each APP provided by the application store
  • a task adjustment unit is configured to adjust the number of threads and the timing interval used by the crawler task according to the publishing time and the number of downloads.
  • the method for replying to the comment information provided in the present application includes: periodically using a pre-built web crawler task to periodically capture the review information of the back-end system of the application store; filtering the captured review information to obtain the bad reviews therein; The review is sent to the customer service system; after the response information of the customer service system is obtained, the response information is synchronized to the back-end system of the application store.
  • the above process uses a web crawler task to periodically capture the review information of the back-end system of the app store, find the negative reviews, and then automatically forward these negative reviews to the customer service, so that the customer service does not need to log in to the app store system background to find the negative reviews. The earth has improved the response efficiency of comments.
  • FIG. 1 is a flowchart of a first embodiment of a method for replying to comment information according to an embodiment of the present application
  • FIG. 2 is a flowchart of a second embodiment of a method for replying to comment information provided by an embodiment of the present application
  • FIG. 3 is a structural diagram of an embodiment of an apparatus for replying to comment information according to an embodiment of the present application
  • FIG. 4 is a schematic diagram of a server according to an embodiment of the present application.
  • the embodiments of the present application provide a method, a storage medium and a server for replying to comment information, which can improve the response efficiency of comments.
  • a first embodiment of a method for replying to comment information in an embodiment of the present application includes:
  • a web crawler is a program or script that automatically grabs information from the World Wide Web in accordance with certain rules.
  • This application uses a web crawler technology to set a scheduled task for crawling evaluation information of the back-end system of the application store.
  • you can set the relevant properties of the crawler task such as whether to use a multi-threaded method, how many threads to start at the same time, the time interval between each time to crawl the data, and the time point to start the crawler task.
  • the number of threads and the timing interval used by the web crawler task are adjusted by the following steps:
  • the release time and number of downloads of each APP can reflect the popularity of downloading APPs, and the higher the popularity of downloading APPs, the greater the number of reviews published by users. Therefore, the number of threads and the timing interval used by the crawler task can be adjusted according to the release time and download times of each APP. For example, if the user's popularity of downloading the APP indicates that the amount of data to be crawled is large, the crawler task can be increased. The number of threads to use or reduce the time interval for crawling data.
  • time interval between adjusting the number of threads and the timing adopted by the crawler task according to the publishing time and the number of downloads may include:
  • An average release time is calculated according to the release time of each APP, and an average download time is calculated according to the download times of each APP;
  • the average number of downloads of each APP if the average number of downloads is greater than a preset number, it indicates that the user has a high download heat, so the number of threads used by the crawler task can be appropriately increased and / or the crawler task can be appropriately reduced Time interval.
  • time point of starting the crawler task can be determined by the following steps:
  • the back-end system can obtain each piece of published comment information and the corresponding posting time, and then count the number of published comment information published in each preset time period according to the posting time. For example, divide the 24 hours of a day into "0 points-1 point, 1 point-2 points, 2 points-3 points " for a total of 24 time periods, and then count the number of published comment information published in each time period , Find the time period in which the most number of published comment information has been posted as the target time period. Finally, a time point for starting the crawler task is selected from the target time period. Setting the time point for starting the crawler task within a time period with the highest probability of a user's comment information can further improve the efficiency of capturing comment information.
  • the review information of the back-end system of the app store is captured, because the captured review information is not all negative reviews, the review information needs to be filtered to find the negative reviews.
  • step 102 may include the following two ways:
  • the review information generally includes the star rating, and the star rating can be used to distinguish positive and negative reviews. For example, five stars represent satisfactory positive reviews, and 0-1 stars indicate negative reviews. Therefore, the review information with a rating star rating lower than a preset star rating may be determined as a bad review.
  • a user When a user expresses dissatisfaction and publishes a negative review, they generally publish some keywords or emoticons used to express negative reviews, such as negative reviews, bad, bad, bad, stuttered, slow response, etc. Symbols can include “bad”, “angry”, “despise”, “failing”, and “halo” emoticons used to express negative emotions. Therefore, whether the review information is a bad review can be determined by detecting whether the review information contains keywords or emoticons used to indicate a negative review.
  • the negative comments are sent to the customer service system. After logging in to the customer service system, customer service personnel can see these negative reviews, so that they can respond to these negative reviews in time to appease the users who posted the negative reviews.
  • the response information can be synchronized to the back-end system of the application store by using the crawler technology, so that the customer who sent the negative review can see the response from the customer service.
  • the method for replying to the review information provided in the embodiment of the present application includes: periodically using a pre-built web crawler task to periodically capture review information of an app store back-end system; filtering the captured review information to obtain a bad review therein; The negative feedback is sent to the customer service system; after the response information of the customer service system is obtained, the response information is synchronized to the back-end system of the application store.
  • the above process uses a web crawler task to periodically capture the review information of the back-end system of the app store, find the negative reviews, and then automatically forward these negative reviews to the customer service, so that the customer service does not need to log in to the application store's system background to find the negative reviews.
  • the earth has improved the response efficiency of comments.
  • a second embodiment of a method for replying to comment information in an embodiment of the present application includes:
  • Steps 201-202 are the same as steps 101-102. For details, refer to the description of steps 101-102.
  • the negative evaluation After the negative evaluation is obtained, it is detected whether the negative evaluation includes a preset target keyword.
  • This embodiment needs to classify each negative evaluation, and the target keywords included in the negative evaluation are the basis for classification. For example, if a negative review contains target keywords such as “account, OTP, password login, login failure, invalid verification code”, the negative review can be classified as a user login. If a negative review contains target keywords such as "expensive, unworthy, and high price”, the negative review can be classified as a price. If the negative evaluation includes a target keyword, step 204 is performed; if the negative evaluation does not include a target keyword, step 205 is directly performed.
  • the negative evaluation includes a target keyword, and at this time, a category label corresponding to the target keyword is added to the negative evaluation.
  • the correspondence between keywords and category tags can be constructed in advance, such as the keyword “expensive, worthless, high price” corresponding to the category tag "price”, the keywords “account, OTP, password login, login failure, Invalid verification code "corresponds to the category label" Login ". That is, if a negative review includes the keyword “expensive”, a category tag "price” is added to the negative review.
  • step 206 Before sending a bad review to the customer service system, it checks if the bad review contains a category tag. If the negative review includes a category label, step 206 is performed; if the negative review does not include a category label, step 207 is performed.
  • 206 Query response information corresponding to the category tag from a pre-built knowledge base, and send the queried response information and the negative evaluation to a customer service system;
  • the negative review includes a category tag.
  • the response information corresponding to the category tag is queried from a pre-built knowledge base, and the queried response information and the negative review are sent to a customer service system.
  • the knowledge base collects conventional reply information of various categories of bad reviews in advance, and these reply information are associated with corresponding category tags. Therefore, after determining the category label of a negative review, the general reply information corresponding to the category label can be found from the knowledge base, and then the query response information and the negative review are sent to the customer service system together.
  • customer service personnel can refer to the response information provided by the knowledge base when responding to conventional questions, and can directly improve the response efficiency of comments by directly adopting the response information or making simple modifications to the response information. After step 206, go to step 208.
  • Steps 207-208 are the same as steps 103-104. For details, refer to the related descriptions of steps 103-104.
  • the negative review may further include:
  • the response information of the customer service system is associated with the category tag "price” and stored in the knowledge base.
  • the regular response content of the knowledge base can be continuously enriched, which is beneficial to the work of negative feedback in the future.
  • associating the response information of the customer service system with the category tag and storing the response in the knowledge base may include:
  • the response effect of the response information can be checked first. If the response is good and the user who posted the bad review is satisfied with the processing result, the bad review will be eliminated. Therefore, the effect of replying to the reply message can be evaluated by judging whether the negative review has been eliminated after a preset time period (such as one week). If the negative evaluation has been eliminated, it indicates that the response of the response information is good. At this time, the response information is stored in the knowledge base, and can be used directly when responding to the same type of negative evaluation next time.
  • a preset time period such as one week
  • the method for replying to the review information provided in the embodiment of the present application includes: periodically using a pre-built web crawler task to periodically capture review information of an app store back-end system; filtering the captured review information to obtain a bad review therein; Whether the negative evaluation includes a preset target keyword; if the negative evaluation includes the target keyword, add a category tag corresponding to the target keyword to the negative evaluation; and send the negative evaluation to the customer service system Before, it is detected whether the negative review contains a category tag; if the negative review contains a category tag, the reply information corresponding to the category tag is queried from a pre-built knowledge base, and the queried reply information and the The negative evaluation is sent to the customer service system; after the response information of the customer service system is obtained, the response information of the customer service system is synchronized to the back-end system of the application store.
  • the embodiment of the present application classifies the collected negative reviews, and constructs a knowledge base in advance to store the conventional reply information corresponding to the negative reviews of each category, and sends the negative reviews and corresponding conventional reply information to the customer service system, so that The personnel can refer to the general reply information to easily determine the current reply content. Compared with the first embodiment of the present application, the response efficiency of the negative reviews is further improved.
  • the above mainly describes a method for replying to comment information, and a device for replying to comment information will be described in detail below.
  • an embodiment of a device for replying to comment information in an embodiment of the present application includes:
  • the review information grabbing module 301 is used to regularly capture the review information of the backend system of the application store using a pre-built web crawler task;
  • the negative evaluation screening module 302 is configured to filter the captured comment information to obtain a negative evaluation therein;
  • a negative evaluation sending module 303 configured to send the negative evaluation to a customer service system
  • the comment reply module 304 is configured to synchronize the reply information to the back-end system of the application store after obtaining the reply information of the customer service system;
  • the comment information grabbing module includes:
  • a parameter obtaining unit configured to obtain the release time and download times of each APP provided by the application store
  • a task adjustment unit is configured to adjust the number of threads and the timing interval used by the crawler task according to the publishing time and the number of downloads.
  • the task adjustment unit may include:
  • a parameter calculation subunit configured to calculate an average release time according to the release time of each APP, and calculate an average download time according to the download times of each APP;
  • a first adjustment subunit configured to increase the number of threads used by the crawler task by a preset ratio and / or reduce the preset ratio if the difference between the average release time and the current time is less than a preset duration; Time interval;
  • the second adjusting subunit is configured to increase the number of threads used by the crawler task according to a preset ratio and / or reduce the timing interval according to a preset ratio if the average download times is greater than a preset number.
  • the negative evaluation screening module may include:
  • a star rating obtaining unit configured to obtain a star rating of the review information
  • a first negative rating determining unit configured to determine review information with a star rating lower than a preset star rating as a negative rating
  • An information detection unit configured to detect whether the comment information includes keywords or emoticons used to indicate a negative evaluation
  • the second negative evaluation determining unit is configured to determine the comment information including the keywords or emoticons as a negative evaluation.
  • apparatus for replying to comment information may further include:
  • a keyword detection module configured to detect whether the negative evaluation includes a preset target keyword
  • a category tag adding module configured to add a category tag corresponding to the target keyword to the negative review if the target keyword is included in the negative review;
  • the negative evaluation sending module may include:
  • a category label detection unit configured to detect whether the negative review contains a category label
  • a first negative review sending unit configured to query the response information corresponding to the category tag from a pre-built knowledge base if the negative review includes a category tag, and send the queried response information and the negative review To customer service system;
  • the second negative evaluation sending unit is configured to send the negative evaluation to a customer service system if the negative evaluation does not include a category label.
  • apparatus for replying to comment information may further include:
  • An information association module is configured to associate the response information of the customer service system with the category tag, and store the response information in the knowledge base.
  • the information association module may include:
  • a negative evaluation elimination judging unit configured to determine whether the negative evaluation has been eliminated after synchronizing the reply information to a preset duration of the back-end system of the application store;
  • An information association unit is configured to associate the response information of the customer service system with the category tag and store it in the knowledge base if the negative evaluation has been eliminated.
  • An embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, any one shown in FIG. 1 or FIG. 2 is implemented. Steps of a reply method for comment information.
  • An embodiment of the present application further provides a server, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor.
  • the processor is implemented when the processor executes the computer-readable instructions. The steps of any method for replying to comment information as shown in FIG. 1 or FIG. 2.
  • FIG. 4 is a schematic diagram of a server provided by an embodiment of the present application.
  • the server 4 of this embodiment includes a processor 40, a memory 41, and computer-readable instructions 42 stored in the memory 41 and executable on the processor 40.
  • the processor 40 executes the computer-readable instructions 42
  • the steps in the embodiment of the method for replying to the above review information are implemented, for example, steps 101 to 104 shown in FIG. 1.
  • the processor 40 executes the computer-readable instructions 42
  • the functions of the modules / units in the foregoing device embodiments are implemented, for example, the functions of the modules 301 to 304 shown in FIG. 3.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本申请涉及网络信息处理技术领域,提出一种评论信息的回复方法、存储介质和服务器。所述评论信息的回复方法包括:使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;对抓取到的评论信息进行筛选,得到其中的差评;将所述差评发送给客服系统;在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统。上述过程采用网络爬虫任务定时抓取应用商店后端系统的评论信息,找到其中的差评,然后将这些差评自动转发给客服,从而使得客服无需自行登录应用商店的系统后台查找差评,极大地提高了评论的回复效率。

Description

一种评论信息的回复方法、存储介质、服务器和装置
本申请要求于2018年9月26日提交中国专利局、申请号为201811122198.5、申请名称为“一种评论信息的回复方法、存储介质和服务器”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及网络信息处理技术领域,尤其涉及一种评论信息的回复方法、存储介质、服务器和装置。
背景技术
用户在手机应用商店(APP STORE)下载APP并使用之后,通常会对该APP进行评论,这些评论包含部分差评。而在接收到用户的差评后,若不对这些差评进行处理,会影响其它用户下载该APP的意愿。针对这个问题,目前通常采用以下方式处理:由客服登录应用商店的系统后台,找到用户反馈的差评;对反馈差评的用户进行回复和安抚,提供消除差评的解决方案。然而,采用这种方式需要客服经常去系统后台查看是否有新的差评,回复效率低。
技术问题
本申请实施例提供了一种评论信息的回复方法、存储介质、服务器和装置,能够提高评论的回复效率。
技术解决方案
本申请实施例的第一方面,提供了一种评论信息的回复方法,包括:
使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
对抓取到的评论信息进行筛选,得到其中的差评;
将所述差评发送给客服系统;
在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
获取所述应用商店提供的各个APP的发布时间和下载次数;
根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
本申请实施例的第二方面,提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可读指令,所述计算机可读指令被处理器执行时实现如本申请实施例的第一方面提出的评论信息的回复方法的步骤。
本申请实施例的第三方面,提供了一种服务器,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现如下步骤:
使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
对抓取到的评论信息进行筛选,得到其中的差评;
将所述差评发送给客服系统;
在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
获取所述应用商店提供的各个APP的发布时间和下载次数;
根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
本申请实施例的第四方面,提供了一种评论信息的回复装置,包括:
评论信息抓取模块,用于使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
差评筛选模块,用于对抓取到的评论信息进行筛选,得到其中的差评;
差评发送模块,用于将所述差评发送给客服系统;
评论回复模块,用于在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
其中,所述评论信息抓取模块包括:
参数获取单元,用于获取所述应用商店提供的各个APP的发布时间和下载次数;
任务调整单元,用于根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
有益效果
本申请提出的评论信息的回复方法包括:使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;对抓取到的评论信息进行筛选,得到其中的差评;将所述差评发送给客服系统;在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统。上述过程采用网络爬虫任务定时抓取应用商店后端系统的评论信息,找到其中的差评,然后将这些差评自动转发给客服,从而使得客服无需自行登录应用商店的系统后台查找差评,极大地提高了评论的回复效率。
附图说明
图1是本申请实施例提供的一种评论信息的回复方法的第一个实施例的流程图;
图2是本申请实施例提供的一种评论信息的回复方法的第二个实施例的流程图;
图3是本申请实施例提供的一种评论信息的回复装置的一个实施例的结构图;
图4是本申请实施例提供的一种服务器的示意图。
本发明的实施方式
本申请实施例提供了一种评论信息的回复方法、存储介质和服务器,能够提高评论的回复效率。
请参阅图1,本申请实施例中一种评论信息的回复方法的第一个实施例包括:
101、使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
首先,使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息。网络爬虫是一种按照一定的规则,自动地抓取万维网信息的程序或者脚本。本申请利用网络爬虫技术,设置定时任务,用于爬取应用商店后端系统的评价信息。在配置爬虫任务时,可以设置爬虫任务的相关属性,比如是否采用多线程方式,同时开启多少个线程去爬,每次去爬取数据的时间间隔和启动爬虫任务的时间点等。
其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
(1)获取所述应用商店提供的各个APP的发布时间和下载次数;
(2)根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
各个APP的发布时间和下载次数能够体现用户下载APP的热度,而下载APP的热度越高,则用户发表评价的数量也越多。因此,可以根据各个APP的发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔,比如,若用户下载APP的热度高,表明需要爬取的数据量大,可以增加爬虫任务采用的线程数量或者减少爬取数据的时间间隔。
进一步的,所述根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔可以包括:
(1)根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
(2)若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
(3)若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
在计算得到各个APP的平均发布时间之后,若该平均发布时间和当前时间之间的差值小于预设时长,表明各个APP大多处于新发布的状态,此时用户的下载热度较高,故可以适当提高爬虫任务采用的线程数量和/或适当降低爬虫任务的定时时间间隔。
类似的,在计算得到各个APP的平均下载次数之后,若该平均下载次数大于预设的次数,表明用户的下载热度较高,故可以适当提高爬虫任务采用的线程数量和/或适当降低爬虫任务的定时时间间隔。
进一步的,启动爬虫任务的时间点可以通过以下步骤确定:
(1)获取后端系统的历史评论信息记录;
(2)根据所述历史评论信息记录确定各个已发布评论信息的发表时间;
(3)按照所述发表时间分别统计各个预设时间段内发表的已发布评论信息的数量;
(4)从所述各个预设时间段中确定发表的已发布评论信息的数量最多的一个目标时间段;
(5)从所述目标时间段中选取启动爬虫任务的时间点。
后端系统可以获取到已发布的每条评论信息以及对应的发表时间,然后根据发表时间分别统计各个预设时间段内发表的已发布评论信息的数量。比如将一天的24个小时划分为“0点-1点、1点-2点、2点-3点…”共计24个时间段,然后统计每一个时间段内发表的已发布评论信息的数量,找出其中发表的已发布评论信息的数量最多的一个时间段作为目标时间段。最后,从所述目标时间段中选取启动爬虫任务的时间点。将启动爬虫任务的时间点设置于用户发表评论信息概率最大的时间段内,能够进一步提高抓取评论信息的效率。
102、对抓取到的评论信息进行筛选,得到其中的差评;
在抓取应用商店后端系统的评论信息之后,由于抓取到的评论信息不全是差评,故需要对这些评论信息进行筛选,找出其中的差评。
进一步的,步骤102可以包括以下两种方式:
方式1:
(1)获取所述评论信息的评分星级;
(2)将评分星级小于预设星级的评论信息确定为差评。
评论信息一般包含星级的评分,而评分星级能够用来区分好评和差评,比如五星代表满意的好评,0-1星表示差评等。因此,可以将评分星级小于预设星级的评论信息确定为差评。
方式2:
(1)检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
(2)将包含所述关键词或表情符的评论信息确定为差评。
当用户表达不满而发表差评时,一般会发表一些用于表示负面评价的关键词或表情符,比如差评、不好、踩、差劲、卡顿、反应慢等负面评价的关键词,表情符则可以包括“差劲”、“生气”、“鄙视”、“衰”和“晕”等用于表达负面情绪的表情符。因此,可以通过检测评论信息中是否包含用于表示负面评价的关键词或表情符来确定该评论信息是否为差评。
103、将所述差评发送给客服系统;
在筛选出评论信息中的差评之后,将所述差评发送给客服系统。客服人员在登录客服系统之后,即可看到这些差评,从而及时对这些差评进行回复,以安抚发表差评的用户。
104、在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统。
在客服人员通过客服系统返回相应的回复信息之后,可以利用爬虫技术将该回复信息同步到所述应用商店的后端系统,使得发送差评的客户能够看到客服的回复。
本申请实施例提出的评论信息的回复方法包括:使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;对抓取到的评论信息进行筛选,得到其中的差评;将所述差评发送给客服系统;在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统。上述过程采用网络爬虫任务定时抓取应用商店后端系统的评论信息,找到其中的差评,然后将这些差评自动转发给客服,从而使得客服无需自行登录应用商店的系统后台查找差评,极大地提高了评论的回复效率。
请参阅图2,本申请实施例中一种评论信息的回复方法的第二个实施例包括:
201、使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
202、对抓取到的评论信息进行筛选,得到其中的差评;
步骤201-202与步骤101-102相同,具体可参照步骤101-102的相关说明。
203、检测所述差评中是否包含预设的目标关键词;
在获得差评之后,检测所述差评中是否包含预设的目标关键词。本实施例需要为各个差评进行分类,而差评中包含的目标关键词则是分类的依据。比如,若某个差评中包含“账号、OTP、密码登录、登录失败、验证码无效”等目标关键词,则可以将该差评归类至用户登录方面。若某个差评中包含“贵、不值、价格高”等目标关键词,则可以将该差评归类至价格方面。若所述差评中包含目标关键词,则执行步骤204;若所述差评中不包含目标关键词,则直接执行步骤205。
204、为所述差评添加与所述目标关键词对应的类别标签;
所述差评中包含目标关键词,此时为所述差评添加与所述目标关键词对应的类别标签。具体的,可以预先构建关键词和类别标签之间的对应关系,比如关键词“贵、不值、价格高”与类别标签“价格”对应,关键词“账号、OTP、密码登录、登录失败、验证码无效”与类别标签“登录”对应。也即,若某个差评中包含关键词“贵”,则为该差评添加类别标签“价格”。
205、检测所述差评是否包含类别标签;
在将差评发送至客服系统之前,会检测差评是否包含类别标签。若所述差评包含类别标签,则执行步骤206;若所述差评不包含类别标签,则执行步骤207。
206、从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
所述差评包含类别标签,此时从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统。所述知识库预先收集各种类别差评的常规回复信息,这些回复信息和相应的类别标签关联。因此,在确定某条差评的类别标签后,可以从该知识库中找到与该类别标签对应的常规回复信息,然后将查询到的回复信息和差评一起发送给客服系统。通过这样设置,客服人员在答复常规问题时,可以参照知识库提供的回复信息,通过直接采用该回复信息或者对该回复信息做简单的修改,能够有效地提高评论的回复效率。在步骤206之后,跳转至步骤208。
207、将所述差评发送给客服系统;
208、在获得所述客服系统的回复信息之后,将所述客服系统的回复信息同步至所述应用商店的后端系统。
步骤207-208与步骤103-104相同,具体可参照步骤103-104的相关说明。
进一步的,若所述差评包含类别标签,则在步骤208之后,还可以包括:
将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
比如,若所述差评包含类别标签“价格”,则将所述客服系统的回复信息与类别标签“价格”关联,并存储至所述知识库中。通过这样设置,能够不断补充丰富知识库的常规答复内容,有利于以后的差评答复工作。
更进一步的,所述将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中可以包括:
(1)在将所述回复信息同步至所述应用商店的后端系统的预设时长后,判断所述差评是否已消除;
(2)若所述差评已消除,则将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
为了避免知识库存储过多的信息,在将客服系统的回复信息存储至知识库之前,可以先检验该回复信息的答复效果。若答复效果好,发表差评的用户满意处理结果,则会消除该差评。因此,可以通过判断在预设时长后(比如一周)差评是否已消除来评价回复信息的答复效果。若差评已消除,表明回复信息的答复效果好,此时将该回复信息存储至知识库,下次答复相同类型的差评时可以直接采用。
本申请实施例提出的评论信息的回复方法包括:使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;对抓取到的评论信息进行筛选,得到其中的差评;检测所述差评中是否包含预设的目标关键词;若所述差评中包含目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;在将差评发送至客服系统之前,检测所述差评是否包含类别标签;若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;在获得所述客服系统的回复信息之后,将所述客服系统的回复信息同步至所述应用商店的后端系统。本申请实施例会将采集到的差评进行分类,并预先构建一个知识库用于存储各个类别的差评对应的常规回复信息,将差评和对应的常规回复信息一起发送给客服系统,使得客服人员能够参照该常规回复信息,方便地确定当次的回复内容。与本申请的第一个实施例相比,进一步提高了差评的回复效率。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
上面主要描述了一种评论信息的回复方法,下面将对一种评论信息的回复装置进行详细描述。
请参阅图3,本申请实施例中一种评论信息的回复装置的一个实施例包括:
评论信息抓取模块301,用于使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
差评筛选模块302,用于对抓取到的评论信息进行筛选,得到其中的差评;
差评发送模块303,用于将所述差评发送给客服系统;
评论回复模块304,用于在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
其中,所述评论信息抓取模块包括:
参数获取单元,用于获取所述应用商店提供的各个APP的发布时间和下载次数;
任务调整单元,用于根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
进一步的,所述任务调整单元可以包括:
参数计算子单元,用于根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
第一调整子单元,用于若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
第二调整子单元,用于若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
进一步的,所述差评筛选模块可以包括:
星级获取单元,用于获取所述评论信息的评分星级;
第一差评确定单元,用于将评分星级小于预设星级的评论信息确定为差评;
信息检测单元,用于检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
第二差评确定单元,用于将包含所述关键词或表情符的评论信息确定为差评。
进一步的,所述评论信息的回复装置还可以包括:
关键词检测模块,用于检测所述差评中是否包含预设的目标关键词;
类别标签添加模块,用于若所述差评中包含所述目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;
所述差评发送模块可以包括:
类别标签检测单元,用于检测所述差评是否包含类别标签;
第一差评发送单元,用于若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
第二差评发送单元,用于若所述差评不包含类别标签,则将所述差评发送给客服系统。
进一步的,所述评论信息的回复装置还可以包括:
信息关联模块,用于将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
更进一步的,所述信息关联模块可以包括:
差评消除判断单元,用于在将所述回复信息同步至所述应用商店的后端系统的预设时长后,判断所述差评是否已消除;
信息关联单元,用于若所述差评已消除,则将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
本申请实施例还提供一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可读指令,所述计算机可读指令被处理器执行时实现如图1或图2表示的任意一种评论信息的回复方法的步骤。
本申请实施例还提供一种服务器,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现如图1或图2表示的任意一种评论信息的回复方法的步骤。
图4是本申请一实施例提供的服务器的示意图。如图4所示,该实施例的服务器4包括:处理器40、存储器41以及存储在所述存储器41中并可在所述处理器40上运行的计算机可读指令42。所述处理器40执行所述计算机可读指令42时实现上述各个评论信息的回复方法实施例中的步骤,例如图1所示的步骤101至104。或者,所述处理器40执行所述计算机可读指令42时实现上述各装置实施例中各模块/单元的功能,例如图3所示模块301至304的功能。

Claims (20)

  1. 一种评论信息的回复方法,其特征在于,包括:
    使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
    对抓取到的评论信息进行筛选,得到其中的差评;
    将所述差评发送给客服系统;
    在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
    其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
    获取所述应用商店提供的各个APP的发布时间和下载次数;
    根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
  2. 根据权利要求1所述的评论信息的回复方法,其特征在于,所述根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔包括:
    根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
    若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
    若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
  3. 根据权利要求1所述的评论信息的回复方法,其特征在于,所述对抓取到的评论信息进行筛选,得到其中的差评包括:
    获取所述评论信息的评分星级;
    将评分星级小于预设星级的评论信息确定为差评;
    或者
    检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
    将包含所述关键词或表情符的评论信息确定为差评。
  4. 根据权利要求1至3中任一项所述的评论信息的回复方法,其特征在于,在对抓取到的评论信息进行筛选,得到其中的差评之后,还包括:
    检测所述差评中是否包含预设的目标关键词;
    若所述差评中包含所述目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;
    所述将所述差评发送给客服系统包括:
    检测所述差评是否包含类别标签;
    若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
    若所述差评不包含类别标签,则将所述差评发送给客服系统。
  5. 根据权利要求4所述的评论信息的回复方法,其特征在于,若所述差评包含类别标签,则在获得所述客服系统的回复信息之后,还包括:
    将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
  6. 根据权利要求5所述的评论信息的回复方法,其特征在于,所述将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中包括:
    在将所述回复信息同步至所述应用商店的后端系统的预设时长后,判断所述差评是否已消除;
    若所述差评已消除,则将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
  7. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被处理器执行时实现如下步骤:
    使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
    对抓取到的评论信息进行筛选,得到其中的差评;
    将所述差评发送给客服系统;
    在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
    其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
    获取所述应用商店提供的各个APP的发布时间和下载次数;
    根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
  8. 根据权利要求7所述的计算机可读存储介质,其特征在于,所述根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔包括:
    根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
    若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
    若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
  9. 根据权利要求7所述的计算机可读存储介质,其特征在于,所述对抓取到的评论信息进行筛选,得到其中的差评包括:
    获取所述评论信息的评分星级;
    将评分星级小于预设星级的评论信息确定为差评;
    或者
    检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
    将包含所述关键词或表情符的评论信息确定为差评。
  10. 根据权利要求7至9中任一项所述的计算机可读存储介质,其特征在于,在对抓取到的评论信息进行筛选,得到其中的差评之后,还包括:
    检测所述差评中是否包含预设的目标关键词;
    若所述差评中包含所述目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;
    所述将所述差评发送给客服系统包括:
    检测所述差评是否包含类别标签;
    若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
    若所述差评不包含类别标签,则将所述差评发送给客服系统。
  11. 一种服务器,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,其特征在于,所述处理器执行所述计算机可读指令时实现如下步骤:
    对抓取到的评论信息进行筛选,得到其中的差评;
    将所述差评发送给客服系统;
    在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
    其中,所述网络爬虫任务采用的线程数量和定时的时间间隔通过以下步骤调整:
    获取所述应用商店提供的各个APP的发布时间和下载次数;
    根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
  12. 根据权利要求11所述的服务器,其特征在于,所述根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔包括:
    根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
    若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
    若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
  13. 根据权利要求11所述的服务器,其特征在于,所述对抓取到的评论信息进行筛选,得到其中的差评包括:
    获取所述评论信息的评分星级;
    将评分星级小于预设星级的评论信息确定为差评;
    或者
    检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
    将包含所述关键词或表情符的评论信息确定为差评。
  14. 根据权利要求11至13中任一项所述的服务器,其特征在于,在对抓取到的评论信息进行筛选,得到其中的差评之后,还包括:
    检测所述差评中是否包含预设的目标关键词;
    若所述差评中包含所述目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;
    所述将所述差评发送给客服系统包括:
    检测所述差评是否包含类别标签;
    若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
    若所述差评不包含类别标签,则将所述差评发送给客服系统。
  15. 根据权利要求14所述的服务器,其特征在于,若所述差评包含类别标签,则在获得所述客服系统的回复信息之后,还包括:
    将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
  16. 一种评论信息的回复装置,其特征在于,包括:
    评论信息抓取模块,用于使用预先构建的网络爬虫任务定时抓取应用商店后端系统的评论信息;
    差评筛选模块,用于对抓取到的评论信息进行筛选,得到其中的差评;
    差评发送模块,用于将所述差评发送给客服系统;
    评论回复模块,用于在获得所述客服系统的回复信息之后,将所述回复信息同步至所述应用商店的后端系统;
    其中,所述评论信息抓取模块包括:
    参数获取单元,用于获取所述应用商店提供的各个APP的发布时间和下载次数;
    任务调整单元,用于根据所述发布时间和下载次数调整所述爬虫任务采用的线程数量和定时的时间间隔。
  17. 根据权利要求16所述的评论信息的回复装置,其特征在于,所述任务调整单元包括:
    参数计算子单元,用于根据所述各个APP的发布时间计算得到平均发布时间,以及根据所述各个APP的下载次数计算得到平均下载次数;
    第一调整子单元,用于若所述平均发布时间和当前时间之间的差值小于预设时长,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔;
    第二调整子单元,用于若所述平均下载次数大于预设次数,则按预设比例提高所述爬虫任务采用的线程数量和/或按预设比例降低所述定时的时间间隔。
  18. 根据权利要求16所述的评论信息的回复装置,其特征在于,所述差评筛选模块包括:
    星级获取单元,用于获取所述评论信息的评分星级;
    第一差评确定单元,用于将评分星级小于预设星级的评论信息确定为差评;
    信息检测单元,用于检测所述评论信息中是否包含用于表示负面评价的关键词或表情符;
    第二差评确定单元,用于将包含所述关键词或表情符的评论信息确定为差评。
  19. 根据权利要求16至18中任一项所述的评论信息的回复装置,其特征在于,还包括:
    关键词检测模块,用于检测所述差评中是否包含预设的目标关键词;
    类别标签添加模块,用于若所述差评中包含所述目标关键词,则为所述差评添加与所述目标关键词对应的类别标签;
    所述差评发送模块包括:
    类别标签检测单元,用于检测所述差评是否包含类别标签;
    第一差评发送单元,用于若所述差评包含类别标签,则从预先构建的知识库中查询与所述类别标签对应的回复信息,并将查询到的回复信息和所述差评发送给客服系统;
    第二差评发送单元,用于若所述差评不包含类别标签,则将所述差评发送给客服系统。
  20. 根据权利要求19所述的评论信息的回复装置,其特征在于,还包括:
    信息关联模块,用于将所述客服系统的回复信息与所述类别标签关联,存储至所述知识库中。
PCT/CN2018/123788 2018-09-26 2018-12-26 一种评论信息的回复方法、存储介质、服务器和装置 Ceased WO2020062672A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811122198.5A CN109388748A (zh) 2018-09-26 2018-09-26 一种评论信息的回复方法、存储介质和服务器
CN201811122198.5 2018-09-26

Publications (1)

Publication Number Publication Date
WO2020062672A1 true WO2020062672A1 (zh) 2020-04-02

Family

ID=65418206

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/123788 Ceased WO2020062672A1 (zh) 2018-09-26 2018-12-26 一种评论信息的回复方法、存储介质、服务器和装置

Country Status (2)

Country Link
CN (1) CN109388748A (zh)
WO (1) WO2020062672A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114417178A (zh) * 2021-12-22 2022-04-29 北京三快在线科技有限公司 一种信息处理方法、装置、电子设备及存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110807091B (zh) * 2019-03-01 2023-08-18 王涵 一种酒店智能问答推荐与决策支持分析方法及系统
CN109684560B (zh) * 2019-03-19 2020-09-18 深圳市蓝禾技术有限公司 商品评价信息的推送方法、系统及服务器
CN110134856A (zh) * 2019-04-12 2019-08-16 平安国际智慧城市科技股份有限公司 应用程序评论的监控方法、装置及存储介质、计算机设备
CN110555716A (zh) * 2019-07-24 2019-12-10 深圳壹账通智能科技有限公司 数据处理方法、装置、计算机设备和存储介质
CN111340583B (zh) * 2020-02-13 2023-05-23 北京字节跳动网络技术有限公司 信息获取方法、装置、计算机设备和存储介质
CN111400437A (zh) * 2020-02-19 2020-07-10 北京三快在线科技有限公司 互联网信息响应方法、装置、电子设备和计算机可读介质
CN111539775B (zh) * 2020-05-07 2024-04-30 网易(杭州)网络有限公司 应用程序的管理方法及设备
CN112598427B (zh) * 2020-12-14 2024-12-17 珠海格力电器股份有限公司 售后服务的方法、装置、电子设备及存储介质
CN114666670B (zh) * 2022-02-24 2023-08-04 北京青藤文化股份有限公司 数据监控方法、装置、设备及计算机可读介质
CN115017388A (zh) * 2022-04-19 2022-09-06 上海华客信息科技有限公司 点评回复方法、系统、设备及存储介质

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106021391A (zh) * 2016-05-11 2016-10-12 广东工业大学 基于Storm的产品评论信息实时采集方法
CN107943787A (zh) * 2017-11-16 2018-04-20 北京百度网讯科技有限公司 收集用户反馈的方法、装置、设备及计算机可读介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130026558A (ko) * 2011-07-27 2013-03-14 (주)케이원리더스 외곽 공간을 이용한 댓글 통합 시스템 및 그 제공방법
KR101655994B1 (ko) * 2016-02-15 2016-09-08 이명환 고객 항의 기반의 온라인 마케팅 시스템 및 방법
CN107025297A (zh) * 2017-04-18 2017-08-08 上海斐讯数据通信技术有限公司 一种聊天机器人及其自动聊天方法
CN107844992A (zh) * 2017-10-31 2018-03-27 广东欧珀移动通信有限公司 评论信息处理方法、装置、终端设备及存储介质
CN108053230A (zh) * 2017-11-29 2018-05-18 无锡雅座在线科技股份有限公司 舆情管理方法及装置
CN108388608B (zh) * 2018-02-06 2020-08-04 金蝶软件(中国)有限公司 基于文本感知的情感反馈方法、装置、计算机设备和存储介质

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106021391A (zh) * 2016-05-11 2016-10-12 广东工业大学 基于Storm的产品评论信息实时采集方法
CN107943787A (zh) * 2017-11-16 2018-04-20 北京百度网讯科技有限公司 收集用户反馈的方法、装置、设备及计算机可读介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114417178A (zh) * 2021-12-22 2022-04-29 北京三快在线科技有限公司 一种信息处理方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN109388748A (zh) 2019-02-26

Similar Documents

Publication Publication Date Title
WO2020062672A1 (zh) 一种评论信息的回复方法、存储介质、服务器和装置
US11843622B1 (en) Providing machine learning models for classifying domain names for malware detection
EP4091110B1 (en) Systems and methods for distributed incident classification and routing
AU2013261007B2 (en) System and method for creating structured event objects
CN111881221B (zh) 物流服务里客户画像的方法、装置和设备
CN109963259B (zh) 一种短信发送方法及装置
CN104836701A (zh) 订单监控方法及装置
US20180322170A1 (en) Event processing system
US10331752B2 (en) Methods and systems for determining query date ranges
US20200218699A1 (en) Systems and computer implemented methods for semantic data compression
US12223709B2 (en) Methods for more effectively moderating one or more images and devices thereof
US20110258177A1 (en) Systems and methods for providing a microdocument framework for storage, retrieval, and aggregation
CN109155747B (zh) 用于管理多个消息的系统、方法和机器可读介质
US11907227B1 (en) System and method for changepoint detection in streaming data
CN111538616B (zh) 异常定位方法、装置、系统与计算机可读存储介质
CN103577251A (zh) 基于事件的互联网计算处理系统及方法
CN112035534A (zh) 一种实时大数据处理方法、装置及电子设备
CN111666298A (zh) 基于flink的用户服务类别检测方法、装置、计算机设备
CN104992318B (zh) 行事历主动推荐事件的方法
WO2015084756A1 (en) Event detection through text analysis using trained event template models
US20220345473A1 (en) Anomaly detection in computing computing system events
CN107291938A (zh) 订单查询系统及方法
JP2010286868A (ja) コミュニティ形成システム、そのコミュニティ形成装置、そのデータ処理方法およびコンピュータプログラム
CN113808589A (zh) 坐席通话方法、计算机设备和存储介质
CN110019269A (zh) 数据校验的方法、系统及终端设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18934749

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 08/07/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18934749

Country of ref document: EP

Kind code of ref document: A1