WO2023015715A1 - 基于用户评论数据处理方法、装置、设备及存储介质 - Google Patents

基于用户评论数据处理方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2023015715A1
WO2023015715A1 PCT/CN2021/123925 CN2021123925W WO2023015715A1 WO 2023015715 A1 WO2023015715 A1 WO 2023015715A1 CN 2021123925 W CN2021123925 W CN 2021123925W WO 2023015715 A1 WO2023015715 A1 WO 2023015715A1
Authority
WO
WIPO (PCT)
Prior art keywords
product
crawling
sales
user
comprehensive evaluation
Prior art date
Application number
PCT/CN2021/123925
Other languages
English (en)
French (fr)
Inventor
李曦杰
王建
陈思贤
刘雁鹏
陈冰
陈小锋
Original Assignee
惠州Tcl云创科技有限公司
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 惠州Tcl云创科技有限公司 filed Critical 惠州Tcl云创科技有限公司
Publication of WO2023015715A1 publication Critical patent/WO2023015715A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR 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
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR 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
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0282Rating or review of business operators or products

Definitions

  • the present disclosure relates to the field of data processing, and in particular to a method, device, device, and storage medium for data processing based on user comments.
  • the technical problem to be solved by the present disclosure is to provide a method, device, terminal device and storage medium based on user comment data processing for the above-mentioned defects of the prior art. It is a question of extracting and analyzing the text evaluation content of all users, and making statistics on the analysis results, and making a data chart that can truly reflect the advantages and disadvantages of all aspects of the product.
  • the first aspect of the present disclosure relates to a method for processing data based on user comments, wherein the method includes:
  • semantic analysis is performed on the obtained after-sales review information, and comprehensive evaluation feedback corresponding to the product is obtained;
  • the step of detecting crawling instructions and controlling the acquisition of after-sale comment information of products on various e-commerce platforms includes:
  • the control crawls the after-sales review information of the products in each e-commerce platform through a web crawler customized and developed for each e-commerce platform.
  • the step of performing semantic analysis on the obtained after-sales comment information based on the keywords of preset favorable ratings, and obtaining comprehensive evaluation feedback corresponding to the product includes:
  • the step of combining preset keywords and semantic analysis results to obtain comprehensive evaluation feedback corresponding to the product includes:
  • the steps of establishing and outputting a favorable rating analysis chart corresponding to the product include:
  • the favorable rating analysis chart includes at least one chart item describing product functions and/or features, and the favorable rating of each chart item.
  • the second aspect of the present disclosure relates to a method for processing data based on user comments, wherein the method includes:
  • semantic analysis is performed on the obtained after-sales review information, and comprehensive evaluation feedback corresponding to the product is obtained;
  • the step of detecting crawling instructions and controlling the acquisition of after-sale comment information of products on various e-commerce platforms includes:
  • the control crawls the after-sales review information of the products in each e-commerce platform through the web crawler customized and developed for each e-commerce platform.
  • the after-sale comment information includes: text, pictures, and platform favorable rating options.
  • the favorable rating option of the platform is a button that pops up when the user confirms the receipt of the goods and is used to select the degree of satisfaction of the goods or services.
  • the described processing method based on user comment data, wherein, before the step of detecting the crawling instruction and controlling the acquisition of the after-sale comment information of the product on each e-commerce platform, the step further includes:
  • the web crawler is pre-customized and developed for crawling after-sales comment information for each e-commerce platform, and the mode of setting the web crawler to trigger data crawling includes timing and automatic
  • the steps of crawling and manual crawling also include:
  • the crawling mode selection function When the crawling mode selection function is set to manual crawling, when the user manually presses the switch of the after-sales review information crawling function, the web crawler corresponding to the e-commerce platform starts to crawl the user's information in the product link. After-sales review information; when the crawling mode selection function is set to timing automatic crawling, according to the preset crawling time interval, the web crawler corresponding to the e-commerce platform starts to crawl the after-sales review information of users in the product connection.
  • the method for processing data based on user comments wherein, before the step of detecting crawling instructions and controlling the acquisition of after-sales comment information of products on each e-commerce platform, the step includes:
  • Pre-set keywords used to reflect product favorable ratings from the perspective of product functions and/or features are preferred.
  • the keywords include words that reflect product functions or features and words that indicate whether product functions or features are well received.
  • the step of performing semantic analysis on the obtained after-sales comment information based on the keywords of preset favorable ratings, and obtaining comprehensive evaluation feedback corresponding to the product includes:
  • the semantic analysis is the analysis and extraction of the meaning of the user's language by using natural language processing related technologies.
  • the step of combining preset keywords and semantic analysis results to obtain comprehensive evaluation feedback corresponding to the product includes:
  • the steps of establishing and outputting a favorable rating analysis chart corresponding to the product include:
  • the favorable rating analysis chart includes at least one chart item describing product functions and/or features, and the favorable rating of each chart item.
  • a third aspect of the present disclosure relates to a device for processing data based on user comments, wherein the device includes:
  • the preset module is used to pre-set keywords used to reflect product praise from the perspective of product functions and/or features;
  • the acquisition module is used to crawl the after-sales review information of products in each e-commerce platform through a web crawler customized and developed for each e-commerce platform;
  • the analysis module is used to perform semantic analysis on the obtained after-sales review information through the preset keywords of praise, and obtain comprehensive evaluation feedback corresponding to the product;
  • the chart creation module is used to create and output a favorable rating analysis chart corresponding to the product according to the comprehensive evaluation feedback corresponding to the product.
  • a fourth aspect of the present disclosure relates to a terminal device, wherein the terminal device includes a memory, a processor, and a user comment-based data processing program stored on the memory and operable on the processor, the processing When the computer executes the user comment-based data processing program, the steps of any one of the user-comment-based data processing methods are implemented.
  • the fifth aspect of the present disclosure relates to a computer-readable storage medium, wherein a user comment-based data processing program is stored thereon, and when the user comment-based data processing program is executed by a processor, any one of the above-mentioned The steps of the user review data processing method.
  • the present disclosure provides a data processing method based on user comments.
  • the method adopts: detecting crawling instructions, controlling and obtaining after-sales comment information of products on various e-commerce platforms; According to the keywords of favorable comments, semantic analysis is carried out on the obtained after-sales review information, and the comprehensive evaluation feedback corresponding to the product is obtained; according to the comprehensive evaluation feedback corresponding to the product, the favorable analysis chart corresponding to the product is established and output.
  • the after-sales review data of the product on the e-commerce platform can be automatically crawled, and a praise rating analysis chart can be established based on the after-sales review data of the product for the manufacturer to analyze the advantages and disadvantages of the product, so as to achieve the effect of efficiently obtaining and analyzing the advantages and disadvantages of the product.
  • Users who want to buy the product provide real and intuitive product evaluation data charts, as well as provide real feedback from users efficiently for manufacturers who produce products, and provide effective reference for subsequent product development.
  • Fig. 1 is a flowchart of a specific implementation of a method for processing data based on user comments provided by an embodiment of the present disclosure.
  • Fig. 2 is a bar chart of product features and favorable ratings of a thermos cup provided by an embodiment of the present disclosure.
  • Fig. 3 is a comparison chart of user comment data collected by manual and this method provided by an embodiment of the present disclosure.
  • Fig. 4 is a comparison chart of processing user comment data by manual and this method provided by the embodiment of the present disclosure.
  • Fig. 5 is a schematic flowchart of a method for processing data based on user comments provided by a further embodiment of the present disclosure.
  • Fig. 6 is a functional block diagram of an apparatus for processing data based on user comments provided by an embodiment of the present disclosure.
  • Fig. 7 is a schematic diagram of an internal structure of a terminal device provided by an embodiment of the present disclosure.
  • an embodiment of the present disclosure provides a user comment-based data processing method.
  • the after-sale comment information of products on various e-commerce platforms can be automatically crawled, and according to the After-sales review information is automatically analyzed and a favorable rating analysis chart is established. It solves the problem of inefficiency when manufacturers use user feedback from e-commerce platforms, achieves efficient and accurate product feedback data analysis and modeling, and provides effective intelligence support for manufacturers in further product development and strategy making.
  • an embodiment of the present disclosure provides a method for processing data based on user comments.
  • the method described in the embodiment of the present disclosure includes the following steps:
  • Step S100 detecting the crawling instruction, and controlling to obtain the after-sales review information of the product on each e-commerce platform;
  • the control acquires after-sales review information of the product on each e-commerce platform, including text, pictures, and favorable rating options of the platform.
  • the favorable rating option is a button popped up by the user for selecting the degree of satisfaction of the commodity or service when confirming the receipt of the goods.
  • control crawls after-sales review information of products in each e-commerce platform through a web crawler developed for each e-commerce platform.
  • a web crawler is customized and developed for each e-commerce platform or commonly used e-commerce platform in advance, and the mode of triggering data crawling by the web crawler is set to include timing automatic crawling and manual crawling.
  • Taobao and JD.com are the most popular e-commerce platforms used by Chinese users.
  • manufacturers need to obtain product feedback, they will pre-create a Taobao web crawler for crawling user after-sales comment information on Taobao products, and a pre-made Taobao crawler for Crawling Jingdong web crawler used to crawl user after-sales review information of products on Jingdong.
  • thermos cup For example, a certain manufacturer A puts a thermos cup on the Taobao store. After advertising, the monthly sales volume of the thermos cup can reach 5,000. After three months, the manufacturer plans to develop and produce the second thermos cup. At this time, it is necessary to obtain and analyze the user feedback of the first thermos cup.
  • the manufacturer crawls the after-sales review information of the water cup on the Taobao platform through a web crawler developed for the Taobao platform. There are only 3,000 pieces of after-sales review information crawled by the crawler that can reflect users' favorable comments on the vacuum flask.
  • the crawling mode selection function is set.
  • the crawling mode selection function is set to manual crawling, only when the user manually presses the switch of the after-sales review information crawling function , the web crawler corresponding to the e-commerce platform starts to crawl the user's after-sales comment information in the product link; when the crawling mode selection function is set to automatic crawling at regular intervals, according to the preset crawling time interval, for example, the data is fetched once a day, then When the timer is full for one hour, the web crawler corresponding to the e-commerce platform starts to crawl the after-sales review information of users in the product link.
  • step S200 perform semantic analysis on the obtained after-sales review information based on the preset keywords of favorable ratings, and obtain comprehensive evaluation feedback corresponding to the product;
  • semantic analysis is carried out in combination with after-sales review information to obtain the comprehensive evaluation feedback of the product on the e-commerce platform.
  • the semantic analysis is the analysis and extraction of the meaning of the user's language using natural language processing related technologies, such as the replacement of the most basic words, words such as uncomfortable, disgusting, bad, etc. can be classified into words such as dislike, beautiful, satisfied, happy, etc. It can be classified as likes, and combined with the preset keywords of praise, by converting the user's after-sales review information into words or terms similar to the keywords of praise, the feedback data can be more accurate and more accurate evaluations can be provided for manufacturers Data enables manufacturers to design according to user preferences when developing and producing products.
  • keywords used to reflect product praise from the perspective of product functions and/or features are preset.
  • the keywords include vocabulary such as battery life, comfort, screen, performance, etc. used to reflect product functions or features, as well as words including good, perfect, like, bad, etc. used to reflect whether a certain function or feature of the product is well received.
  • the step of performing semantic analysis on the obtained after-sales review information based on the keywords of the preset favorable rating, and obtaining the comprehensive evaluation feedback corresponding to the product includes:
  • the step of obtaining the comprehensive evaluation feedback corresponding to the product by combining preset keywords and semantic analysis results includes:
  • manufacturer A crawls 3,000 pieces of valid after-sales review information from the Taobao platform, and processes the 3,000 pieces of information one by one in combination with preset keywords and semantic analysis.
  • one of the messages is "Baby is very beautiful, the child likes it very much”
  • the product feature of the after-sales review information is the appearance, and the user shows that he likes it, and then find the product appearance and praise by comparing the preset keywords
  • Two key words; one of the longer after-sales comments is "The order placed last week arrived today. After opening, the water bottle smells very bad, and the heat preservation effect is not good. The customer service said a few words and ignored me.
  • thermos cup Use the contacts of users who bought the first thermos cup because they liked the appearance of the product to promote the new thermos cup to the outside world, or further develop products with better heat preservation effect because of the poor heat preservation effect. All aspects of the thermos cup are improved to the extent that users like it.
  • step S300 according to the comprehensive evaluation feedback corresponding to the product, an analysis chart of favorable ratings corresponding to the product is established and output.
  • the comprehensive evaluation feedback is converted and a favorable rating analysis chart corresponding to the product is established.
  • the favorable rating analysis chart includes at least one chart item describing product features and/or characteristics, and the favorable rating of the product feature chart item, and the form of the chart includes but is not limited to a histogram, a pie chart, a line chart , three-dimensional data map, etc.
  • the comprehensive evaluation feedback of the thermos cup of the existing manufacturer A its product feature items include product appearance, thermal insulation effect, product size, logistics and after-sales service, and its scores are 4.54, 4.24, 4.35, 4.44, 4.12 points
  • the abscissa includes five items including product appearance, heat preservation effect, product size, logistics and after-sales service, and the ordinate represents the degree of praise rating.
  • the method for processing data based on user comments in this specific application embodiment includes the following steps:
  • Step S10 start, enter step S11;
  • Step S11 configure the basic information of the crawler task in advance, the basic information includes (product, cycle), enter step S12;
  • Step S12 configure the website URL, that is, the network address, and enter step S13;
  • Step S13 configure the early warning mailbox, and enter step S14;
  • Step S14 start and run the crawler task, and enter step S15;
  • Step S15 crawl user comment information, enter step S16;
  • Step S16 analyze the user's comment information through natural language processing related technologies, and enter step S17;
  • Step S17 perform statistics on the crawled and analyzed user comment information data and draw a chart, and enter step S18;
  • Step S18 automatically send an early warning email to the preset early warning email address for the low-scoring item, and enter step S20;
  • the user can automatically crawl the user comment information of the products on the e-commerce platform through this method and automatically draw it into an intuitive data table, and automatically send an early warning email to
  • the preset early warning mailbox reminds the person in charge to deal with it, which improves the efficiency of manufacturers in obtaining real comments and feedback from users, and the automatic email reminder function further ensures the effect of finding and solving product deficiencies.
  • pre-configure the crawler task for crawling data such as configuring basic information such as product name, crawling cycle, and feedback mailbox, so that the manufacturer can accurately know that the crawler is used for crawling when using the crawler task.
  • Which product of which e-commerce platform to fetch, and how often each crawl cycle is, such as once a week, once a day, and once a month. Further configure the corresponding website URL and the link address of the user comment data of the corresponding product to be crawled, and configure the early warning mailbox to notify the relevant person in charge by email in a timely manner when a certain data is too low or the crawled data is abnormal , For example, when it is found that the overall user evaluation has become low, it may be considered that a competitor is maliciously posting bad reviews. When one of the scores is too low, it may also be a defect in the product itself, and the product needs to be adjusted in time .
  • the crawler task is started, and the crawler crawls the user comment information of the preset website, and analyzes the user comment information through natural language processing related technologies to obtain each valid user.
  • the meaning of the comment For example, when the crawled product data is a water pen, extract the meaning that the user wants to express, such as good-looking appearance, smooth writing, and slight smell, etc., perform data statistics on the obtained evaluation and draw it as an intuitive chart, and automatically identify it For all projects with lower scores or significantly lower scores than the last crawled data, the corresponding person in charge will be notified by preset email to remind them to further find out the reason and solve it.
  • manufacturers can not only efficiently obtain and analyze user comments, but also judge whether there are malicious bad reviews or serious defects in products by identifying user evaluation information, and timely send emails to comrades.
  • the person in charge enables manufacturers to avoid risks in a timely manner.
  • an embodiment of the present disclosure provides a data processing device based on user comments, which includes: a preset module 610 , an acquisition module 620 , an analysis module 630 , and a chart creation module 640 .
  • the preset module 610 is used to pre-set keywords used to reflect product praise from the perspective of product functions and/or features;
  • the acquisition module 620 is used to customize and develop keywords for each e-commerce platform
  • the web crawler crawls the after-sales review information of products in various e-commerce platforms;
  • the analysis module 630 is used to perform semantic analysis on the obtained after-sales review information through preset keywords of favorable ratings, and obtain comprehensive information corresponding to the product.
  • Evaluation feedback the chart creation module 640 is used to create and output a favorable rating analysis chart corresponding to the product according to the comprehensive evaluation feedback corresponding to the product.
  • the present disclosure further provides a terminal device, the functional block diagram of which may be shown in FIG. 7 .
  • the terminal equipment includes a processor, a memory, a network interface, and a display screen connected through a system bus.
  • the processor of the terminal device is used to provide calculation and control capabilities.
  • the memory of the terminal device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system and computer programs.
  • the internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium.
  • the network interface of the terminal device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, data processing based on user comments is realized.
  • the display screen of the terminal device may be a liquid crystal display screen or an electronic ink display screen.
  • FIG. 7 is only a block diagram of a partial structure related to the disclosed solution, and does not constitute a limitation on the terminal equipment to which the disclosed solution is applied.
  • the specific terminal equipment More or fewer components than shown in the figures may be included, or certain components may be combined, or have a different arrangement of components.
  • a terminal device includes a memory, a processor, and a data processing program based on user comments stored on the processor and operable on the processor.
  • the processor performs the following steps:
  • semantic analysis is performed on the obtained after-sales review information, and the comprehensive evaluation feedback corresponding to the product is obtained;
  • the step of detecting the crawling instruction and controlling the acquisition of after-sales comment information of the product on each e-commerce platform includes:
  • the control crawls the after-sales review information of the products in each e-commerce platform through a web crawler customized and developed for each e-commerce platform.
  • the step of detecting the crawling instruction and controlling the acquisition of the after-sales comment information of the product on each e-commerce platform it also includes:
  • step of detecting the crawling instruction and controlling the acquisition of the after-sales comment information of the product on each e-commerce platform includes:
  • Pre-set keywords used to reflect product favorable ratings from the perspective of product functions and/or features are preferred.
  • the step of performing semantic analysis on the obtained after-sales review information based on the keywords of the preset favorable rating, and obtaining the comprehensive evaluation feedback corresponding to the product includes:
  • the step of obtaining the comprehensive evaluation feedback corresponding to the product by combining preset keywords and semantic analysis results includes:
  • the steps of establishing and outputting the favorable rating analysis chart corresponding to the product include:
  • the favorable rating analysis chart includes at least one chart item describing product functions and/or features, and the favorable rating of each chart item.
  • Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
  • SRAM Static RAM
  • DRAM Dynamic RAM
  • SDRAM Synchronous DRAM
  • DDRSDRAM Double Data Rate SDRAM
  • ESDRAM Enhanced SDRAM
  • SLDRAM Synchronous Chain Synchlink DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM
  • the present disclosure discloses a data processing method, device, device, and storage medium based on user comments.
  • the method includes: detecting crawling instructions, controlling and obtaining after-sales comment information of products on various e-commerce platforms; According to the keywords of favorable comments, semantic analysis is carried out on the obtained after-sales review information, and the comprehensive evaluation feedback corresponding to the product is obtained; according to the comprehensive evaluation feedback corresponding to the product, the favorable analysis chart corresponding to the product is established and output. It aims to solve the problem that there is no solution in the prior art that can automatically extract and analyze the text evaluation content of all users in the product link, and make statistics on the analysis results to make a data chart that can truly reflect the advantages and disadvantages of all aspects of the product. Provide buyers with real and intuitive commodity evaluation data, and provide commodity manufacturers with efficient and accurate user feedback data.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本公开涉及基于用户评论数据处理方法、装置、设备及存储介质,所述方法包括:检测到爬取指令,控制获取产品在各电商平台的售后评论信息;基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。旨在解决现有技术中没有能够自动将商品链接中全部用户的文字评价内容提取和分析,并将其分析结果进行统计,制作为能够真实体现商品各方面优缺点的数据图表的方案的问题。为买家提供真实、直观的商品评价数据,为商品厂家提供高效、准确的用户反馈数据。

Description

基于用户评论数据处理方法、装置、设备及存储介质
优先权
所述PCT专利申请要求申请日为2021年8月12日,申请号为202110922557.0的中国专利优先权,本专利申请结合了上述专利的技术方案。
技术领域
本公开涉及数据处理领域,尤其涉及的是基于用户评论数据处理方法、装置、设备及存储介质。
背景技术
随着互联网的不断发展,用户越来越倾向于通过互联网购买商品,相应的各个渠道的用户反馈数量也在快速增长,而目前最常见的用户反馈渠道为电商平台的用户评价,但通过平台本身的评价体系例如好评率或差评个数等参数,无法获得买家的真实评价,而买家的真实评价往往在于文字评价部分。
现有技术中没有能够自动将商品链接中全部用户的文字评价内容提取和分析,并将其分析结果进行统计,制作为能够真实体现商品各方面优缺点的数据图表的方案。无法为准备购买商品的用户提供真实、直观的评价数据,也无法为商品厂家提供真实、迅速的用户反馈,为之后的研发策略做参考。
因此,现有技术还有待改进和发展。
公开内容
本公开要解决的技术问题在于,针对现有技术的上述缺陷,提供一种基于用户评论数据处理方法、装置、终端设备及存储介质,本公开解决了现有技术中缺少能够自动将商品链接中全部用户的文字评价内容提取和分析,并将其分析结果进行统计,制作为能够真实体现商品各方面优缺点的数据图表的方案的问题。
为了解决上述技术问题,本公开采用的技术方案如下:
本公开第一方面涉及一种基于用户评论数据处理方法,其中,所述方法包括:
预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取;
预先设置用于从产品功能和/或特点角度体现产品好评度的关键词;
检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
所述的基于用户评论数据处理方法,其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤包括:
当检测到数据爬取指令,控制通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息。
所述的基于用户评论数据处理方法,其中,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
通过语言处理算法对爬取的售后评论信息进行语义分析;
结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
所述的基于用户评论数据处理方法,其中,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
将至少一个好评度整理为产品对应的综合评价反馈。
所述的基于用户评论数据处理方法,其中,所述根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出的步骤包括:
根据得到的综合评价反馈,建立与产品对应的好评度分析图表;
所述好评度分析图表包括至少一项描述产品功能和/或特点的图表项,以及各图表项的好评度。
本公开第二方面涉及一种基于用户评论数据处理方法,其中,所述方法包括:
检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
所述的基于用户评论数据处理方法,其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤包括:
当检测到数据爬取指令,控制通过对每个电商平台定制开发的网络爬虫,爬取各个 电商平台中产品的售后评论信息。
所述的基于用户评论数据处理方法,其中,所述售后评论信息包括:文字,图片以及平台的好评度选项。
所述的基于用户评论数据处理方法,其中,所述平台的好评度选项为用户在确认收货时弹出的用于选择商品或服务满意程度的按钮。
所述的基于用户评论数据处理方法,其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前还包括:
预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取。
所述的基于用户评论数据处理方法,其中,所述预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取的步骤之前还包括:
设置爬取模式选择功能,当设置爬取模式选择功能为手动爬取时,当用户手动按下售后评论信息爬取功能的开关时,对应电商平台的网络爬虫开始爬取产品链接中用户的售后评论信息;当设置爬取模式选择选择功能为定时自动爬取时,根据预设的爬取时间间隔,对应电商平台的网络爬虫开始爬取产品连接中用户的售后评论信息。
所述的基于用户评论数据处理方法,其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前包括:
预先设置用于从产品功能和/或特点角度体现产品好评度的关键词。
所述的基于用户评论数据处理方法,其中,所述关键词包括体现产品功能或特点的词汇和体现产品功能或特点是否受到好评的词汇。
所述的基于用户评论数据处理方法,其中,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
通过语言处理算法对爬取的售后评论信息进行语义分析;
结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
所述的基于用户评论数据处理方法,其中,所述语义分析为利用自然语言处理相关技术对用户语言含义的分析以及提取。
所述的基于用户评论数据处理方法,其中,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
将至少一个好评度整理为产品对应的综合评价反馈。
所述的基于用户评论数据处理方法,其中,所述根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出的步骤包括:
根据得到的综合评价反馈,建立与产品对应的好评度分析图表;
所述好评度分析图表包括至少一项描述产品功能和/或特点的图表项,以及各图表项的好评度。
本公开的第三方面涉及一种基于用户评论数据处理装置,其中,所述装置包括:
预设模块,用于预先设置用于从产品功能和/或特点角度体现产品好评度的关键词;
获取模块,用于通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息;
分析模块,用于通过预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
图表创建模块,用于根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
本公开的第四方面涉及一种终端设备,其中,所述终端设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的基于用户评论数据处理程序,所述处理器执行所述基于用户评论数据处理程序时,实现任一项所述的基于用户评论数据处理方法的步骤。
本公开的第五方面涉及一种计算机可读存储介质,其中,其上存储有基于用户评论数据处理程序,所述基于用户评论数据处理程序被处理器执行时,实现任一项所述的基于用户评论数据处理方法的步骤。
有益效果:与现有技术相比,本公开提供了一种基于用户评论数据处理方法,所述方法采用:检测到爬取指令,控制获取产品在各电商平台的售后评论信息;基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。通过本方法可自动爬取产品在电商平台的售后评论数据,并根据产品的售后评论数据建立好评度分析图表供厂商分析产品的优缺点,达到高效获取并分析产品优缺点的效果,为想要购买该商品的用户提供真实、直观的商品评价数据图表,以及为生产商品的厂家高效率的提供用户的真实反馈,为后续产品的开发做有效参考。
附图说明
为了更清楚地说明本公开实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本公开中记载的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本公开实施例提供的基于用户评论数据处理方法的具体实施方式的流程图。
图2是本公开实施例提供的保温杯的产品特征与好评度的柱图。
图3是本公开实施例提供的利用人工与本方法收集用户评论数据的对比图。
图4是本公开实施例提供的利用人工与本方法处理用户评论数据的对比图。
图5是本公开进一步的实施例提供的基于用户评论数据处理方法的流程示意图。
图6是本公开实施例提供的基于用户评论数据处理装置的原理框图。
图7是本公开实施例提供的终端设备的内部结构原理图。
具体实施方式
为使本公开的目的、技术方案及优点更加清楚、明确,以下参照附图并举实施例对本公开进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本公开,并不用于限定本公开。
需要说明,若本公开实施例中有涉及方向性指示(诸如上、下、左、右、前、后……),则该方向性指示仅用于解释在某一特定姿态(如附图所示)下各部件之间的相对位置关系、运动情况等,如果该特定姿态发生改变时,则该方向性指示也相应地随之改变。
另外,若本公开实施例中有涉及“第一”、“第二”等的描述,则该“第一”、“第二”等的描述仅用于描述目的,而不能理解为指示或暗示其相对重要性或者隐含指明所指示的技术特征的数量。由此,限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。另外,各个实施例之间的技术方案可以相互结合,但是必须是以本领域普通技术人员能够实现为基础,当技术方案的结合出现相互矛盾或无法实现时应当认为这种技术方案的结合不存在,也不在本公开要求的保护范围之内。
随着互联网的不断发展,用户对于产品的发声渠道不断增加,相应的用户反馈数量也在快速增长。如何能够高效利用用户在电商平台的反馈并挖掘出产品的各维度信息成为了各大厂商重点关注的问题。目前,最常见的用户反馈渠道为电商平台的用户评论,部分网站根据用户的正负面评价将反馈划分到不同类别,另一些网站会根据用户关注较 多的产品特征单独划分出新的类别来囊括相关反馈。但是,各厂商在利用这类渠道的用户评论数据时往往比较低效,因为需要将所有的用户反馈进行逐条分析,以获取用户对于产品的关注点、痛点及其他维度信息,而且每当有一批新的数据进来,都要做这种重复性的工作,这种低效性对于更新迭代较快的产品影响更为严重,而一个高效的产品用户反馈分析能够给产品带来很多先发优势。
为了解决上述问题,本公开实施例提供一种基于用户评论数据处理方法,根据本实施例的基于用户评论数据处理方法,可自动爬取产品在各电商平台的售后评论信息,并根据所述售后评论信息自动分析并建立好评度分析图表。解决了厂商在利用电商平台的用户反馈时效率低下的问题,达到高效、精准的产品反馈数据分析与建模,为厂商在进一步研发产品以及做策略时提供了有效的情报支撑。
示例性方法
如图1中所示,本公开实施例提供一种基于用户评论数据处理方法。在本公开实施例中所述方法包括如下步骤:
步骤S100、检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
在本实施例中,当检测到由定时自动开启或手动开启的爬取指令时,控制获取产品在各电商平台的售后评论信息,包括文字、图片以及平台的好评度选项。所述好评度选项为用户在确认收货时弹出的用于选择商品或服务满意程度的按钮。
具体地,当检测到数据爬取指令,控制通过对每个电商平台开发的网络爬虫,爬取各个电商平台中产品的售后评论信息。
其中,预先针对每个电商平台或常用电商平台分别定制开发一个网络爬虫,并设置该网络爬虫触发数据爬取的方式包括定时自动爬去以及手动爬取。
例如中国用户使用的较多的电商平台为淘宝以及京东,则在厂商需要获取产品的反馈情况时,预先制作用于爬取淘宝产品的用户售后评论信息的淘宝网络爬虫,以及预先制作用于爬取用于爬取京东上产品的用户售后评论信息的京东网络爬虫。
举例说明,某厂商A将生产的一款保温杯上架至了淘宝店铺,经过宣传该保温杯的月销量能达到五千个,而过了三个月后该厂商准备研发并生产第二款保温杯,此时就需要对第一款保温杯的用户反馈进行获取和分析。厂商通过针对淘宝平台开发的网络爬虫对淘宝平台该水杯的售后评论信息进行爬取,例如爬取到15000用户的售后评论信息,固定去掉“该用户未评论”的售后评论信息后,实际通过网络爬虫爬取到的可以体现用户对保温杯好评度的售后评论信息仅剩下3000条。
进一步地,为了提高和控制爬取信息的自由度,设置爬取模式选择功能,当设置爬取模式选择功能为手动爬取时,仅有当用户手动按下售后评论信息爬取功能的开关时,对应电商平台的网络爬虫开始爬取产品链接中用户的售后评论信息;当设置爬取模式选择选择功能为定时自动爬取时,根据预设的爬取时间间隔例如数据一天取一次,则当定时器记满一个小时后,对应电商平台的网络爬虫开始爬取产品连接中用户的售后评论信息。
进一步地,步骤S200、基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
在本实施例中,根据预设的用户对产品某方面功能或特点进行好评度评价的关键词,结合售后评论信息进行语义分析,得到产品在该电商平台中的综合评价反馈。所述语义分析为利用自然语言处理相关技术对用户语言含义的分析以及提取,例如最基本的词汇的替换,难受、讨厌、很烂、等词汇可归为不喜欢,漂亮、满意、开心等词汇可归为喜欢,并且结合预设的好评度的关键词,通过将用户的售后评论信息转换为与好评度关键词相近的词语或用语,可使反馈数据更加准确,为厂商提供更加精准的评价数据,使厂商在研发和制作产品时能够针对性的抓住用户的喜好进行设计。
其中,预先设置用于从产品功能和/或特点角度体现产品好评度的关键词。所述关键词包括续航、舒适度、屏幕、性能等用于体现产品功能或特点的词汇,以及包括好、完美、喜欢、不好等用于体现产品某功能或特点是否受到好评的词汇。
具体地,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
通过语言处理算法对爬取的售后评论信息进行语义分析;
结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
其中,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
将至少一个好评度整理为产品对应的综合评价反馈。
举例说明,厂商A通过在淘宝平台爬取到3000条有效的售后评论信息,并且结合预设的关键词与语义分析对3000条信息逐条的进行处理。例如其中一条信息为“宝贝很漂亮,孩子很喜欢”,则通过语义分析得到该售后评论信息评论的产品特征为外观,且用户表现为喜欢,再通过对比预设的关键词找到产品外观以及好评两个关键词;其中一条 较长的售后评论信息为“上上周下的单今天才到,打开之后水瓶异味很严重,保温效果也不好,客服说两句就不理人了,差评!”则通过语义分析读取用户的售后评价信息可以得到物流慢、有气味、保温差、服务态度差等四个评价,则结合预设的关键词找到该售后评论信息中包含的四个产品特征分别为、物流、异味、保温效果以及售后服务,以及其评价都为差;以及还有一条售后评论信息为“宝贝很好,用来焖小米和鸡蛋很好吃”,则通过语义分析以及结合预设的关键词也可得到评价的产品特征为保温效果,以及其评价为好。
当所有的售后评论信息都通过语义分析后的得到了若干组产品特征以及其评价的评论信息,例如产品外观中评价好占了80%,一般占了17%,差占了3%,保温效果中评价保温占了70%,一般占了22%,差占了8%。则如果分别为好、一般、差赋予五分、三分、一分的分数值,并分别计算每个产品特征的得分,得到产品外观得分为4.54分,保温效果为4.24分,将得到的分数进行整理则得到该保温杯的综合评价反馈,厂商在下一次研发保温杯时可以根据该综合评价反馈进行决策,例如产品外观得分最高,则下一款保温杯的产品开发策略为制作多款好看的、符合男女老少审美的保温,利用第一款保温杯中因为喜欢产品外观而购买的用户的人脉将新款保温杯向外推广,或因为保温效果较差则进一步研发保温效果更好的产品,将保温杯的各方面都提高到能够使用户喜欢程度。
进一步地,步骤S300、根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
在本实施例中,为进一步得到更加直观的产品数据,使厂商了解到产品的受众点、缺点等产品信息,将所述综合评价反馈进行转换并建立产品对应的好评度分析图表。
所述好评度分析图表中包括至少一项描产品特征和/或特点的图表项,以及该产品特征图表项的好评度,且所述图表的形式包括但不限于柱状图、饼图、折线图、三维立体数据图等。
举例说明,如图2所示,现有厂商A的保温杯的综合评价反馈,其产品特征项数包括产品外观、保温效果、产品尺寸、物流以及售后服务,且其评分分别为4.54、4.24、4.35、4.44、4.12分,则当该保温杯的综合评价反馈建立成柱状图时横坐标分别有包括产品外观、保温效果、产品尺寸、物流以及售后服务的五项,纵坐标代表各项好评度的评分。
如图3以及图4所示,通过实际对比本公开方法以及人工收集处理信息的时间成本、沟通成本、工作效率以及局限性,得到结果为通过本公开的基于用户评论数据处理方法 比起人工收集处理信息的方法效率提高了二十倍,而成本也大大的降低了,并且人工处理则有可能提前也有可能延后,具有不确定性,而通过设备收集与处理的办法可以计算得到准确地完成时间。
以下通过一具体应用实施例对本公开方法做进一步详细说明:
如图5所示,本具体应用实施例的基于用户评论数据处理方法,包括如下步骤:
步骤S10、开始,进入步骤S11;
步骤S11、预先对爬虫任务基本信息进行配置,所述基本信息包括(产品,周期),进入步骤S12;
步骤S12、配置网站URL,即网络地址,进入步骤S13;
步骤S13、配置预警邮箱,进入步骤S14;
步骤S14、爬虫任务开启运行,进入步骤S15;
步骤S15、爬取用户评论信息,进入步骤S16;
步骤S16、通过自然语言处理相关技术分析用户的评论信息,进入步骤S17;
步骤S17、将爬取以及分析的用户评论信息的数据进行统计并绘制图表,进入步骤S18;
步骤S18、将低分项目通过邮件自动发送预警邮件到预设的预警邮箱,进入步骤S20;
步骤S20、结束。
由上可见,在本公开具体应用实施例中,用户通过本方法实现了自动爬取电商平台商品的用户评论信息并自动绘制为直观的数据表格,并且识别低分项目自动的发送预警邮件到预设的预警邮箱提醒对应负责人进行处理,提升了厂商在获取用户真实的评论反馈时的效率,并且通过邮件自动提醒功能进一步确保了寻找产品不足并加以解决的效果。如图5所示,预先对用于爬取数据的爬虫任务进行配置,例如配置产品名、爬取周期以及反馈邮箱等基本信息,使厂商在使用爬虫任务时能准确知道该爬虫是用于爬取哪个电商平台的哪个产品,以及每次爬取的周期为多久一次,例如一周一次、一天一次以及一个月一次。进一步配置对应的网站URL及所要爬取的对应产品的用户评论数据的链接地址,以及配置预警邮箱,用于当某项数据过低或爬取的数据出现异常时及时的通过邮件通知相关负责人,例如当发现用户评价整体变低时,可认为有可能是竞争对手在恶意刷差评,当其中某项分数过低时也可能是产品本身出现了某项缺陷,需要及时的进行产品的调整。进一步地,当设置完爬虫任务、网站URL以及预警邮箱时,开启爬虫任务,爬虫爬取预设网站的用户评论信息,并通过自然语言处理相关技术分析用户的评论信 息,得到每一个有效的用户评论的意思。例如当爬取的产品数据为水笔时,提取出用户想要表达的意思例如,外形好看、书写流畅、稍有气味等评价,将获取的评价进行数据统计并绘制为直观的图表,并自动识别所有项目中分数较低或较上一次爬取数据分数严重变低的项目,并通过预设邮件通知相应负责人,提醒其进一步查明原因并进行解决。
通过本实施例的基于用户评论数据处理方法,厂商不仅可以高效的获取并分析用户评论,还可通过识别用户评价信息判断是否有恶意差评或产品出现严重缺陷等问题,并及时的通过邮件同志负责人,使厂商能够及时规避风险。
示例性装置
如图6中所示,本公开实施例提供一种基于用户评论数据处理装置,该装置包括:预设模块610、获取模块620、分析模块630、图表创建模块640。具体地,所述预设模块610,用于预先设置用于从产品功能和/或特点角度体现产品好评度的关键词;所述获取模块620,用于通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息;所述分析模块630,用于通过预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;所述图表创建模块640,用于根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
基于上述实施例,本公开还提供了一种终端设备,其原理框图可以如图7所示。该终端设备包括通过系统总线连接的处理器、存储器、网络接口、显示屏。其中,该终端设备的处理器用于提供计算和控制能力。该终端设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机程序。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该终端设备的网络接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种基于用户评论数据处理。该终端设备的显示屏可以是液晶显示屏或者电子墨水显示屏。
本领域技术人员可以理解,图7中示出的原理框图,仅仅是与本公开方案相关的部分结构的框图,并不构成对本公开方案所应用于其上的终端设备的限定,具体的终端设备可以包括比图中所示更多或更少的部件,或者组合某些部件,或者具有不同的部件布置。
在一个实施例中,提供了一种终端设备,终端设备包括存储器、处理器及存储在处理器上并可在处理器上运行的基于用户评论数据处理程序,处理器执行如下步骤:
检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对 应的综合评价反馈;
根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤包括:
当检测到数据爬取指令,控制通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息。
其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前还包括:
预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取。
其中,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前包括:
预先设置用于从产品功能和/或特点角度体现产品好评度的关键词。
其中,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
通过语言处理算法对爬取的售后评论信息进行语义分析;
结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
其中,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
将至少一个好评度整理为产品对应的综合评价反馈。
其中,所述根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出的步骤包括:
根据得到的综合评价反馈,建立与产品对应的好评度分析图表;
所述好评度分析图表包括至少一项描述产品功能和/或特点的图表项,以及各图表项的好评度。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一非易失性计算机可读取存储介质中,该计算机程序在执行时,可包括如上述各方法的实施例的流程。其中,本公开所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引 用,均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限,RAM以多种形式可得,诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
综上所述,本公开公开了基于用户评论数据处理方法、装置、设备及存储介质,所述方法包括:检测到爬取指令,控制获取产品在各电商平台的售后评论信息;基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。旨在解决现有技术中没有能够自动将商品链接中全部用户的文字评价内容提取和分析,并将其分析结果进行统计,制作为能够真实体现商品各方面优缺点的数据图表的方案的问题。为买家提供真实、直观的商品评价数据,为商品厂家提供高效、准确的用户反馈数据。
应当理解的是,本公开公开的应用不限于上述的举例,对本领域普通技术人员来说,可以根据上述说明加以改进或变换,所有这些改进和变换都应属于本公开所附权利要求的保护范围。

Claims (20)

  1. 一种基于用户评论数据处理方法,其特征在于,所述方法包括:
    预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取;
    预先设置用于从产品功能和/或特点角度体现产品好评度的关键词;
    检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
    基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
    根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
  2. 根据权利要求1所述的基于用户评论数据处理方法,其特征在于,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤包括:
    当检测到数据爬取指令,控制通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息。
  3. 根据权利要求1所述的基于用户评论数据处理方法,其特征在于,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
    通过语言处理算法对爬取的售后评论信息进行语义分析;
    结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
  4. 根据权利要求3所述的基于用户评论数据处理方法,其特征在于,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
    结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
    将至少一个好评度整理为产品对应的综合评价反馈。
  5. 根据权利要求4所述的基于用户评论数据处理方法,其特征在于,所述根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出的步骤包括:
    根据得到的综合评价反馈,建立与产品对应的好评度分析图表;
    所述好评度分析图表包括至少一项描述产品功能和/或特点的图表项,以及各图表项的好评度。
  6. 一种基于用户评论数据处理方法,其特征在于,所述方法包括:
    检测到爬取指令,控制获取产品在各电商平台的售后评论信息;
    基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
    根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
  7. 根据权利要求6所述的基于用户评论数据处理方法,其特征在于,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤包括:
    当检测到数据爬取指令,控制通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息。
  8. 根据权利要求7所述的基于用户评论数据处理方法,其特征在于,所述售后评论信息包括:文字,图片以及平台的好评度选项。
  9. 根据权利要求8所述的基于用户评论数据处理方法,其特征在于,所述平台的好评度选项为用户在确认收货时弹出的用于选择商品或服务满意程度的按钮。
  10. 根据权利要求7所述的基于用户评论数据处理方法,其特征在于,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前还包括:
    预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取。
  11. 根据权利要求10所述的基于用户评论数据处理方法,其特征在于,所述预先针对每个电商平台定制开发用于爬取的售后评论信息的网络爬虫,并设置所述网络爬虫触发数据爬取的方式包括定时自动爬取以及手动爬取的步骤之前还包括:
    设置爬取模式选择功能,当设置爬取模式选择功能为手动爬取时,当用户手动按下售后评论信息爬取功能的开关时,对应电商平台的网络爬虫开始爬取产品链接中用户的售后评论信息;当设置爬取模式选择选择功能为定时自动爬取时,根据预设的爬取时间间隔,对应电商平台的网络爬虫开始爬取产品连接中用户的售后评论信息。
  12. 根据权利要求6所述的基于用户评论数据处理方法,其特征在于,所述检测到爬取指令,控制获取产品在各电商平台的售后评论信息的步骤之前包括:
    预先设置用于从产品功能和/或特点角度体现产品好评度的关键词。
  13. 根据权利要求12所述的基于用户评论数据处理方法,其特征在于,所述关键词包括体现产品功能或特点的词汇和体现产品功能或特点是否受到好评的词汇。
  14. 根据权利要求12所述的基于用户评论数据处理方法,其特征在于,所述基于预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈的步骤包括:
    通过语言处理算法对爬取的售后评论信息进行语义分析;
    结合预设的关键词与语义分析结果得到产品对应的综合评价反馈。
  15. 根据权利要求14所述的基于用户评论数据处理方法,其特征在于,所述语义分析为利用自然语言处理相关技术对用户语言含义的分析以及提取。
  16. 根据权利要求14所述的基于用户评论数据处理方法,其特征在于,所述结合预设的关键词与语义分析结果得到产品对应的综合评价反馈的步骤包括:
    结合预设的关键词与语义分析结果得到售后评论信息对应的好评度;
    将至少一个好评度整理为产品对应的综合评价反馈。
  17. 根据权利要求16所述的基于用户评论数据处理方法,其特征在于,所述根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出的步骤包括:
    根据得到的综合评价反馈,建立与产品对应的好评度分析图表;
    所述好评度分析图表包括至少一项描述产品功能和/或特点的图表项,以及各图表项的好评度。
  18. 一种基于用户评论数据处理装置,其特征在于,所述装置包括:
    预设模块,用于预先设置用于从产品功能和/或特点角度体现产品好评度的关键词;
    获取模块,用于通过对每个电商平台定制开发的网络爬虫,爬取各个电商平台中产品的售后评论信息;
    分析模块,用于通过预设的好评度的关键词,对获取的售后评论信息进行语义分析,得到与产品对应的综合评价反馈;
    图表创建模块,用于根据与产品对应的综合评价反馈,建立与产品对应的好评度分析图表并输出。
  19. 一种终端设备,其特征在于,所述终端设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的基于用户评论数据处理程序,所述处理器执行所述基于用户评论数据处理程序时,实现如权利要求6-17任一项所述的基于用户评论数据处理方法的步骤。
  20. 一种计算机可读存储介质,其特征在于,其上存储有基于用户评论数据处理程序,所述基于用户评论数据处理程序被处理器执行时,实现如权利要求6-17任一项所述的基于用户评论数据处理方法的步骤。
PCT/CN2021/123925 2021-08-12 2021-10-14 基于用户评论数据处理方法、装置、设备及存储介质 WO2023015715A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110922557.0A CN113793169A (zh) 2021-08-12 2021-08-12 基于用户评论数据处理方法、装置、设备及存储介质
CN202110922557.0 2021-08-12

Publications (1)

Publication Number Publication Date
WO2023015715A1 true WO2023015715A1 (zh) 2023-02-16

Family

ID=78875936

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/123925 WO2023015715A1 (zh) 2021-08-12 2021-10-14 基于用户评论数据处理方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN113793169A (zh)
WO (1) WO2023015715A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114493687A (zh) * 2022-01-14 2022-05-13 广东好太太智能家居有限公司 一种晾衣设备产品痛点分析方法、装置及可读介质
CN114330340B (zh) * 2022-03-14 2022-07-19 浙江口碑网络技术有限公司 评价信息的处理方法、电子设备及可读存储介质
CN114327688B (zh) * 2022-03-15 2022-07-08 Tcl通讯科技(成都)有限公司 基于微前端的数据处理方法、框架、存储介质及终端设备
CN116467393A (zh) * 2023-04-12 2023-07-21 北京博晓通科技有限公司 数据处理方法、装置、计算机设备及存储介质
CN116127105B (zh) * 2023-04-18 2023-07-04 潍坊学院 一种大数据平台的数据汇集方法及装置
CN116883014A (zh) * 2023-07-12 2023-10-13 深圳科迪新汇信息科技有限公司 基于分布式ai模型的客户满意度测评系统
CN117151797A (zh) * 2023-09-05 2023-12-01 中铁商业保理有限公司 基于综合数据分析的企业信用评估系统
CN116957633B (zh) * 2023-09-19 2023-12-01 武汉创知致合科技有限公司 一种基于智能家居场景的产品设计用户体验评价方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160140634A1 (en) * 2014-11-17 2016-05-19 Institute For Information Industry System, method and non-transitory computer readable medium for e-commerce reputation analysis
CN108230085A (zh) * 2017-11-27 2018-06-29 重庆邮电大学 一种基于用户评论的商品评价系统及方法
CN108446813A (zh) * 2017-12-19 2018-08-24 清华大学 一种电商服务质量综合评价的方法
CN110705286A (zh) * 2019-09-24 2020-01-17 青木数字技术股份有限公司 一种基于评论信息的数据处理方法及装置
CN113010639A (zh) * 2021-02-26 2021-06-22 济南浪潮高新科技投资发展有限公司 一种基于电商平台的商品分析方法及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107527231A (zh) * 2017-07-27 2017-12-29 温州市鹿城区中津先进科技研究院 基于自然语言分析的电商用户满意度评价方法
CN107977798B (zh) * 2017-12-21 2023-09-12 中国计量大学 一种电子商务产品质量的风险评价方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160140634A1 (en) * 2014-11-17 2016-05-19 Institute For Information Industry System, method and non-transitory computer readable medium for e-commerce reputation analysis
CN108230085A (zh) * 2017-11-27 2018-06-29 重庆邮电大学 一种基于用户评论的商品评价系统及方法
CN108446813A (zh) * 2017-12-19 2018-08-24 清华大学 一种电商服务质量综合评价的方法
CN110705286A (zh) * 2019-09-24 2020-01-17 青木数字技术股份有限公司 一种基于评论信息的数据处理方法及装置
CN113010639A (zh) * 2021-02-26 2021-06-22 济南浪潮高新科技投资发展有限公司 一种基于电商平台的商品分析方法及装置

Also Published As

Publication number Publication date
CN113793169A (zh) 2021-12-14

Similar Documents

Publication Publication Date Title
WO2023015715A1 (zh) 基于用户评论数据处理方法、装置、设备及存储介质
US9965791B1 (en) Systems, apparatuses, and methods for extracting inventory from unstructured electronic messages
US7788131B2 (en) Advertising keyword cross-selling
WO2015188699A1 (zh) 推荐项目的方法和装置
CN109416611A (zh) 具有项目选择的数字用户界面
WO2021017306A1 (zh) 根据用户画像的个性化搜索方法、系统、设备及存储介质
US20090287714A1 (en) Method and Apparatus for Community-Based Comparison Shopping Based on Social Bookmarking
JP2005196469A (ja) データ表示サーバ,データ表示方法,およびそのプログラム
CN109949172B (zh) 社交账号影响力评价方法、装置及存储介质
CN112148973A (zh) 一种信息推送的数据处理方法及装置
US20240152972A1 (en) Marketing to consumers using data obtained from abandoned gps searches
JP5523222B2 (ja) 商品情報提供システム、商品情報提供装置、商品情報提供方法及びプログラム
CN109064296A (zh) 一种辅助加餐的方法、装置和系统
CN111680213A (zh) 信息推荐方法、数据处理方法及装置
CN110717095B (zh) 服务项目的推送方法及装置
CN110209944B (zh) 一种股票分析师推荐方法、装置、计算机设备和存储介质
CN112862553A (zh) 一种商品推荐的方法和装置
CN113704630B (zh) 一种信息推送方法、装置、可读存储介质和电子设备
Kim et al. How network externality leads to the success of mobile instant messaging business?
KR20200065754A (ko) 도서 추천 방법 및 이를 지원하는 서비스 장치
JP2020154821A (ja) 記事推奨システム
CN113609375A (zh) 内容推荐方法及装置、存储介质及电子装置
WO2020233586A1 (zh) 食物账单的推送方法、装置和计算机可读存储介质
JP2004362539A (ja) 閲覧コンテンツ変更装置、閲覧コンテンツ変更方法、閲覧コンテンツ変更プログラム、および閲覧コンテンツ変更プログラムを記録したコンピュータ読み取り可能な記録媒体
WO2022011982A1 (zh) 物品流通跟踪处理方法及相关组件

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: 21953321

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE