WO2022021400A1 - 一种电商评论鉴别标记系统 - Google Patents

一种电商评论鉴别标记系统 Download PDF

Info

Publication number
WO2022021400A1
WO2022021400A1 PCT/CN2020/106370 CN2020106370W WO2022021400A1 WO 2022021400 A1 WO2022021400 A1 WO 2022021400A1 CN 2020106370 W CN2020106370 W CN 2020106370W WO 2022021400 A1 WO2022021400 A1 WO 2022021400A1
Authority
WO
WIPO (PCT)
Prior art keywords
comment
module
identification
review
commerce
Prior art date
Application number
PCT/CN2020/106370
Other languages
English (en)
French (fr)
Inventor
陈钦鹏
Original Assignee
深圳齐心集团股份有限公司
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 深圳齐心集团股份有限公司 filed Critical 深圳齐心集团股份有限公司
Priority to PCT/CN2020/106370 priority Critical patent/WO2022021400A1/zh
Publication of WO2022021400A1 publication Critical patent/WO2022021400A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques

Definitions

  • the invention belongs to the field of e-commerce, in particular to an e-commerce review identification marking system.
  • e-commerce has become a widely used business trade method. Buyers and sellers mainly conduct transactions through e-commerce web pages or software. Since e-commerce does not have traditional physical stores and does not have high requirements on the number of sales personnel, it can control operating costs more than traditional transaction models, so it has a greater price advantage. However, in order to increase their sales, there are many unscrupulous merchants who hire professional review evaluation teams to create a large number of fake reviews to falsely promote their products, thereby deceiving consumers to increase their real sales.
  • the embodiment of the present invention provides an e-commerce review identification and marking system, which aims to solve the problem of the lack of accurate and effective related equipment to realize the identification of product review quality in the prior art.
  • an e-commerce comment identification and marking system includes: a comment document construction module, a comment mapping module, a comment quality identification module, and an ID extraction module; wherein, the comment document construction module is used to capture Get the review data, and at the same time, classify the review data according to the product category to construct a product review document corresponding to the product; the review mapping module is used to extract sensitive keywords from the product review document, and compare the extracted sensitive keywords with the described The comment information in the product review document establishes a mapping relationship; the comment quality identification module is used for quality identification of the comment information in the established mapping relationship, and marks the identified false comments; the ID extraction module is used for Extract the ID information of the flagged fake reviews and tag them, and tag the same and similar IDs in other product review areas in the store.
  • the output terminals of the review document building module are respectively connected to the input terminals of the review mapping module and the ID extraction module; the review quality identification module is respectively connected to the output terminals of the review mapping module and the ID extraction module.
  • the review quality identification module includes:
  • a receiving unit configured to receive the comment information in the mapping relationship established by the comment mapping module
  • the false comment identification unit is used for quality identification of the comment information in the established mapping relationship, and marks the identified false comments.
  • it also includes: a marking start time inputting module for inputting the marking start time for the false comments and IDs marked in the comment quality identification module.
  • it also includes: a storage module for storing the marked ID information.
  • a marker ID timing deletion module configured to calculate the time value of the marker ID stored in the storage module according to the marker start time and the current system time entered by the marker ID, and compare the time value with the preset time value.
  • the set time threshold is compared, and when the time value is greater than the preset time threshold, the ID of the mark is deleted from the storage module.
  • the time threshold is 30-60 days.
  • the method further includes: a data redundancy judgment module, connected to the comment quality identification module and the storage module, for determining whether the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • a data redundancy judgment module connected to the comment quality identification module and the storage module, for determining whether the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • it also includes: a same ID deletion module, for deleting the ID identified in the comment quality identification module when the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • a same ID deletion module for deleting the ID identified in the comment quality identification module when the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • the e-commerce review identification and marking system captures review data through a review document construction module, and at the same time classifies the review data according to commodity categories to construct product review documents corresponding to the commodities; Sensitive keywords are extracted from the document, and a mapping relationship is established between the extracted sensitive keywords and the comment information in the product review document; The fake reviews are marked, and finally the ID information of the marked fake reviews is extracted and marked through the ID extraction module, and the same and similar IDs are marked in the other product review areas in the store, which can identify and mark the fake reviews and product reviews in the product reviews. ID, the judgment result is highly reliable.
  • FIG. 1 is a schematic structural diagram of an e-commerce review identification marking system provided by an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a review quality identification module provided by an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of another e-commerce review identification marking system provided by an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of another e-commerce review identification marking system provided by an embodiment of the present invention.
  • the e-commerce review identification and marking system captures review data through a review document construction module, and at the same time classifies the review data according to commodity categories to construct product review documents corresponding to the commodities; Sensitive keywords are extracted from the document, and a mapping relationship is established between the extracted sensitive keywords and the comment information in the product review document; The fake reviews are marked, and finally the ID information of the marked fake reviews is extracted and marked through the ID extraction module, and the same and similar IDs are marked in the other product review areas in the store, which can identify and mark the fake reviews and product reviews in the product reviews. ID, the judgment result is highly reliable.
  • an e-commerce comment identification and marking system 1 includes: a comment document construction module 11, a comment mapping module 12, a comment quality identification module 13, and an ID extraction module 14; wherein , the review document building module 11 is used to capture the review data, and at the same time, the review data is classified according to the product category to construct a product review document corresponding to the product; the review mapping module 12 is used to extract from the product review document.
  • Sensitive keywords establish a mapping relationship between the extracted sensitive keywords and the comment information in the product review document;
  • the comment quality identification module 13 is used for quality identification of the comment information in the established mapping relationship, and after identification mark the false comments;
  • the ID extraction module 14 is used to extract the ID information of the marked false comments, mark them, and mark the same and similar IDs in the comment area of other products in the store; Get the review data, and at the same time classify the review data according to the product category to construct a product review document corresponding to the product; and extract sensitive keywords from the product review document through the review mapping module, and associate the extracted sensitive keywords with the product review document.
  • the quality identification of the comment information in the established mapping relationship is carried out, and the identified fake comments are marked, and finally the ID information of the marked fake comments is extracted through the ID extraction module. , and mark it, and mark the same and similar IDs in other product review areas in the store, which can identify and mark false reviews and IDs in product reviews, and the judgment results are highly reliable.
  • the output end of the comment document construction module 11 is respectively connected with the input end of the comment mapping module 12 and the ID extraction module 14; the comment quality identification module 13 is respectively connected with the comment mapping module 12 and The output of the ID extraction module 14 is connected.
  • the comment quality identification module 13 includes: a receiving unit 131 for receiving comment information in the mapping relationship established by the comment mapping module; and a false comment identification unit 132 for The quality of the review information in the established mapping relationship is identified, and the identified fake reviews are marked.
  • the system 1 further includes: a mark start time entry module 15 , a storage module 16 and a mark ID timing deletion module 17 .
  • the marking start time input module is used to input the marking start time for the false comments and IDs marked in the comment quality identification module.
  • the storage module is used to store the marked ID information.
  • the marked ID timing deletion module is used to calculate the time value of the marked ID stored in the storage module according to the marked start time and the current system time entered by the marked ID, and compare the time value with the preset time threshold. The comparison is performed, and when the time value is greater than the preset time threshold, the ID of the mark is deleted from the storage module.
  • the time threshold may be 30 to 60 days.
  • the marking start time entered by the marking start time input module to an ID identified in the comment quality identification module is 2011-06-06, and the current system time is 2011-07- 06, then the described mark ID timing deletion module deletes the ID of this mark stored in the storage module; another example, when the time threshold is 45 days, the described mark start time input module marks the mark in the comment quality identification module.
  • the marked start time of an ID entry is 2011-06-06, and the current system time is 2011-07-21, then the marked ID timing deletion module deletes the marked ID stored in the storage module;
  • the time threshold is 60 days, and the marking starting time of the marking starting time input module to a marking ID identified in the comment quality identification module is 2011-06-06, and the current system time is 2011-08-06 , the tag ID timing deletion module deletes the tag ID stored in the storage module.
  • the system 1 further includes: a data redundancy judgment module 18 and an identical ID deletion module 19 .
  • the data redundancy judgment module 18 is connected to the comment quality identification module 13 and the storage module 16, and is used to judge whether the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • the same ID deletion module 19 is used to delete the ID identified in the comment quality identification module when the ID identified in the comment quality identification module is the same as the ID stored in the storage module.
  • the storage module stores a marked ID of 123456
  • the database redundancy module recognizes that the marked ID in the comment quality identification module is 123456
  • the same ID deletion module deletes the ID marked in the comment quality identification module as 123456. ID removed.
  • the e-commerce review identification and marking system captures review data through a review document construction module, and at the same time classifies the review data according to commodity categories to construct product review documents corresponding to the commodities; Sensitive keywords are extracted from the document, and a mapping relationship is established between the extracted sensitive keywords and the comment information in the product review document; The fake reviews are marked, and finally the ID information of the marked fake reviews is extracted and marked through the ID extraction module, and the same and similar IDs are marked in the other product review areas in the store, which can identify and mark the fake reviews and product reviews in the product reviews. ID, the judgment result is highly reliable.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种电商评论鉴别标记系统(1)，包括：评论文档构建模块(11)、评论映射模块(12)、评论质量鉴别模块(13)以及ID提取模块(14)；其中，所述评论文档构建模块(11)，用于抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；所述评论映射模块(12)，用于从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；所述评论质量鉴别模块(13)，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记；所述ID提取模块(14)，用于提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记，能够鉴别标记出产品评论中的虚假评论和ID，判断结果可靠性高。

Description

一种电商评论鉴别标记系统 技术领域
本发明属于电子商务领域，尤其涉及一种电商评论鉴别标记系统。
背景技术
在当代，随着互联网的普及，电子商务已经成为一种被广泛利用的商业贸易方式。买卖双方主要是通过电商的网页或者是软件进行交易活动。由于电子商务没有传统的实体店面，对销售人员的数量要求也不高，所以相比传统交易模式更能够控制运营成本，因而有着更大的价格优势。但是，有很多不法商家为了提高自己的销量从而雇佣专业刷评价团队也制造大量的虚假评论来对自己的商品进行虚假的宣传，从而欺骗消费者来提高自己的真实销量。
目前电子商务的发展迅猛，体量巨大，电商环境中的卖家数量众多，用户在进行购买决定时难以判断商品描述的真实性，对商品评价的依赖度很高，由于卖家评价作弊而造成的商品的性能好评度虚高的情况引起的买家利益损失的情况严重。在这样的情况下，如何对电子商务中商家的评价作弊行为进行识别和判断成电子商务发展过程中亟待解决的问题；在判断虚假评论过程中如何提高判断的准确性，避免误判情况的发生也是十分重要的考量因素；目前现有技术中还缺乏准确有效的相关设备实现产品评论质量的鉴别。
发明内容
本发明实施例提供一种电商评论鉴别标记系统，旨在解决现有技术中还缺乏准确有效的相关设备实现产品评论质量的鉴别的问题。
本发明实施例是这样实现的，一种电商评论鉴别标记系统，包括：评论文档构建模块、评论映射模块、评论质量鉴别模块以及ID提取模块；其中，所述 评论文档构建模块，用于抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；所述评论映射模块，用于从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；所述评论质量鉴别模块，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记；所述ID提取模块，用于提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记。
优选地，所述评论文档构建模块输出端分别与所述评论映射模块和ID提取模块的输入端连接；所述评论质量鉴别模块分别与所述评论映射模块和ID提取模块的输出端连接。
优选地，所述评论质量鉴别模块，包括：
接收单元，用于接收评论映射模块建立映射关系中的评论信息；以及
虚假评论标识单元，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记。
优选地，还包括：标记起始时间录入模块，用于对评论质量鉴别模块内标记的虚假评论和ID录入标记起始时间。
优选地，还包括：存储模块，用于存储被标记的ID信息。
优选地，还包括：标记ID定时删除模块，用于根据标记ID录入的标记起始时间和当前系统时间，计算出该标记的ID在存储模块内存储的时间值，并将该时间值与预设的时间阈值进行比对，当该时间值大于预设的时间阈值时，则从存储模块内删除该标记的ID。
优选地，所述时间阈值为30～60天。
优选地，还包括：数据冗余判断模块，与所述评论质量鉴别模块和存储模块连接，用于判断评论质量鉴别模块内标识的ID与存储模块内存储的ID是否相同。
优选地，还包括：相同ID删除模块，用于当评论质量鉴别模块内标识的 ID与存储模块内存储的ID相同时，则删除评论质量鉴别模块内标识的ID。
本发明实施例提供的电商评论鉴别标记系统，通过评论文档构建模块抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；并通过评论映射模块从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；然后通过评论质量鉴别模块对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记，最后通过ID提取模块提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记，能够鉴别标记出产品评论中的虚假评论和ID，判断结果可靠性高。
附图说明
为了更清楚地说明本发明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，下面描述中的附图是本发明的一些实施例，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。
以下附图仅旨在于对本发明做示意性说明和解释，并不限定本发明的范围。
图1是本发明实施例提供的一种电商评论鉴别标记系统的结构示意图；
图2是本发明实施例提供的评论质量鉴别模块的结构示意图；
图3是本发明实施例提供的另一种电商评论鉴别标记系统的结构示意图；
图4是本发明实施例提供的又一种电商评论鉴别标记系统的结构示意图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
本发明实施例提供的电商评论鉴别标记系统，通过评论文档构建模块抓取 评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；并通过评论映射模块从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；然后通过评论质量鉴别模块对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记，最后通过ID提取模块提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记，能够鉴别标记出产品评论中的虚假评论和ID，判断结果可靠性高。
以下结合具体实施例对本发明的具体实现进行详细描述。
如图1所示，在本发明实施例中，一种电商评论鉴别标记系统1，包括：包括：评论文档构建模块11、评论映射模块12、评论质量鉴别模块13以及ID提取模块14；其中，所述评论文档构建模块11，用于抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；所述评论映射模块12，用于从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；所述评论质量鉴别模块13，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记；所述ID提取模块14，用于提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记；通过评论文档构建模块抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；并通过评论映射模块从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；然后通过评论质量鉴别模块对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记，最后通过ID提取模块提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记，能够鉴别标记出产品评论中的虚假评论和ID，判断结果可靠性高。
在本发明实施例中，所述评论文档构建模块11输出端分别与所述评论映射模块12和ID提取模块14的输入端连接；所述评论质量鉴别模块13分别与所 述评论映射模块12和ID提取模块14的输出端连接。
在本发明实施例中，如图2所示，所述评论质量鉴别模块13，包括：接收单元131，用于接收评论映射模块建立映射关系中的评论信息；以及虚假评论标识单元132，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记。
在本发明实施例中，如图3所示，所述系统1还包括：标记起始时间录入模块15、存储模块16和标记ID定时删除模块17。其中，所述标记起始时间录入模块，用于对评论质量鉴别模块内标记的虚假评论和ID录入标记起始时间。所述存储模块，用于存储被标记的ID信息。所述标记ID定时删除模块，用于根据标记ID录入的标记起始时间和当前系统时间，计算出该标记的ID在存储模块内存储的时间值，并将该时间值与预设的时间阈值进行比对，当该时间值大于预设的时间阈值时，则从存储模块内删除该标记的ID。
在本实施例中，所述时间阈值可为30～60天。例如，当所述时间阈值为30天，所述标记起始时间录入模块对评论质量鉴别模块内标识的一ID录入的标记起始时间为2011-06-06，当前系统时间为2011-07-06，则所述标记ID定时删除模块删除存储模块内存储的该标记的ID；又如，当所述时间阈值为45天，所述所述标记起始时间录入模块对评论质量鉴别模块内标记的一ID录入的标记起始时间为2011-06-06，当前系统时间为2011-07-21，则所述标记ID定时删除模块删除存储模块内存储的该标记的ID；再如，当所述时间阈值为60天，所述所述标记起始时间录入模块对评论质量鉴别模块内标识的一标记ID录入的标记起始时间为2011-06-06，当前系统时间为2011-08-06，则所述标记ID定时删除模块删除存储模块内存储的该标记的ID。
在本发明实施例中，如图4所示，所述系统1还包括：数据冗余判断模块18和相同ID删除模块19。其中，所述数据冗余判断模块18，与所述评论质量鉴别模块13和存储模块16连接，用于判断评论质量鉴别模块内标识的ID与存储模块内存储的ID是否相同。所述相同ID删除模块19，用于当评论质量鉴别 模块内标识的ID与存储模块内存储的ID相同时，则删除评论质量鉴别模块内标识的ID。例如，当存储模块内存储有一标记的ID为123456，所述数据库冗余模块识别出评论质量鉴别模块内标记的ID为123456，则相同ID删除模块将评论质量鉴别模块内标识的ID为123456的ID删除。
上述发明实施例提供的电商评论鉴别标记系统，通过评论文档构建模块抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；并通过评论映射模块从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；然后通过评论质量鉴别模块对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记，最后通过ID提取模块提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记，能够鉴别标记出产品评论中的虚假评论和ID，判断结果可靠性高。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (9)

  1. 一种电商评论鉴别标记系统，其特征在于，包括：评论文档构建模块、评论映射模块、评论质量鉴别模块以及ID提取模块；其中，所述评论文档构建模块，用于抓取评论数据，同时将评论数据按商品类别进行分类构建与商品相对应的产品评论文档；所述评论映射模块，用于从产品评论文档内提取敏感关键词，将提取的敏感关键词与所述产品评论文档内的评论信息建立映射关系；所述评论质量鉴别模块，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记；所述ID提取模块，用于提取被标记的虚假评论的ID信息，并进行标记，并在店内其他产品评论区对相同和相似ID进行标记。
  2. 如权利要求1所述的电商评论鉴别标记系统，其特征在于，所述评论文档构建模块输出端分别与所述评论映射模块和ID提取模块的输入端连接；所述评论质量鉴别模块分别与所述评论映射模块和ID提取模块的输出端连接。
  3. 如权利要求1所述的电商评论鉴别标记系统，其特征在于，所述评论质量鉴别模块，包括：
    接收单元，用于接收评论映射模块建立映射关系中的评论信息；以及
    虚假评论标识单元，用于对建立映射关系中的评论信息进行质量鉴别，并对鉴别后的虚假评论进行标记。
  4. 如权利要求3所述的电商评论鉴别标记系统，其特征在于，还包括：标记起始时间录入模块，用于对评论质量鉴别模块内标记的虚假评论和ID录入标记起始时间。
  5. 如权利要求4所述的电商评论鉴别标记系统，其特征在于，还包括：存储模块，用于存储被标记的ID信息。
  6. 如权利要求5所述的电商评论鉴别标记系统，其特征在于，还包括：标记ID定时删除模块，用于根据标记ID录入的标记起始时间和当前系统时间，计算出该标记的ID在存储模块内存储的时间值，并将该时间值与预设的时间阈值进行比对，当该时间值大于预设的时间阈值时，则从存储模块内删除该标记 的ID。
  7. 如权利要求6所述的电商评论鉴别标记系统，其特征在于，所述时间阈值为30～60天。
  8. 如权利要求7所述的电商评论鉴别标记系统，其特征在于，还包括：数据冗余判断模块，与所述评论质量鉴别模块和存储模块连接，用于判断评论质量鉴别模块内标识的ID与存储模块内存储的ID是否相同。
  9. 如权利要求8所述的电商评论鉴别标记系统，其特征在于，还包括：相同ID删除模块，用于当评论质量鉴别模块内标识的ID与存储模块内存储的ID相同时，则删除评论质量鉴别模块内标识的ID。
PCT/CN2020/106370 2020-07-31 2020-07-31 一种电商评论鉴别标记系统 WO2022021400A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/106370 WO2022021400A1 (zh) 2020-07-31 2020-07-31 一种电商评论鉴别标记系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/106370 WO2022021400A1 (zh) 2020-07-31 2020-07-31 一种电商评论鉴别标记系统

Publications (1)

Publication Number Publication Date
WO2022021400A1 true WO2022021400A1 (zh) 2022-02-03

Family

ID=80037017

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/106370 WO2022021400A1 (zh) 2020-07-31 2020-07-31 一种电商评论鉴别标记系统

Country Status (1)

Country Link
WO (1) WO2022021400A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114626885A (zh) * 2022-03-17 2022-06-14 华院分析技术(上海)有限公司 一种基于大数据的零售管理方法和系统
CN115983873A (zh) * 2022-12-29 2023-04-18 珠海蓝度科技有限公司 一种基于大数据的用户数据分析管理系统及方法
CN117312634A (zh) * 2023-11-29 2023-12-29 大文传媒集团(山东)有限公司 人工智能数据整合传播处理系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8131655B1 (en) * 2008-05-30 2012-03-06 Bitdefender IPR Management Ltd. Spam filtering using feature relevance assignment in neural networks
CN107577759A (zh) * 2017-09-01 2018-01-12 安徽广播电视大学 用户评论自动推荐方法
CN109377080A (zh) * 2018-11-13 2019-02-22 平安科技（深圳）有限公司 商品评价方法、装置、计算机设备及存储介质
CN110457558A (zh) * 2019-07-31 2019-11-15 沃民高新科技（北京）股份有限公司 网络水军的识别方法及装置、存储介质和处理器

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8131655B1 (en) * 2008-05-30 2012-03-06 Bitdefender IPR Management Ltd. Spam filtering using feature relevance assignment in neural networks
CN107577759A (zh) * 2017-09-01 2018-01-12 安徽广播电视大学 用户评论自动推荐方法
CN109377080A (zh) * 2018-11-13 2019-02-22 平安科技（深圳）有限公司 商品评价方法、装置、计算机设备及存储介质
CN110457558A (zh) * 2019-07-31 2019-11-15 沃民高新科技（北京）股份有限公司 网络水军的识别方法及装置、存储介质和处理器

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114626885A (zh) * 2022-03-17 2022-06-14 华院分析技术(上海)有限公司 一种基于大数据的零售管理方法和系统
CN115983873A (zh) * 2022-12-29 2023-04-18 珠海蓝度科技有限公司 一种基于大数据的用户数据分析管理系统及方法
CN117312634A (zh) * 2023-11-29 2023-12-29 大文传媒集团(山东)有限公司 人工智能数据整合传播处理系统
CN117312634B (zh) * 2023-11-29 2024-02-20 大文传媒集团(山东)有限公司 人工智能数据整合传播处理系统

Similar Documents

Publication Publication Date Title
WO2022021400A1 (zh) 一种电商评论鉴别标记系统
CN111428599B (zh) 票据识别方法、装置和设备
US20140258169A1 (en) Method and system for automated verification of customer reviews
US20050210016A1 (en) Confidence-based conversion of language to data systems and methods
US11138372B2 (en) System and method for reporting based on electronic documents
JPWO2019008766A1 (ja) 証憑処理システムおよび証憑処理プログラム
CN104881795A (zh) 一种电商虚假评价的判断识别方法
US20050210046A1 (en) Context-based conversion of language to data systems and methods
CN111914729A (zh) 凭证关联方法、装置、计算机设备及存储介质
CN115496555A (zh) 一种智能化跨境电商网站安全质量评估方法及系统
CN114118816A (zh) 一种风险评估方法、装置、设备及计算机存储介质
WO2019006642A1 (zh) 一种电子商务产品评论质量鉴别系统
KR102650139B1 (ko) 인공지능 기반 온라인 위조상품 단속 시스템 및 방법
CN107392654A (zh) 一种电子商务产品评论质量鉴别系统
WO2023045535A1 (zh) 识别图片的方法和装置
JP6507459B2 (ja) 会計処理システム
CN115908062A (zh) 一种知识产权全周期管理系统
TWM575887U (zh) 智能會計帳務系統
US20180046663A1 (en) System and method for completing electronic documents
TWM553835U (zh) 自動智慧化帳務系統
CN115470177A (zh) 文件处理方法、装置、设备及计算机存储介质
CN113763143A (zh) 审计处理方法、计算机设备及存储装置
CN113626655A (zh) 提取文件中信息的方法、计算机设备及存储装置
CN112232894A (zh) 基于增值税发票的数据分析方法
KR102533239B1 (ko) 인공지능 기반 위조 상품 탐지를 위한 이미지 라이브러리 시스템 및 가품 탐지 방법

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: 20947136

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20947136

Country of ref document: EP

Kind code of ref document: A1