CN112651768A - E-commerce analysis method and system based on block chain - Google Patents

E-commerce analysis method and system based on block chain Download PDF

Info

Publication number
CN112651768A
CN112651768A CN202011406667.3A CN202011406667A CN112651768A CN 112651768 A CN112651768 A CN 112651768A CN 202011406667 A CN202011406667 A CN 202011406667A CN 112651768 A CN112651768 A CN 112651768A
Authority
CN
China
Prior art keywords
module
classification
different
user
block chain
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.)
Pending
Application number
CN202011406667.3A
Other languages
Chinese (zh)
Inventor
王治东
谢绍韫
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.)
Suzhou Black Cloud Intelligent Technology Co ltd
Original Assignee
Suzhou Black Cloud Intelligent 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 Suzhou Black Cloud Intelligent Technology Co ltd filed Critical Suzhou Black Cloud Intelligent Technology Co ltd
Priority to CN202011406667.3A priority Critical patent/CN112651768A/en
Publication of CN112651768A publication Critical patent/CN112651768A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/31—Indexing; Data structures therefor; Storage structures
    • G06F16/313—Selection or weighting of terms for indexing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/374—Thesaurus
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates

Landscapes

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

Abstract

The invention discloses an e-commerce analysis method and system based on a block chain. Classifying product comments according to different user geographic positions, segmenting words after preprocessing the product comments in different geographic positions, obtaining a keyword set by using a TF-IDF model, classifying the keyword set, carrying out dictionary matching on emotional tendencies to obtain classifications of different emotional tendencies, analyzing different emotional tendencies by an LDA topic model, and proposing suggestions for preventing user loss by combining the geographic positions of the users; and storing the classification and facies analysis results in the block chain, and updating the basic word bank in the block chain according to the classification results. The system comprises a web crawler module, an analysis module and a block chain storage module. According to the method, by increasing the geographical distribution of the users and the classification of the product comments, different sales strategies and after-sale services can be conveniently carried out on the users in different geographical positions by the merchants; by using the block chain, the safety and the privacy of data storage are improved.

Description

E-commerce analysis method and system based on block chain
Technical Field
The invention relates to the technical field of e-commerce, in particular to an e-commerce analysis method and system based on a block chain.
Background
With the continuous development of the internet and the logistics industry, online shopping becomes the first choice of most people; meanwhile, the freedom degree of the netizens for publishing the information is high, and the propagation speed of the netizens is increased by geometric multiple. Based on the above, the product reviews after online shopping have great influence on the brand image and public praise of the product, the requirements and the preferences of the user can be known through the product reviews, the information is collected and analyzed, and brand service can be optimized, product iteration is accelerated, and the market occupation ratio can be presumed. However, the amount of data of the product review information is enormous, and it is very difficult to find useful information therefrom. The existing large e-commerce platforms often lack the functions of product comment classification and user geographic distribution analysis for analyzing commodity evaluation information, do not have the functions of product comment classification and user geographic distribution analysis, and cannot perform different sales strategies and after-sales services for users in different areas. Meanwhile, in data storage, the existing security and privacy are poor, and data cannot be stored safely.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a block chain-based e-commerce analysis method and system which can improve the safety and privacy of data storage, classify product reviews and analyze user geographical distribution.
In order to solve the technical problem, the invention provides an e-commerce analysis method based on a block chain, which comprises the following steps:
step 1: obtaining product comment information and corresponding user geographical position information, and classifying the product comment information according to different user geographical positions;
step 2: product comment information in different geographic positions is preprocessed, and word segmentation is carried out on preprocessed data to obtain a word set;
and step 3: carrying out data mining on the word set by using a TF-IDF model, and removing non-key words to obtain a keyword set;
and 4, step 4: classifying the keyword set, and performing dictionary matching of emotional tendency on the classified keywords to obtain classification of different emotional tendency;
and 5: analyzing different emotional tendencies through an LDA topic model, analyzing the reason of user loss and providing a suggestion for preventing the user loss by combining the geographical position of the user;
step 6: and classifying the emotional tendency in different geographic positions and storing corresponding analysis results in the block chain, and updating the basic word bank in the block chain according to the classification results of different emotional tendency.
Further, the product comment information is preprocessed in the step 2, so that the comment information is deduplicated.
Further, in the step 2, the preprocessed data is segmented, and a segmentation method is adopted by using a jieba word segmentation device.
Further, in the step 3, a TF-IDF model is used to perform data mining on the word set, specifically, a word list to be removed is defined, a skleran module under Python is used to implement TF-IDF, and a keyword set suitable for classification is selected.
Further, the method for classifying the keyword set in step 4 is to perform clustering by using k-means, specifically, perform k-means clustering on the keyword set to obtain a classification and a centroid of each class, calculate a distance from each keyword to the centroid, select a class closest to the centroid, and add the class.
Further, in the step 5, different emotional tendencies are analyzed through the LDA topic model, specifically, potential topics in the data set are mined through the LDA topic model, and then concentrated attention points and related feature words of the data set are analyzed, and the reason for user loss is analyzed on the basis.
The invention also provides an e-commerce analysis system based on the block chain, which is characterized in that: comprises a web crawler module, an analysis module and a block chain storage module,
the web crawler module is used for acquiring product comment information and corresponding user geographical position information, providing the information to the analysis module and updating the information at regular time;
the analysis module analyzes the reason of the user loss and provides a suggestion for preventing the user loss by combining the geographical position of the user;
the block chain storage module is used for storing the classification of different emotional tendencies and geographic positions and corresponding analysis results on the block chain.
Further, the analysis module comprises a geographic classification module, a comment classification module and an attrition prediction module,
the geographic classification module classifies the information of the geographic positions of the users extracted by the web crawler module and provides the comment classification module with the comment information of the products in different geographic positions;
the comment classification module classifies the product comment information in different geographic positions to obtain classifications of different emotional tendencies and provides classification results to the loss prediction module;
the loss prediction module analyzes different emotional tendencies and provides suggestions for preventing user loss by combining the geographical position of the user.
Furthermore, the comment classification module performs word segmentation on the product comment information in different geographic positions, performs data mining on a word set by using a TF-IDF model on the basis of the word segmentation to obtain a keyword set, and performs dictionary matching of emotional tendency after the keyword set is classified to obtain classification of different emotional tendency.
Further, the geographic classification module generates a corresponding distribution graph according to different user geographic position information, and associates the distribution graph with comment data to search local user comment information according to the geographic position.
The invention has the beneficial effects that: according to the method, the geographical distribution of the users and the classification of the product comments are increased, the reason of the user loss can be analyzed, the suggestion for preventing the user loss is provided by combining the geographical positions of the users, and different sales strategies and after-sale services can be conveniently carried out on the users in different geographical positions by merchants; by using blockchain storage, the security and the privacy of data storage are improved.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood and to implement them in accordance with the contents of the description, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings.
Drawings
Fig. 1 is a flow chart of a block chain based e-commerce analysis method.
Fig. 2 is a block chain-based framework diagram of an e-commerce analysis system.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
In the description of the present invention, it should be understood that the term "comprises/comprising" is intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to the listed steps or elements but may alternatively include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
As shown in the flowchart of fig. 1, the block chain-based e-commerce analysis method of the present invention includes the following steps:
step 1: and obtaining the product comment information and the corresponding user geographical position information, and classifying the product comment information according to different user geographical positions. In this embodiment, product review information is preprocessed to remove duplicate comment information. If the comment information of different shoppers is completely repeated, the probability is that a single line is swiped, the information is generally meaningless, so that the completely repeated comment information needs to be deduplicated, and useful comment information and corresponding user geographical position information are reserved.
Step 2: product comment information in different geographic positions is preprocessed, and word segmentation is carried out on preprocessed data to obtain a word set. In this embodiment, the preprocessed data is subjected to word segmentation, and the word segmentation method is implemented by using a jieba word segmentation device.
And step 3: and (4) carrying out data mining on the word set by using a TF-IDF model, and removing non-key words to obtain a keyword set. Defining a word list needing to be removed, using a sklern module under Python to realize TF-IDF, and selecting a keyword set suitable for classification. In this embodiment, the word list to be removed is defined to include english, numbers, panning, down jackets, and the like, and then the tfidern module under Python is used to implement TF-IDF, that is, the tfidfvactor function provided by the skleern is used to create the object of the word steering amount, and the fit _ transform function converts the keyword list into the word frequency matrix. TF-IDF (term frequency-inverse document frequency) is a common weighting technology for information retrieval and data mining, is commonly used for mining keywords in articles, has simple and efficient algorithm, and is often used for text data cleaning. If a word or a word appears in an article with a high frequency TF and rarely appears in other articles, the word or the word is considered to have a good classification capability and is suitable for classification. Wherein TF represents the frequency of occurrences of terms in the article; IDF represents the word discrimination, which is greater if there are fewer documents containing a word.
And 4, step 4: classifying the keyword set, in this embodiment, classifying the keyword set by using k-means, specifically, performing k-means clustering on the keyword set to obtain a classification and a centroid of each type, then calculating a distance from each keyword to the centroid, and selecting the closest class to add to complete classification. The method for classifying the keyword set can also be used for enabling the words classified by the jieba word classifier to collide with a basic word stock stored in the block chain and classifying the words into corresponding classifications. The word frequency statistics is carried out on various keywords, word cloud is drawn to check word segmentation effects, the word cloud provides a certain degree of first impression, the attention can be attracted compared with other bar graphs, the word frequency of texts or websites can be analyzed rapidly, and the texts or the websites can be displayed in various styles. Performing dictionary matching of emotional tendency on the classified keywords to obtain classification of different emotional tendency; the emotional tendency is the attitude expressed by the user to the commodity, namely positive emotion, negative emotion and neutral emotion. The dictionary matching method used in this embodiment is a word set (beta version) for web-emotion analysis, which can better process the product comment information and perform corresponding optimization on the word set. Words frequently used on online shopping comments such as high cost performance, fast logistics, good service and the like are added into the positive emotion word list according to the emotional tendency to obtain three types of positive emotion, negative emotion and neutral emotion.
And 5: different emotional tendencies are analyzed through an LDA topic model, the reason of user loss is analyzed, and a suggestion for preventing the user loss is provided by combining the geographical position of the user. Potential subjects in the data set are mined through the LDA subject model, concentrated attention points and relevant feature words of the data set are further analyzed, and the reason of user loss is analyzed on the basis. The topics under different emotional tendencies and the high-frequency characteristic words in the topics are mined, the concentrated attention points and the related characteristic words are found, namely, for key indexes of the lost users, such as cost performance, logistics, service and the like, a basic basis is provided for analyzing the reason of the user loss according to different indexes, and suggestions for preventing the user loss are provided according to the basic basis. And (3) constructing LDA theme models for positive emotion, negative emotion and neutral emotion respectively by using Gensim of Python, analyzing the themes and high-frequency feature words in the themes to obtain the reason for user churn, and proposing suggestions on how to increase user viscosity and prevent user churn of the product according to the geographical position of the user. For example, the overall quality of customer service and the quality of service can be improved according to the after-sales service index. Words and probabilities obtained by LDA theme analysis on negative emotions are the basic reason of user loss, while LDA theme analysis on positive emotions is the basic reason of product purchase by users, and suggestion is provided to keep current service.
Step 6: and classifying the emotional tendency in different geographic positions and storing corresponding analysis results in the block chain, and updating the basic word bank in the block chain according to the classification results of different emotional tendency.
As shown in the frame diagram of fig. 2, the block chain-based e-commerce analysis system of the present invention includes a web crawler module, an analysis module, and a block chain storage module. The web crawler module is used for acquiring product comment information and corresponding user geographical position information, providing the information to the analysis module and updating the information at regular time; the analysis module analyzes the reason of the user loss and provides a suggestion for preventing the user loss by combining the geographical position of the user; the block chain storage module is used for storing the classification of different emotional tendencies and geographic positions and corresponding analysis results on the block chain, and the storage results on the block chain are used for improving the safety and the privacy of data storage. Blockchain storage is a decentralized and data-sharing storage system constructed by using blockchains, and is an effective combination of blockchains and storage systems. The data stored in the storage device has the characteristics of being incapable of being forged, having trace in the whole process, being traceable, being publicly transparent, being maintained in a collective mode and the like. The information stored on the blockchain is public, but the account identity information is highly encrypted and can only be accessed under the authorization of the data owner, thereby ensuring the security of the data and the privacy of individuals.
In this embodiment, the analysis module includes a geo-classification module, a review classification module, and an attrition prediction module. The geographic classification module classifies the information of the geographic positions of the users extracted by the web crawler module and provides the comment classification module with the comment information of the products in different geographic positions; the comment classification module classifies the product comment information in different geographic positions to obtain classifications of different emotional tendencies and provides classification results to the loss prediction module; the loss prediction module analyzes different emotional tendencies and provides suggestions for preventing user loss by combining the geographical position of the user.
In this embodiment, the comment classification module performs word segmentation on product comment information in different geographic positions, performs data mining on a word set by using a TF-IDF model on the basis of the word segmentation to obtain a keyword set, and performs dictionary matching of emotional tendencies after classifying the keyword set to obtain classifications of different emotional tendencies.
In this embodiment, the geo-categorizing module generates a corresponding distribution graph according to different user geographical location information, associates the distribution graph with comment data, and is configured to search local user comment information according to a geographical location, and input a place name to obtain corresponding local user comment information. Through the distribution diagram, the merchant can check and analyze the comments of all merchants corresponding to the products, and for the merchants, the merchant can visually compare the advantages and disadvantages of the merchants with those of the same party to find out the places needing improvement.
The invention has the beneficial effects that: according to the method, the geographical distribution of the users and the classification of the product comments are increased, the reason of the user loss can be analyzed, the suggestion for preventing the user loss is provided by combining the geographical positions of the users, and different sales strategies and after-sale services can be conveniently carried out on the users in different geographical positions by merchants; by using blockchain storage, the security and the privacy of data storage are improved.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.

Claims (10)

1.一种基于区块链的电商分析方法，其特征在于，包括以下步骤：1. a block chain-based e-commerce analysis method, is characterized in that, comprises the following steps: 步骤1：获取产品评论信息和对应的用户地理位置信息，根据不同的用户地理位置对产品评论信息进行归类；Step 1: Obtain product review information and corresponding user geographic location information, and classify product review information according to different user geographic locations; 步骤2：对不同地理位置下的产品评论信息进行预处理，对预处理后的数据进行分词，得到词语集；Step 2: Preprocess the product review information in different geographical locations, and perform word segmentation on the preprocessed data to obtain a word set; 步骤3：使用TF-IDF模型对词语集进行数据挖掘，去除非关键词语得到关键字集；Step 3: Use the TF-IDF model to perform data mining on the word set, and remove the non-key words to obtain the keyword set; 步骤4：对关键字集进行分类，对分类后的关键字进行情感倾向的词典匹配，得到不同情感倾向的分类；Step 4: Categorize the keyword set, perform dictionary matching of emotional tendencies on the classified keywords, and obtain classifications of different emotional tendencies; 步骤5：通过LDA主题模型分析不同的情感倾向，分析用户流失的原因并结合用户所处的地理位置提出预防用户流失的建议；Step 5: Analyze different emotional tendencies through the LDA topic model, analyze the reasons for user churn, and put forward suggestions to prevent user churn based on the user's geographic location; 步骤6：将不同地理位置下的情感倾向分类、以及相应的分析结果存储在区块链中，并根据不同情感倾向的分类结果更新区块链中的基础词库。Step 6: The classification of emotional tendencies in different geographical locations and the corresponding analysis results are stored in the blockchain, and the basic thesaurus in the blockchain is updated according to the classification results of different emotional tendencies. 2.根据权利要求1所述的基于区块链的电商分析方法，其特征在于：所述步骤2中对产品评论信息进行预处理，为对评论信息进行去重。2 . The blockchain-based e-commerce analysis method according to claim 1 , wherein in the step 2, the product review information is preprocessed to deduplicate the review information. 3 . 3.根据权利要求1所述的基于区块链的电商分析方法，其特征在于：所述步骤2中对预处理后的数据进行分词，采用的分词方法为使用jieba分词器。3. The blockchain-based e-commerce analysis method according to claim 1, characterized in that: in the step 2, the preprocessed data is segmented, and the segmented method adopted is to use a jieba tokenizer. 4.根据权利要求1所述的基于区块链的电商分析方法，其特征在于：所述步骤3中使用TF-IDF模型对词语集进行数据挖掘，具体为定义需要去除的词语列表，使用Python下的sklearn模块实现TF-IDF，选择出适合用来分类的关键字集。4. The e-commerce analysis method based on block chain according to claim 1, is characterized in that: in described step 3, use TF-IDF model to carry out data mining on word set, specifically for defining the word list that needs to be removed, using The sklearn module under Python implements TF-IDF and selects a keyword set suitable for classification. 5.根据权利要求1所述的基于区块链的电商分析方法，其特征在于：所述步骤4中对关键字集进行分类的方法为采用k-means进行聚类，具体为对关键字集进行k-means聚类得到分类和各类的质心，再计算每个关键字到质心的距离，选择距离最近的类加入，完成分类。5. The blockchain-based e-commerce analysis method according to claim 1, wherein the method for classifying the keyword set in the step 4 is to use k-means to perform clustering, specifically for Perform k-means clustering on the set to obtain the classification and the centroid of each category, and then calculate the distance from each keyword to the centroid, select the class with the closest distance to join, and complete the classification. 6.根据权利要求1所述的基于区块链的电商分析方法，其特征在于：所述步骤5中通过LDA主题模型分析不同的情感倾向，具体为通过LDA主题模型挖掘数据集中的潜在主题，进而分析数据集的集中关注点及其相关特征词，在此基础上分析用户流失的原因。6. The blockchain-based e-commerce analysis method according to claim 1, wherein in the step 5, different emotional tendencies are analyzed by LDA topic model, specifically mining potential topics in the data set by LDA topic model , and then analyze the focus of the dataset and its related feature words, and then analyze the reasons for user churn. 7.一种基于区块链的电商分析系统，其特征在于：包括网络爬虫模块、分析模块和区块链存储模块，7. A blockchain-based e-commerce analysis system, characterized in that: comprising a web crawler module, an analysis module and a blockchain storage module, 所述网络爬虫模块用于获取产品评论信息和对应的用户地理位置信息，将信息提供给所述分析模块，并定时更新；The web crawler module is used to obtain product review information and corresponding user geographic location information, provide the information to the analysis module, and update it regularly; 所述分析模块分析用户流失的原因并结合用户所处的地理位置提出预防用户流失的建议；The analysis module analyzes the reasons for user loss and proposes suggestions for preventing user loss in combination with the geographic location of the user; 所述区块链存储模块用于将不同情感倾向和地理位置的分类、以及相应的分析结果存储在区块链上。The block chain storage module is used to store the classification of different emotional tendencies and geographic locations, and the corresponding analysis results on the block chain. 8.根据权利要求7所述的基于区块链的电商分析系统，其特征在于：所述分析模块包括地理分类模块、评论分类模块和流失预测模块，8. The blockchain-based e-commerce analysis system according to claim 7, wherein the analysis module comprises a geographic classification module, a comment classification module and a loss prediction module, 所述地理分类模块对网络爬虫模块提取到用户地理位置信息进行分类，并将处于不同地理位置下的产品评论信息提供给所述评论分类模块；The geographic classification module classifies the user geographic location information extracted by the web crawler module, and provides product review information at different geographic locations to the review classification module; 所述评论分类模块对处于不同地理位置下的产品评论信息进行分类，得到不同情感倾向的分类，并将分类结果提供给所述流失预测模块；The review classification module classifies product review information in different geographic locations, obtains classifications of different emotional tendencies, and provides the classification results to the loss prediction module; 所述流失预测模块对不同情感倾向进行分析，并结合用户所处的地理位置提出预防用户流失的建议。The churn prediction module analyzes different emotional tendencies, and proposes suggestions for preventing user churn in combination with the user's geographic location. 9.根据权利要求8所述的基于区块链的电商分析系统，其特征在于：所述评论分类模块对处于不同地理位置下的产品评论信息进行分词，在此基础上使用TF-IDF模型对词语集进行数据挖掘得到关键字集，对关键字集分类后进行情感倾向的词典匹配，得到不同情感倾向的分类。9. The blockchain-based e-commerce analysis system according to claim 8, wherein the comment classification module performs word segmentation on product comment information in different geographic locations, and uses the TF-IDF model on this basis. Data mining is performed on the word set to obtain the keyword set, and after classifying the keyword set, the dictionary matching of sentiment tendency is performed to obtain the classification of different sentiment tendencies. 10.根据权利要求8所述的基于区块链的电商分析系统，其特征在于：所述地理分类模块根据不同的用户地理位置信息生成相应的分布图，并将分布图与评论数据关联，用于根据地理位置查找当地的用户评论信息。10. The blockchain-based e-commerce analysis system according to claim 8, wherein the geographic classification module generates a corresponding distribution map according to different user geographic location information, and associates the distribution map with the comment data, Used to find local user review information based on geographic location.
CN202011406667.3A 2020-12-04 2020-12-04 E-commerce analysis method and system based on block chain Pending CN112651768A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011406667.3A CN112651768A (en) 2020-12-04 2020-12-04 E-commerce analysis method and system based on block chain

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011406667.3A CN112651768A (en) 2020-12-04 2020-12-04 E-commerce analysis method and system based on block chain

Publications (1)

Publication Number Publication Date
CN112651768A true CN112651768A (en) 2021-04-13

Family

ID=75350570

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011406667.3A Pending CN112651768A (en) 2020-12-04 2020-12-04 E-commerce analysis method and system based on block chain

Country Status (1)

Country Link
CN (1) CN112651768A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113450153A (en) * 2021-07-02 2021-09-28 京东科技控股股份有限公司 Data processing method and device
CN115795040A (en) * 2023-02-10 2023-03-14 成都桉尼维尔信息科技有限公司 User portrait analysis method and system

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107943909A (en) * 2017-11-17 2018-04-20 合肥工业大学 User demand trend method for digging and device, storage medium based on comment data
CN108388660A (en) * 2018-03-08 2018-08-10 中国计量大学 A kind of improved electric business product pain spot analysis method
CN108446813A (en) * 2017-12-19 2018-08-24 清华大学 A kind of method of electric business service quality overall merit
CN108491377A (en) * 2018-03-06 2018-09-04 中国计量大学 A kind of electric business product comprehensive score method based on multi-dimension information fusion
CN109345266A (en) * 2018-09-18 2019-02-15 众安信息技术服务有限公司 Evaluation processing method and processing device based on block chain
KR20200022144A (en) * 2018-08-22 2020-03-03 울산과학기술원 System and Method for Analyzing Housing Market using Development of Emotion Dictionary
CN111767725A (en) * 2020-06-24 2020-10-13 中国平安财产保险股份有限公司 Data processing method and device based on emotion polarity analysis model
CN111813944A (en) * 2020-09-09 2020-10-23 北京神州泰岳智能数据技术有限公司 Live comment analysis method and device, electronic equipment and storage medium
CN112015721A (en) * 2020-09-14 2020-12-01 广州立信电子科技有限公司 E-commerce platform storage database optimization method based on big data

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107943909A (en) * 2017-11-17 2018-04-20 合肥工业大学 User demand trend method for digging and device, storage medium based on comment data
CN108446813A (en) * 2017-12-19 2018-08-24 清华大学 A kind of method of electric business service quality overall merit
CN108491377A (en) * 2018-03-06 2018-09-04 中国计量大学 A kind of electric business product comprehensive score method based on multi-dimension information fusion
CN108388660A (en) * 2018-03-08 2018-08-10 中国计量大学 A kind of improved electric business product pain spot analysis method
KR20200022144A (en) * 2018-08-22 2020-03-03 울산과학기술원 System and Method for Analyzing Housing Market using Development of Emotion Dictionary
CN109345266A (en) * 2018-09-18 2019-02-15 众安信息技术服务有限公司 Evaluation processing method and processing device based on block chain
CN111767725A (en) * 2020-06-24 2020-10-13 中国平安财产保险股份有限公司 Data processing method and device based on emotion polarity analysis model
CN111813944A (en) * 2020-09-09 2020-10-23 北京神州泰岳智能数据技术有限公司 Live comment analysis method and device, electronic equipment and storage medium
CN112015721A (en) * 2020-09-14 2020-12-01 广州立信电子科技有限公司 E-commerce platform storage database optimization method based on big data

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
朱晓霞;宋嘉欣;孟建芳;: "基于主题―情感挖掘模型的微博评论情感分类研究", 情报理论与实践, no. 05, 21 December 2018 (2018-12-21), pages 159 - 164 *
陈俊宇;郑列;: "基于R语言的商品评论情感可视化分析", 湖北工业大学学报, no. 01, 15 February 2020 (2020-02-15), pages 110 - 113 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113450153A (en) * 2021-07-02 2021-09-28 京东科技控股股份有限公司 Data processing method and device
CN115795040A (en) * 2023-02-10 2023-03-14 成都桉尼维尔信息科技有限公司 User portrait analysis method and system

Similar Documents

Publication Publication Date Title
Pawar et al. Twitter sentiment analysis: A review
Diaby et al. Taxonomy-based job recommender systems on Facebook and LinkedIn profiles
Kausar et al. Sentiment classification based on machine learning approaches in amazon product reviews
Sivanantham Sentiment analysis on social media for emotional prediction during COVID‐19 pandemic using efficient machine learning approach
Wu et al. Determining the factors affecting customer satisfaction using an extraction-based feature selection approach
CN112651768A (en) E-commerce analysis method and system based on block chain
Sharma et al. Integrated feature engineering based deep learning model for predicting customer’s review helpfulness
Bharathi et al. Sentiment classification of tourist’s opinion on tourist places of interest in South India using tweet reviews
Gollagi et al. Recursive feature elimination based multi-variate Naïve Bayes classification for product recommendation
Kundu et al. Personalized Word Recommendation System Using Sentiment Analysis
Karthikeyan et al. Sentimental analysis and classification of restaurant reviews
Usha et al. Sentiment analysis on business data using machine learning
Jishag et al. Automated review analyzing system using sentiment analysis
Szmydt Contextual personality-aware recommender system versus big data recommender system
Khan et al. A multi-model approach to aspect-based sentiment analysis of car reviews
Gajduk et al. Opinion mining of text documents written in Macedonian language
Wang et al. Prediction of Sensitive Consumer Behavior Based on Random Forest with Grid Search Cross-Validation
Hawladar et al. Amazon product reviews sentiment analysis using supervised learning algorithms
Lee Automatically learning user needs from online reviews for new product design
Yueyang et al. Detecting opinion polarities using ensemble of classification algorithms
Nag et al. Fusing Sentiment Analysis with Hybrid Collaborative Algorithms for Enhanced Recommender Systems
Ferilli et al. Towards sentiment and emotion analysis of user feedback for digital libraries
Sumathi et al. Opinion mining analysis in banking system using rough feature selection technique from social media text
Nandal A Hybrid Model for Review Analysis Using Deep Learning
Bhopale et al. Temporal topic modeling of scholarly publications for future trend forecasting

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20210413

RJ01 Rejection of invention patent application after publication