CN113449507A - Quality improvement method and device, electronic equipment and storage medium - Google Patents

Quality improvement method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN113449507A
CN113449507A CN202110775302.6A CN202110775302A CN113449507A CN 113449507 A CN113449507 A CN 113449507A CN 202110775302 A CN202110775302 A CN 202110775302A CN 113449507 A CN113449507 A CN 113449507A
Authority
CN
China
Prior art keywords
defect
importance
theme
demand
determining
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.)
Granted
Application number
CN202110775302.6A
Other languages
Chinese (zh)
Other versions
CN113449507B (en
Inventor
潘星
王辉雄
尤薇佳
张曼丽
蔡华利
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.)
Beihang University
Original Assignee
Beihang University
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 Beihang University filed Critical Beihang University
Priority to CN202110775302.6A priority Critical patent/CN113449507B/en
Publication of CN113449507A publication Critical patent/CN113449507A/en
Application granted granted Critical
Publication of CN113449507B publication Critical patent/CN113449507B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • 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
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • 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
    • 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
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • 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
    • G06Q30/0282—Rating or review of business operators or products
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

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

Abstract

The application provides a quality improvement method and device, an electronic device and a computer readable storage medium, wherein the method comprises the following steps: obtaining a plurality of customer demand characteristic words and a plurality of product defect characteristic words from a plurality of pieces of comment data; determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong; processing the comment data through order regression analysis, and determining a first importance degree of each demand theme and a second importance degree of each defect theme; determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters; a first number of corrective measures of greatest importance are taken as quality improvement measures. According to the scheme, the requirements and defects most concerned by customers can be mined from the comment data, and the improvement measures are determined according to the requirements and defects, so that quality improvement can be effectively achieved.

Description

Quality improvement method and device, electronic equipment and storage medium
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a quality improvement method and apparatus, an electronic device, and a computer-readable storage medium.
Background
As consumer demands for product quality increase, the emphasis of competition among products is gradually shifting from price to quality. Various quality defects inevitably exist during product design, manufacture and service. The explosive growth of e-commerce sales channels enables enterprises to timely learn about consumers' opinions about their products or services through online reviews of e-commerce platforms. The review data often includes the most significant needs and most objectionable deficiencies of consumers. How to utilize review data to mine information for improving quality is a difficult problem to be solved urgently.
Disclosure of Invention
It is an object of embodiments of the present application to provide a quality improvement method and apparatus, an electronic device, and a computer-readable storage medium for determining measures for improvement based on review data.
In one aspect, an embodiment of the present application provides a quality improvement method, including:
obtaining a plurality of customer demand characteristic words and a plurality of product defect characteristic words from a plurality of pieces of comment data;
determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong;
processing the comment data through order regression analysis, and determining a first importance degree of each demand theme and a second importance degree of each defect theme;
determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters;
a first number of corrective measures of greatest importance are taken as quality improvement measures.
In one embodiment, the obtaining a plurality of customer demand characteristic words and a plurality of product defect characteristic words from a plurality of pieces of comment data includes:
performing word segmentation processing on each piece of comment data to obtain a plurality of word segmentation units and the part of speech of each word segmentation unit;
screening out a plurality of word segmentation units with parts of speech as target parts of speech;
screening a plurality of word segmentation units corresponding to the target part of speech according to an Apriori algorithm to obtain an appointed word segmentation unit;
and dividing the appointed word segmentation unit into a plurality of customer demand characteristic words and a plurality of product defect characteristic words through a defect word list.
In an embodiment, before the filtering the word segmentation units corresponding to the target part of speech according to Apriori algorithm, the method further includes:
calculating evaluation parameters for a plurality of word segmentation units corresponding to the target part of speech through an evaluation function;
and screening a second number of word segmentation units with the maximum evaluation parameters to perform screening based on the Apriori algorithm.
In one embodiment, before the dividing the designated word segmentation unit into a plurality of customer requirement characteristic words and a plurality of product defect characteristic words through a defect word list, the method further comprises:
acquiring a plurality of conventional defect words and a plurality of specific defect words of the comment data indication product, and constructing an initial defect word list according to the conventional defect words and the specific defect words;
processing the plurality of pieces of comment data according to a word vector algorithm based on the initial defect word list to obtain a plurality of corpus defect words, and putting the corpus defect words into the initial defect word list to obtain the defect word list.
In one embodiment, the determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong includes:
searching a demand theme corresponding to each customer demand characteristic word in a preset demand theme configuration word bank to obtain a plurality of demand themes;
and searching a defect theme corresponding to each product defect characteristic word in a preset defect theme configuration word library to obtain a plurality of defect themes.
In one embodiment, the calculation parameters include a first relation matrix between the demand topic and the defect topic, a second relation matrix between the defect topic and a plurality of preset defect reasons, and a third relation matrix between the defect reasons and a plurality of preset corrective measures;
the determining the importance of a plurality of corrective measures according to the first importance of each demand topic, the second importance of each defect topic and the preset calculation parameters comprises:
determining a third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix;
determining a fourth importance degree of each defect reason according to the third importance degree of each defect theme and the second relation matrix;
and determining the importance of each corrective measure according to the fourth importance of each defect reason and the third relation matrix.
On the other hand, the embodiment of the present application further provides a quality improvement apparatus, including:
the acquisition module is used for acquiring a plurality of customer demand characteristic words and a plurality of product defect characteristic words from the plurality of pieces of comment data;
the first determining module is used for determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong;
the second determining module is used for processing the comment data through ordered regression analysis and determining the first importance of each demand theme and the second importance of each defect theme;
the third determining module is used for determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters;
an improvement module for taking a first number of corrective measures of greatest importance as quality improvement measures.
In one embodiment, the calculation parameters include a first relation matrix between the demand topic and the defect topic, a second relation matrix between the defect topic and a plurality of preset defect reasons, and a third relation matrix between the defect reasons and a plurality of preset corrective measures; the third determining module is further configured to:
determining a third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix;
determining a fourth importance degree of each defect reason according to the third importance degree of each defect theme and the second relation matrix;
and determining the importance of each corrective measure according to the fourth importance of each defect reason and the third relation matrix.
Further, an embodiment of the present application further provides an electronic device, where the electronic device includes:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the quality improvement method described above.
In addition, the present application also provides a computer-readable storage medium, which stores a computer program, and the computer program can be executed by a processor to implement the quality improvement method.
According to the technical scheme, a plurality of customer demand characteristic words and a plurality of product defect characteristic words are obtained from a plurality of comment data, a plurality of demand themes are determined according to the customer demand characteristic words, a plurality of defect subjects are determined according to the product defect characteristic words, after a first importance degree of each demand theme and a second importance degree of each defect theme are determined from the comment data through orderly regression analysis, the importance degrees of a plurality of preset correction measures can be calculated, and therefore the first number of correction measures with the largest importance degree are selected as quality improvement measures; through the measures, the requirements and the defects most concerned by customers can be mined from the comment data, and the improvement measures are determined according to the requirements and the defects, so that quality improvement can be effectively realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly described below.
Fig. 1 is a schematic view of an application scenario of a quality improvement method according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an electronic device according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of a quality improvement method provided by an embodiment of the present application;
fig. 4 is a schematic flowchart of a method for acquiring feature words according to an embodiment of the present application;
FIG. 5 is a flowchart illustrating a method for determining importance of corrective measures according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a defect report quality room according to an embodiment of the present application;
FIG. 7 is a schematic diagram of a defect cause analysis quality room according to an embodiment of the present disclosure;
FIG. 8 is a schematic diagram of a quality room for corrective action analysis provided by an embodiment of the present application;
fig. 9 is a block diagram of a quality improvement apparatus provided in an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
Like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Fig. 1 is a schematic application scenario diagram of a quality improvement method provided in an embodiment of the present application. As shown in fig. 1, the application scenario includes a server 20 and a server 30; the server 20 may be a server, a server cluster or a cloud computing center carrying a shopping website, and is configured to provide online comment data for a certain commodity or a certain service on the shopping website to the server 30; the server 30 may be a server, a server cluster, or a cloud computing center, and may determine quality improvement measures based on the review data.
As shown in fig. 2, the present embodiment provides an electronic apparatus 1 including: at least one processor 11 and a memory 12, one processor 11 being exemplified in fig. 2. The processor 11 and the memory 12 are connected by a bus 10, and the memory 12 stores instructions executable by the processor 11, and the instructions are executed by the processor 11, so that the electronic device 1 can execute all or part of the flow of the method in the embodiments described below. In an embodiment, the electronic device 1 may be the server 30 described above for performing the quality improvement method.
The Memory 12 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk.
The present application also provides a computer readable storage medium storing a computer program executable by a processor 11 to perform the quality improvement method provided by the present application.
Referring to fig. 3, a flow chart of a quality improvement method provided for an embodiment of the present application is schematically illustrated, and as shown in fig. 3, the method may include the following steps 310 to 350.
Step 310: and acquiring a plurality of customer demand characteristic words and a plurality of product defect characteristic words from the plurality of pieces of comment data.
The comment data can be online comments left after a consumer purchases a product on the e-commerce platform, and the comment data can comprise comment texts and comment scores. For example, for the comment data of the mobile phone, the comment text can be 'fine and fine in work, fine and smooth in screen, and very smooth in system', and the comment score can be five out of five.
The customer demand characteristic word may be a word for explaining the customer demand. For example, for electronic products, the customer demand characteristics words may include "price", "cheap", "discount", "handwriting", "recorder", "battery quality", "sound effect", "earphone", and the like.
The product defect feature words may be words used to describe product defects. For example, for an electronic product, the product defect feature words may include "leakage current", "dead halt", "blue screen", "unable to start", "poor heat dissipation", "poor manufacturing", "bad spot", and the like.
For a product, the server side can obtain a plurality of pieces of comment data of the product from the e-commerce platform, and obtain a plurality of customer demand characteristic words and a plurality of product defect characteristic words from comment texts of the comment data.
Step 320: and determining a plurality of demand subjects to which the plurality of customer demand characteristic words belong and a plurality of defect subjects to which the plurality of product defect characteristic words belong.
The demand topic is category information summarizing customer demand characteristic words, and different customer demand characteristic words can belong to the same demand topic. The defect theme is the category information summarizing the product defect characteristic words, and different product defect characteristic words can belong to the same defect theme.
The server side can judge the requirement theme to which each customer requirement characteristic word belongs, and therefore a plurality of requirement themes are obtained. The server side can judge the defect theme to which each product defect word belongs, and therefore a plurality of defect themes are obtained.
For example, for electronic products, the demand topic may include "price," "appearance," "function," "screen," "battery," "sound," "hardware," "software," "service," "accessory," and the like; the defect subjects may include "leakage", "crash", "failure", "no screen display", "no start", "no internet access", "poor screen", "poor work", "poor keyboard", "poor heat dissipation", "slow speed", and the like.
Step 330: and processing the comment data through the ordered regression analysis to determine the first importance of each demand theme and the second importance of each defect theme.
The server side can build a data set according to the occurrence frequency of customer demand characteristic words corresponding to each demand theme in the comment data, the occurrence frequency of product defect characteristic words corresponding to each defect theme, the emotion value corresponding to each customer demand characteristic word and comments in the comment data in a grading mode, and conduct order regression analysis on the data set to determine the first importance degree of each demand theme and the second importance degree of each defect theme.
The emotion value of the customer demand characteristic word can be determined according to the word of the customer demand characteristic word in the comment text context; if the vocabulary of the customer demand characteristic words in the context of the comment text appears in the preset positive emotion vocabulary, the emotion value of the customer demand characteristic words is 1 point; and if the vocabulary of the customer demand characteristic words in the comment text context appears in the preset negative emotion vocabulary, the emotion value of the customer demand characteristic words is-1.
Step 340: and determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters.
After determining the first importance of each demand topic and the second importance of each defect topic, the server may determine, according to the calculation parameter, the importance corresponding to a plurality of preset corrective measures. Here, the corrective action may be an improvement action for each stage of the production process of the product. Exemplary corrective actions may include "skill training enhancement", "motivation mechanism adoption", "periodic maintenance of equipment", "detection method effect enhancement", "production process supervision", "management using control charts", "management using 5S", "supervision enhancement", "reasonable sampling method formulation", "enterprise products", "competitive products", "improved products", and the like.
Step 350: a first number of corrective measures of greatest importance are taken as quality improvement measures.
After determining the importance corresponding to each corrective measure, the server may sort the plurality of corrective measures according to the order of the importance from large to small, and select the first number of corrective measures with the largest importance as the quality improvement measure. Wherein the first number may be a preconfigured empirical value.
Illustratively, after ranking the correction measures based on the importance, the server selects 3 correction measures with the largest importance as quality improvement measures, and can output the quality improvement measures to a user terminal (such as a mobile phone, a tablet computer, a computer, and the like), so as to prompt the user to improve the product quality based on the quality improvement measures.
Through the measures, the customer demand characteristic words and the product defect characteristic words for improving the quality can be mined from a large amount of comment data, and after the demand subjects to which the customer demand characteristic words belong and the defect subjects to which the product defect characteristic words belong are determined, the importance of each demand subject and the importance of each defect subject are determined from the comment data, so that the importance of each corrective measure can be determined, and a plurality of corrective measures with high importance are selected as quality improvement measures.
In an embodiment, referring to fig. 4, a flowchart of a method for acquiring a feature word provided in an embodiment of the present application is shown, as shown in fig. 4, when the server executes step 310, the server may execute the following steps 311 to 314.
Step 311: and performing word segmentation processing on each piece of comment data to obtain a plurality of word segmentation units and the part of speech of each word segmentation unit.
The server can perform word segmentation processing on the comment text in each piece of comment data through a word segmentation algorithm, so that a plurality of word segmentation units and the part of speech of the word segmentation units are obtained. Here, the word segmentation unit is the minimum result obtained by word segmentation. The word segmentation Algorithm may be BERT (proportional Encoder responses from transformations) + CRF (Conditional Random Field Algorithm).
Step 312: and screening a plurality of word segmentation units with parts of speech as target parts of speech.
Wherein the target part of speech is a part of speech closely related to customer requirements or product defects, and the target part of speech can be a combination of one or more of names, adjectives and verbs.
The service end can screen out the word segmentation units with the part of speech as the target part of speech, and filter the word segmentation units with other parts of speech.
Step 313: and screening a plurality of word segmentation units corresponding to the target part of speech according to an Apriori algorithm to obtain the appointed word segmentation units.
The server side can realize association mining based on an Apriori algorithm, and can generate rules from a transaction set meeting the preset minimum support degree to find a frequent item set. Here, the transaction set is a comment text set; the frequent item set is a word segmentation unit set. And the server screens out high-frequency word segmentation units capable of expressing key product characteristics from the frequent item set according to preset minimum support, minimum field support and independent support to form a key item set. The participle units in the key term set may be considered to be designated participle units.
Step 314: and dividing the appointed word segmentation unit into a plurality of customer demand characteristic words and a plurality of product defect characteristic words through a defect word list.
Wherein the defect vocabulary contains a plurality of words representing defects.
The server side can search each appointed word segmentation unit in the defect word list, and on one hand, if the appointed word segmentation unit is searched, the appointed word segmentation unit is a product defect characteristic word; on the other hand, if the appointed segmentation unit is not found, the appointed segmentation unit is the customer demand characteristic word.
Through the measures, the customer demand characteristic words and the product defect characteristic words can be extracted from the comment data.
In an embodiment, before the server filters the multiple word segmentation units corresponding to the target part-of-speech according to Apriori algorithm, the server may further calculate evaluation parameters for the multiple word segmentation units corresponding to the target part-of-speech through an evaluation function. Here, the evaluation function may be used to evaluate the degree of importance of the word segmentation unit in the comment data. Illustratively, the evaluation function may be represented by the following formula (1):
Figure BDA0003154163840000111
wherein, T represents a set of word segmentation units with part of speech as a target part of speech; d represents a set of comment texts in the comment data; t is tiRepresenting the ith word segmentation unit; g is a preset constant and can be 0.2; tf isijIs a word segmentation unit tiAppear in comment text djA probability of (1); dijRefers to comment text djLength of (d);
Figure BDA0003154163840000112
the average length of the comment texts in the set D is obtained; df is aiMeaning containing word-segmentation units tiThe number of comment texts; and I is the number of words in all comment texts.
After the evaluation parameters of the word segmentation units are calculated, the server can sort the word segmentation units according to the evaluation parameters, and screen out a second number of word segmentation units with the largest evaluation parameters. Here, the second number may be a pre-configured empirical value. For example, the second number may be 5000, and the server may select 5000 participle units with the largest evaluation parameter from all the participle units with the part of speech as the target part of speech.
The subsequent server may further filter the second number of word segmentation units based on Apriori algorithm.
Through the measures, the server side can screen all word segmentation units with the part of speech as the target part of speech, so that the subsequent calculation amount screened by an Apriori algorithm is reduced.
In an embodiment, the server may generate the defective vocabulary before dividing the specified word segmentation unit by the defective vocabulary.
The server side can obtain a plurality of conventional defect words and a plurality of specific defect words of the comment data indicating products. Here, the general defect word may be a general word describing a defect of a product; the specific defect word can be a word provided by an industry expert and capable of describing the defect of a specific part of the product more accurately. The server can form an initial defect word list according to a plurality of conventional defect words and a plurality of specific defect words.
The server side can process the comment data according to a word vector algorithm based on the initial defect word list to obtain a plurality of corpus defect words. The Word vector algorithm may be a Word2vec (Word to vector) algorithm. The server side can calculate word vectors for conventional defective words and specific defective words in the initial defective word list based on a word vector algorithm, a plurality of comment data are used as linguistic data and belong to the word vector algorithm, and new words similar to the calculated word vectors are identified from the comment data through the word vector algorithm and are used as linguistic data defective words. The server side can put the corpus defect words identified from the comment data into the initial defect word list, and therefore the defect word list is obtained.
In an embodiment, when the server executes step 320, the server may search a preset requirement theme configuration word library for a requirement theme corresponding to each customer requirement feature word, so as to obtain a plurality of requirement themes. The demand topic configuration word library comprises a plurality of demand topics and customer demand characteristic words corresponding to each demand topic. Referring to the following table 1, a thesaurus is configured for a requirement topic shown in the present application:
Figure BDA0003154163840000131
TABLE 1
For each customer demand characteristic word, the server side can search a demand theme corresponding to the customer demand characteristic word in the demand theme configuration word bank. After searching each customer demand characteristic word, all demand topics corresponding to all customer demand characteristic words can be determined. The requirement theme determined by the server side can be consistent with all requirement themes in the requirement theme configuration word bank, and can also be less than the requirement themes in the requirement theme configuration word bank.
The server side can search the defect theme corresponding to each product defect feature word in a preset defect theme configuration word bank, and therefore a plurality of defect themes are obtained. The defect topic configuration word library comprises a plurality of defect topics and product defect characteristic words corresponding to each defect topic. Referring to table 2 below, a thesaurus is configured for a defect topic shown in the present application:
Figure BDA0003154163840000132
Figure BDA0003154163840000141
TABLE 2
Aiming at each product defect feature word, the server side can search a defect topic corresponding to the product defect feature word in a defect topic configuration word library. After each product defect feature word is searched, all defect topics corresponding to all product defect feature words can be determined. The defect theme determined by the server side can be consistent with all defect themes in the defect theme configuration word bank, or can be less than the defect themes in the defect theme configuration word bank.
In one embodiment, the calculation parameters may include a first relation matrix between the requirement theme and the defect theme, a second relation matrix between the defect theme and a predetermined plurality of defect causes, and a third relation matrix between the defect causes and a predetermined plurality of corrective actions.
Referring to fig. 5, a flowchart of a method for determining importance of corrective measures according to an embodiment of the present application is shown, and as shown in fig. 5, the method includes the following steps 341 to 343.
Step 341: and determining the third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix.
The first relation matrix represents the degree of relation between each demand theme and each defect theme, and the degree of relation may include three types, namely "strong", "medium", and "weak".
The server can fill the first importance of each demand theme, the second importance of each defect main body and the relation degree between each demand theme and each defect main body in a preset defect report quality room, and further calculate the third importance of each defect theme.
For any defect topic, the server side can determine the weight coefficient of each demand topic under the defect topic according to the degree of relationship between the defect topic and each demand topic. Here, the weighting coefficients corresponding to different degrees of relationship may be configured in advance. Illustratively, the weight coefficient corresponding to the relationship degree "strong" is 3, the weight coefficient corresponding to the relationship degree "medium" is 2, and the weight coefficient corresponding to the relationship degree "strong" is 1. The server side can perform weighted summation on the first importance of the demand theme based on the weight coefficients of different demand themes, and multiply the weighted summation result with the second importance of the defect theme to obtain a third importance of the defect theme.
Referring to fig. 6, a schematic diagram of a defect report quality room according to an embodiment of the present disclosure is shown in fig. 6, where a left wall of the defect report quality room is filled with a first importance of each demand body, a ceiling is filled with a second importance of each defect body, and a column of "product defect importance" is a third importance calculated.
The calculation process is illustrated in fig. 6, where the weight coefficient corresponding to the relationship degree "strong" is 3, the weight coefficient corresponding to the relationship degree "medium" is 2, the weight coefficient corresponding to the relationship degree "strong" is 1, the relationship degree between the defect topic "electric leakage" and the requirement topic "appearance" is medium, and the corresponding weight coefficient is 2; the relation degree of the defect theme 'electric leakage' and the demand theme 'tone quality' is 'strong', and the corresponding weight coefficient is 3; the relation degree of the defect theme 'electric leakage' and the requirement theme 'hardware' is 'strong', and the corresponding weight coefficient is 3; the degree of relation between the defect theme 'leakage' and the demand theme 'service' is 'weak', and the corresponding weight coefficient is 1. The third significance of the defect topic "leaky" is (0.938 × 2+1.322 × 3+1.058 × 3+2.243 × 1) × 2.543 ═ 28.6.
Step 342: and determining the fourth importance of the defect reasons according to the third importance of each defect theme and the second relation matrix.
The second relation matrix represents the degree of relation between each defect topic and each defect reason, and the degree of relation may include three types of "strong", "medium", and "weak".
The server side can fill the third importance of each defect theme, the relation degree between each defect theme and each defect reason in the preset defect reason analysis quality room, and further calculate the fourth importance of each defect reason.
For any defect reason, the server side can determine the weight coefficient of each defect topic under the defect reason according to the degree of the relationship between the defect reason and each defect topic. Here, the weighting coefficients corresponding to different degrees of relationship may be configured in advance. Illustratively, the weight coefficient corresponding to the relationship degree "strong" is 3, the weight coefficient corresponding to the relationship degree "medium" is 2, and the weight coefficient corresponding to the relationship degree "strong" is 1. The server side can carry out weighted summation on the third importance of the defect theme based on the weight coefficients of different defect themes, so as to obtain a fourth importance of the defect reason.
Fig. 7 is a schematic diagram of a defect cause analysis quality room according to an embodiment of the present application, and as shown in fig. 7, a left wall of the defect cause analysis quality room is filled with a third importance of each defect topic, and a column of "defect cause importance" is a fourth importance calculated.
The calculation process is described with reference to fig. 7, where the weight coefficient corresponding to the degree of relationship "strong" is 3, the weight coefficient corresponding to the degree of relationship "medium" is 2, and the weight coefficient corresponding to the degree of relationship "strong" is 1. The relationship degrees between the defect cause "lack of expertise" and the defect subjects "leak electricity", "crash", "malfunction", "screen does not show", "power-off", "internet-access-impossible", "screen-poor", "poor-operation", "poor heat dissipation" are "strong", "medium", "weak", respectively, and in this case, the fourth importance of the defect cause "lack of expertise" is 28.6 + 3.5 + 3+16 + 2+34.4 + 1+25.1 + 3+21.5 + 2+15.9 + 1+20.9 + 1+9.3 ═ 345.1.
Step 343: and determining the importance of each corrective measure according to the fourth importance of each defect cause and the third relation matrix.
Wherein, the third relation matrix represents the relation degree between each defect reason and each correction measure, and the relation degree can include three types of "strong", "medium" and "weak".
The server can analyze the fourth importance of each defect reason and the relation degree between each defect reason and each correction measure in the quality room by presetting the correction measures, and further calculate the importance of each correction measure.
For any correction measure, the server may determine the weight coefficient of each defect cause under the correction measure according to the degree of relationship between the correction measure and each defect cause. Here, the weighting coefficients corresponding to different degrees of relationship may be configured in advance. Illustratively, the weight coefficient corresponding to the relationship degree "strong" is 3, the weight coefficient corresponding to the relationship degree "medium" is 2, and the weight coefficient corresponding to the relationship degree "strong" is 1. The server-side can carry out weighted summation on the fourth importance of the defect reason based on the weight coefficient which does not pass through the defect reason, thereby obtaining the importance of the corrective measure.
Fig. 8 is a schematic diagram of a corrective action analysis quality room according to an embodiment of the present application, where the fourth importance of each defect cause is filled in the left wall of the corrective action analysis quality room, and the column of "importance of corrective action" is the calculated importance of each corrective action.
The calculation process is described with reference to fig. 8, where the weight coefficient corresponding to the degree of relationship "strong" is 3, the weight coefficient corresponding to the degree of relationship "medium" is 2, and the weight coefficient corresponding to the degree of relationship "strong" is 1. The degrees of relationship between the correction measures "training on the enhanced skills" and the causes of defects "lack of professional skills", "operation is not standardized", "equipment aging", "process design is not perfect", "process parameters are not proper", "supervision work is not strict" are respectively "strong", "weak", "strong", "weak", and "weighted skill training" are 345.1 + 3+442 + 1+155.9 + 1+205.7 + 3+114.7 + 3+330.6 +1 ═ 2925.
Fig. 9 is a block diagram of a quality improvement apparatus according to an embodiment of the present invention, which may include, as shown in fig. 9:
an obtaining module 910, configured to obtain a plurality of customer demand feature words and a plurality of product defect feature words from a plurality of pieces of review data;
a first determining module 920, configured to determine a plurality of demand topics to which the plurality of customer demand feature words belong, and a plurality of defect topics to which the plurality of product defect feature words belong;
a second determining module 930, configured to process the comment data through an ordered regression analysis, and determine a first importance of each demand topic and a second importance of each defect topic;
a third determining module 940, configured to determine the importance of the plurality of corrective measures according to the first importance of each demand topic, the second importance of each defect topic, and a preset calculation parameter;
an improvement module 950 for taking the first number of corrective actions of greatest importance as quality improvement actions.
In an embodiment, the improving module 950 is further configured to:
determining a third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix;
determining a fourth importance degree of each defect reason according to the third importance degree of each defect theme and the second relation matrix;
and determining the importance of each corrective measure according to the fourth importance of each defect reason and the third relation matrix.
The implementation process of the functions and actions of each module in the above device is specifically described in the implementation process of the corresponding step in the above quality improvement method, and is not described herein again.
In the embodiments provided in the present application, the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

Claims (10)

1. A method of quality improvement, comprising:
obtaining a plurality of customer demand characteristic words and a plurality of product defect characteristic words from a plurality of pieces of comment data;
determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong;
processing the comment data through order regression analysis, and determining a first importance degree of each demand theme and a second importance degree of each defect theme;
determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters;
a first number of corrective measures of greatest importance are taken as quality improvement measures.
2. The method of claim 1, wherein obtaining a plurality of customer demand characteristic words and a plurality of product defect characteristic words from a plurality of pieces of review data comprises:
performing word segmentation processing on each piece of comment data to obtain a plurality of word segmentation units and the part of speech of each word segmentation unit;
screening out a plurality of word segmentation units with parts of speech as target parts of speech;
screening a plurality of word segmentation units corresponding to the target part of speech according to an Apriori algorithm to obtain an appointed word segmentation unit;
and dividing the appointed word segmentation unit into a plurality of customer demand characteristic words and a plurality of product defect characteristic words through a defect word list.
3. The method according to claim 2, wherein before the filtering the word segmentation units corresponding to the target part of speech according to Apriori algorithm, the method further comprises:
calculating evaluation parameters for a plurality of word segmentation units corresponding to the target part of speech through an evaluation function;
and screening a second number of word segmentation units with the maximum evaluation parameters to perform screening based on the Apriori algorithm.
4. The method of claim 2, wherein before said dividing said designated participle unit into a plurality of customer demand characteristic words and a plurality of product defect characteristic words by a defect vocabulary, said method further comprises:
acquiring a plurality of conventional defect words and a plurality of specific defect words of the comment data indication product, and constructing an initial defect word list according to the conventional defect words and the specific defect words;
processing the plurality of pieces of comment data according to a word vector algorithm based on the initial defect word list to obtain a plurality of corpus defect words, and putting the corpus defect words into the initial defect word list to obtain the defect word list.
5. The method of claim 1, wherein the determining a plurality of demand topics to which the plurality of customer demand feature words belong and a plurality of defect topics to which the plurality of product defect feature words belong comprises:
searching a demand theme corresponding to each customer demand characteristic word in a preset demand theme configuration word bank to obtain a plurality of demand themes;
and searching a defect theme corresponding to each product defect characteristic word in a preset defect theme configuration word library to obtain a plurality of defect themes.
6. The method of claim 1, wherein the calculation parameters comprise a first relationship matrix between the demand topic and the defect topic, a second relationship matrix between the defect topic and a predetermined plurality of defect causes, and a third relationship matrix between the defect causes and a predetermined plurality of corrective actions;
the determining the importance of a plurality of corrective measures according to the first importance of each demand topic, the second importance of each defect topic and the preset calculation parameters comprises:
determining a third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix;
determining a fourth importance degree of each defect reason according to the third importance degree of each defect theme and the second relation matrix;
and determining the importance of each corrective measure according to the fourth importance of each defect reason and the third relation matrix.
7. A quality improvement device, comprising:
the acquisition module is used for acquiring a plurality of customer demand characteristic words and a plurality of product defect characteristic words from the plurality of pieces of comment data;
the first determining module is used for determining a plurality of demand topics to which the plurality of customer demand characteristic words belong and a plurality of defect topics to which the plurality of product defect characteristic words belong;
the second determining module is used for processing the comment data through ordered regression analysis and determining the first importance of each demand theme and the second importance of each defect theme;
the third determining module is used for determining the importance of a plurality of corrective measures according to the first importance of each demand theme, the second importance of each defect theme and preset calculation parameters;
an improvement module for taking a first number of corrective measures of greatest importance as quality improvement measures.
8. The apparatus of claim 7, wherein the calculation parameters comprise a first relationship matrix between the demand topic and the defect topic, a second relationship matrix between the defect topic and a predetermined plurality of defect causes, and a third relationship matrix between the defect causes and a predetermined plurality of corrective actions; the third determining module is further configured to:
determining a third importance of each defect theme according to the first importance of each demand theme, the second importance of each defect theme and the first relation matrix;
determining a fourth importance degree of each defect reason according to the third importance degree of each defect theme and the second relation matrix;
and determining the importance of each corrective measure according to the fourth importance of each defect reason and the third relation matrix.
9. An electronic device, characterized in that the electronic device comprises:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the quality improvement method of any one of claims 1-6.
10. A computer-readable storage medium, characterized in that the storage medium stores a computer program executable by a processor to perform the quality improvement method of any one of claims 1-6.
CN202110775302.6A 2021-07-08 2021-07-08 Quality improvement method and device, electronic equipment and storage medium Active CN113449507B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110775302.6A CN113449507B (en) 2021-07-08 2021-07-08 Quality improvement method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110775302.6A CN113449507B (en) 2021-07-08 2021-07-08 Quality improvement method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN113449507A true CN113449507A (en) 2021-09-28
CN113449507B CN113449507B (en) 2023-02-17

Family

ID=77815556

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110775302.6A Active CN113449507B (en) 2021-07-08 2021-07-08 Quality improvement method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN113449507B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140278769A1 (en) * 2013-03-12 2014-09-18 Oracle International Corporation Method and system for performing trend analysis of themes in social data
CN110826323A (en) * 2019-10-24 2020-02-21 新华三信息安全技术有限公司 Comment information validity detection method and device
CN111258905A (en) * 2020-01-19 2020-06-09 中信银行股份有限公司 Defect positioning method and device, electronic equipment and computer readable storage medium
CN112417845A (en) * 2020-11-18 2021-02-26 武汉大学 Text evaluation method and device, electronic equipment and storage medium
CN113032554A (en) * 2019-12-24 2021-06-25 Tcl集团股份有限公司 Decision making system and computer readable storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140278769A1 (en) * 2013-03-12 2014-09-18 Oracle International Corporation Method and system for performing trend analysis of themes in social data
CN110826323A (en) * 2019-10-24 2020-02-21 新华三信息安全技术有限公司 Comment information validity detection method and device
CN113032554A (en) * 2019-12-24 2021-06-25 Tcl集团股份有限公司 Decision making system and computer readable storage medium
CN111258905A (en) * 2020-01-19 2020-06-09 中信银行股份有限公司 Defect positioning method and device, electronic equipment and computer readable storage medium
CN112417845A (en) * 2020-11-18 2021-02-26 武汉大学 Text evaluation method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN113449507B (en) 2023-02-17

Similar Documents

Publication Publication Date Title
US11645319B1 (en) Systems and methods for identifying issues in electronic documents
US11709875B2 (en) Prioritizing survey text responses
CN108320171B (en) Hot-sold commodity prediction method, system and device
US9117006B2 (en) Recommending keywords
US20050154692A1 (en) Predictive selection of content transformation in predictive modeling systems
CN104933081A (en) Search suggestion providing method and apparatus
CN111639247A (en) Method, apparatus, device and computer-readable storage medium for evaluating quality of review
US20220138770A1 (en) Method and apparatus for analyzing sales conversation based on voice recognition
US20040267554A1 (en) Methods and systems for semi-automated job requisition
US20180218427A1 (en) Recommendation integrated online digital sales service chat system
CN110706028A (en) Commodity evaluation emotion analysis system based on attribute characteristics
CN109947902A (en) A kind of data query method, apparatus and readable medium
Irawaty et al. Vectorizer comparison for sentiment analysis on social media youtube: A case study
CN110489531B (en) Method and device for determining high-frequency problem
CN116579351B (en) Analysis method and device for user evaluation information
CN115577890A (en) Intelligent quality management method, electronic device and storage medium
CN113449507B (en) Quality improvement method and device, electronic equipment and storage medium
CN116151840A (en) User service data intelligent management system and method based on big data
CN112445908A (en) Commodity comment information display method and device, electronic equipment and storage medium
CN111859946A (en) Method and device for ranking comments and machine-readable storage medium
CN115660451A (en) Supplier risk early warning method, device, equipment and medium based on RPA
CN112200602B (en) Neural network model training method and device for advertisement recommendation
CN110795537B (en) Method, device, equipment and medium for determining improvement strategy of target commodity
CN110929123A (en) E-commerce product competition analysis method and system
Das et al. Aspect aware optimized opinion analysis of online product reviews

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
GR01 Patent grant
GR01 Patent grant