CN108764617A - A kind of enterprises under network environment reputation evaluation method - Google Patents

A kind of enterprises under network environment reputation evaluation method Download PDF

Info

Publication number
CN108764617A
CN108764617A CN201810336472.2A CN201810336472A CN108764617A CN 108764617 A CN108764617 A CN 108764617A CN 201810336472 A CN201810336472 A CN 201810336472A CN 108764617 A CN108764617 A CN 108764617A
Authority
CN
China
Prior art keywords
enterprise
evaluated
public opinion
data
score
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
CN201810336472.2A
Other languages
Chinese (zh)
Inventor
牛振东
陈杰
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.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
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 Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN201810336472.2A priority Critical patent/CN108764617A/en
Publication of CN108764617A publication Critical patent/CN108764617A/en
Pending legal-status Critical Current

Links

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
    • 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
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The present invention relates to the corporate reputation evaluation methods under a kind of network environment, belong to computer application technology.Concrete operation step is：Step 1: determining enterprise to be evaluated.Step 2: obtaining net matchmaker's image index score of enterprise to be evaluated.Step 3: obtaining the talent input index score of enterprise to be evaluated.Step 4: obtaining enterprise's public opinion evaluation index score of enterprise to be evaluated.Step 5: by corporate reputation evaluation model, the corporate reputation evaluation score value of enterprise to be evaluated is obtained.Step 6: according to corporate reputation score, the reputation of enterprise to be evaluated is evaluated.Corporate reputation evaluation method under network environment proposed by the present invention is compared with the prior art, the method of the present invention keeps the reputation of comment enterprise more comprehensive, official's data and netizen's content-data can be integrated, and to evaluate corporate reputation on a large scale, the performance of the reputation situation and Target Enterprise of different enterprises under each dimension is fully understanded convenient for user.

Description

A kind of enterprises under network environment reputation evaluation method
Technical field
The present invention relates to the corporate reputation evaluation methods under a kind of network environment, belong to computer application technology.
Background technology
Corporate reputation under network environment is that the internet that enterprise obtains in market management activity and public opinion gives Generally acknowledged prestige and reputation is stakeholder (including public organization, supplier, customer, investor, the rival of enterprise And enterprises employee) formed and what is grown up a kind of recognizing the Enterprise Integrated impression constituted by emotional responses and rationality.
Traditional corporate reputation evaluation is the degree of understanding based on industry field expert to enterprise mostly, from the management of enterprise Horizontal, product wind comments situation, corporate tax information etc. to carry out comprehensive marking.Such methods are in face of comprehensive enterprise When, expertise will be not enough to accurate comprehensive measurement enterprise.Meanwhile expert analysis mode usually requires the investigation and analysis of long period Can just obtain more objective and fair as a result, not being suitable for the scene to big data on internet.In addition, under network environment with enterprise The relevant data volume of industry is big, manually extracts and to arrange information work amount huge, it is difficult within the reasonable time to the information of enterprise into Row analysis.Therefore, the evaluation method of tradition dependence domain expert is not suitable for evaluating the corporate reputation under network environment.
Currently, in having document, there is not the related record for carrying out evaluation method under network environment to corporate reputation also.
Invention content
The purpose of the present invention is to propose to the corporate reputation evaluation methods under a kind of network environment.It can be calculated by this method Go out to be more in line with the evaluation result of internet environment than relying on domain expert's method.
The purpose of the present invention is what is be achieved through the following technical solutions.
Corporate reputation evaluation method under a kind of network environment of the present invention, concrete operation step are：
Step 1: determining enterprise to be evaluated.
Step 2: obtaining net matchmaker's image index score of enterprise to be evaluated.
On the basis of step 1 operates, news data related with enterprise to be evaluated is obtained from webpage, in conjunction with network News evaluation index item obtains enterprise network matchmaker's image index score.Specially：
Step 2.1：According to enterprise name to be evaluated, principal products of business and leader's information, determines enterprise's keyword, pass through Enterprise's keyword obtains news data related with enterprise to be evaluated from webpage.It is described related with enterprise to be evaluated The quantity of news data indicates with symbol m, m >=100.
Step 2.2：It is made a summary to news data described in step 2.1 using autoabstract technology, obtains summary data.
Step 2.3：According to Internet news evaluation index, the summary data obtained to step 2.2 is for statistical analysis, obtains The score of each Internet news evaluation index.The Internet news evaluation index is as shown in table 1.
1 Internet news evaluation index table of table
Step 3: obtaining the talent input index score of enterprise to be evaluated.
On the basis of step 1 operates, is obtained from webpage and once took office in or now took office in the letter of enterprise personnel to be evaluated It counts evidence one by one, in conjunction with enterprise's talent's input-occupancy-output analysis, obtains enterprise talent input-occupancy-output analysis score.Specially：
Step 3.1：According to enterprise name to be evaluated, enterprise name keyword is determined, by the enterprise name keyword, Personnel's resume data related with enterprise to be evaluated are obtained from webpage.Personnel's resume data related with enterprise to be evaluated Quantity indicated with symbol n, n >=100.
Step 3.2：According to enterprise's talent's input-occupancy-output analysis, pair personnel's resume data related with enterprise to be evaluated count Analysis, obtains each enterprise talent input-occupancy-output analysis score.The enterprise talent input-occupancy-output analysis is specifically as shown in table 2.
2 enterprise talent input-occupancy-output analysis table of table
Serial number Enterprise's talent's input-occupancy-output analysis Value range
1 Employee's education composition ratio [0,1]
2 Employee's previous graduate college composition [0,1]
3 Male to female ratio in employee [0,1]
4 Employee's average operation time limit [0,1]
5 Employee's average age Positive real number
6 The average salary of employee Positive real number
7 The ratio of direct competitive opponent enterprise is gone in labor turnover [0,1]
8 The patent number that enterprise possesses Positive integer
9 The copyright number that enterprise possesses Positive integer
10 The national major project number that enterprise participates in Positive integer
11 The national science fund number of Enterprise Application Positive integer
Step 4: obtaining enterprise's public opinion evaluation index score of enterprise to be evaluated.
On the basis of step 1 operates, public opinion data related with enterprise to be evaluated are obtained from webpage, in conjunction with enterprise The confidence level of public opinion evaluation index and network public opinion data obtains enterprise's public opinion evaluation index score.The public opinion data packet Include microblogging and comment.Concrete operations are：
Step 4.1：By enterprise's keyword described in step 2.1, network related with enterprise to be evaluated is obtained from webpage Public opinion data.The quantity of the network public opinion data related with enterprise to be evaluated indicates with symbol p, p >=100.
Step 4.2：The confidence level of the network public opinion data is calculated using the confidence level computing technique of comment, is obtained To the confidence level of the network public opinion data.
Step 4.2.1：Duplicate removal processing is carried out to the network public opinion data.Specially：Using natural language processing technique In local salted hash Salted to network public opinion data carry out Hash calculation, if the cryptographic Hash of two network public opinion data is identical, It is judged as repeating speech, weeds out.
Step 4.2.2：Author's personal information and other users according to network public opinion data is to the network public opinion data The quantity of forwarding or comment, calculates the confidence level of network public opinion data.Specific method is：If the work of network public opinion data The registion time of person is less than 1 week or the overall network public opinion data number of publication is less than 50, then the author of the network public opinion data Confidence level is relatively low, and value is [0,0.3]；Otherwise, value is [0.7,1]；Emotion point is carried out for the content of network public opinion data Analysis, the emotion tendentiousness of text if negative content and in forwarding and commenting on is inconsistent, then the feelings of the network public opinion data It is relatively low to feel confidence level, value [0,0.3]；Otherwise, value [0.7,1].Using the product of author's confidence level and emotion confidence level as The confidence level of the network public opinion data.
Step 4.3：It is for statistical analysis to the network public opinion data according to enterprise's public opinion evaluation index, obtain each The statistical value of enterprise's public opinion evaluation index.Enterprise's public opinion evaluation index is as shown in table 3.
3 enterprise's public opinion evaluation index table of table
Serial number Enterprise's public opinion evaluation index Value range
1 Exposure rate of the enterprise in microblogging [0,1]
2 Microblogging ratio in front in microblogging [0,1]
3 Negative microblogging ratio in microblogging [0,1]
4 With the relevant front microblog topic ratio of enterprise [0,1]
5 With the relevant negative microblog topic ratio of enterprise [0,1]
6 The ratio of working strength involved in microblogging [0,1]
7 The ratio of the microblogging of Enterprise Public Relations event [0,1]
8 With the ratio of enterprise relevant forum's note [0,1]
9 With the ratio of enterprise relevant positive forum's note [0,1]
10 With the ratio of enterprise relevant negative forum's note [0,1]
11 The negative money order receipt to be signed and returned to the sender ratio of social benefit about employee [0,1]
12 Negative money order receipt to be signed and returned to the sender ratio about training and Growth Development [0,1]
13 About the negative money order receipt to be signed and returned to the sender ratio for promoting space [0,1]
14 Negative money order receipt to be signed and returned to the sender ratio about job stability [0,1]
15 Negative money order receipt to be signed and returned to the sender ratio about work climate [0,1]
Step 4.4：The score for each enterprise's public opinion evaluation index that step 4.3 is obtained, is multiplied by what step 4.2 obtained The confidence level of network public opinion data obtains each enterprise's public opinion evaluation index score of enterprise to be evaluated.
Step 5: by corporate reputation evaluation model, the corporate reputation evaluation score value of enterprise to be evaluated is obtained.
On the basis of step 2 to step 4 operates, by the corporate reputation evaluation model as shown in formula (1), obtain The corporate reputation score of enterprise to be evaluated, is indicated with symbol Y.Corporate reputation score Y is higher, indicates that the reputation of enterprise is better.
Wherein, aiIndicate the corresponding coefficient of i-th Internet news evaluation index, aiFor artificial setting value, i ∈ [1,13]； AiIndicate i-th Internet news evaluation index score；bjIndicate the corresponding coefficient of jXiang enterprises talent's input-occupancy-output analysis, bjIt is artificial Setting value, j ∈ [1,11]；BjIndicate jXiang enterprise talent input-occupancy-output analysis scores； ckIndicate kXiang enterprises public opinion evaluation index Corresponding coefficient, ckFor artificial setting value, k ∈ [1,15]；CkIndicate kXiang enterprises public opinion evaluation index score.
Step 6: according to corporate reputation score, the reputation of enterprise to be evaluated is evaluated.
On the basis of step 5 operates, reputation evaluation is carried out to enterprise to be evaluated according to table 4.
4 corporate reputation evaluation table of table
Serial number Corporate reputation score Y Reputation is evaluated
1 Y≥1.50 A
2 Y∈[1.00,1.50) B
3 Y∈[0.50,1) C
4 Y<0.50 D
Operation through the above steps, you can obtain the reputation evaluation of enterprise to be evaluated.
Advantageous effect
Compared with the prior art, the method for the present invention makes to comment corporate reputation evaluation method under network environment proposed by the present invention It is more comprehensive by the reputation of enterprise, official's data and netizen's content-data can be integrated, and make extensive evaluation enterprise sound Reputation is possibly realized, and the performance of the reputation situation and Target Enterprise of different enterprises under each dimension is fully understanded convenient for user.
Description of the drawings
Fig. 1 is the operational flowchart of the corporate reputation evaluation method under network environment in the specific embodiment of the invention.
Specific implementation mode
According to above-mentioned technical proposal, the present invention is described in detail with embodiment below in conjunction with the accompanying drawings.
Reputation evaluation is carried out to enterprise to be evaluated using the corporate reputation evaluation method under network environment proposed by the present invention, Its operating process is as shown in Figure 1, concrete operation step is as follows：
Step 1: determining enterprise to be evaluated.
Enterprise to be evaluated is " Beijing network technology Co., Ltd ".Time range：September 14 days to 2018 3 in 2017 The moon 14.
Step 2: obtaining net matchmaker's image index score of enterprise to be evaluated.
On the basis of step 1 operates, news data related with enterprise to be evaluated is obtained from webpage, in conjunction with network News evaluation index item obtains enterprise network matchmaker's image index score.Specially：
Step 2.1：According to enterprise name to be evaluated, principal products of business and leader's information, enterprise's keyword is determined.It is to be evaluated The name keyword of valence enterprise is " Beijing network technology Co., Ltd " and " Hua Pinbo is farsighted "；Product keyword is that " Boss is straight Engage " and " being certain about net "；Administrative staff's keyword is " Zhao " and " Tang so-and-so ".By enterprise's keyword, searched using Baidu Rope obtains 2964 news datas related with enterprise to be evaluated, i.e. m=2964.
Step 2.2：It is made a summary to news data described in step 2.1 using autoabstract technology, obtains summary data.
Step 2.3：According to Internet news evaluation index, the summary data obtained to step 2.2 is for statistical analysis, obtains The score of each Internet news evaluation index, as shown in table 5.
5 Internet news evaluation index Score Lists of table
Step 3: obtaining the talent input index score of enterprise to be evaluated.
On the basis of step 1 operates, is obtained from webpage and once took office in or now took office in the letter of enterprise personnel to be evaluated It counts evidence one by one, in conjunction with enterprise's talent's input-occupancy-output analysis, obtains enterprise talent input-occupancy-output analysis score.Specially：
Step 3.1：According to enterprise name to be evaluated, enterprise name keyword is determined, by the enterprise name keyword, Personnel's resume data related with enterprise to be evaluated are obtained from webpage, obtain the resume of 113 employees of the said firm altogether, wherein 84 male employees, 29 female employees, leave office 15 employees, and 2 people are at present in rival company.Beijing worker in 2017 is average 7706 yuan/month of wage.The quantity n=113 of personnel's resume data related with enterprise to be evaluated.
Step 3.2：According to enterprise's talent's input-occupancy-output analysis, pair personnel's resume data related with enterprise to be evaluated count Analysis, obtains each enterprise talent input-occupancy-output analysis score, as shown in table 6.
6 enterprise talent input-occupancy-output analysis Score Lists of table
Serial number Enterprise's talent's input-occupancy-output analysis Value range
1 Employee's education composition ratio 37/113
2 Employee's previous graduate college composition 29/113
3 Male to female ratio in employee 2.89
4 Employee's average operation time limit 3.7
5 Employee's average age 28
6 The average salary of employee 12440
7 The ratio of direct competitive opponent enterprise is gone in labor turnover 0.13
8 The patent number that enterprise possesses 0
9 The copyright number that enterprise possesses 0
10 The national major project number that enterprise participates in 0
11 The national science fund number of Enterprise Application 0
Wherein, it is 7706 according to the average salary that Beijing announces for 2017, the average salary of employee is scored at 12440/ 7706 be 1.61.
Step 4: obtaining enterprise's public opinion evaluation index score of enterprise to be evaluated.
On the basis of step 1 operates, public opinion data related with enterprise to be evaluated are obtained from webpage, in conjunction with enterprise The confidence level of public opinion evaluation index and network public opinion data obtains enterprise's public opinion evaluation index score.The public opinion data packet Include microblogging and comment.Concrete operations are：
Step 4.1：By enterprise's keyword described in step 2.1, network related with enterprise to be evaluated is obtained from webpage Public opinion data.
3000 microblog datas obtained in Sina weibo are analyzed, the data as the 1 to 7th index in table 3 Source.Since the keyword of the enterprise is in forum of portal website (community of Sohu, ends of the earth forum, forum of Sina, forum of Netease) Related money order receipt to be signed and returned to the sender number is very little, therefore is not counted in range of value with forum money order receipt to be signed and returned to the sender index of correlation (7-15) in table 3.
Step 4.2：The confidence level of the network public opinion data is calculated using the confidence level computing technique of comment, is obtained To the confidence level of the network public opinion data.Specially：
Step 4.2.1：Duplicate removal processing is carried out to the network public opinion data.Specially：Using natural language processing technique In local salted hash Salted to network public opinion data carry out Hash calculation, if the cryptographic Hash of two network public opinion data is identical, It is judged as repeating speech, weeds out.
Step 4.2.2：Author's personal information and other users according to network public opinion data is to the network public opinion data The quantity of forwarding or comment, calculates the confidence level of network public opinion data.Specific method is：If the work of network public opinion data The registion time of person is less than 1 week or total microblog number is less than 50, then author's confidence level of the network public opinion data is relatively low, value It is 0.3；Otherwise, value 0.7；Sentiment analysis is carried out for content of microblog, if negative content and in forwarding and commenting on Emotion tendentiousness of text is inconsistent, then the emotion confidence level of the microblogging is relatively low, value 0.3；Otherwise, value 0.9.Author is credible Confidence level of the product of degree and emotion confidence level as the network public opinion data.Step 4.3：Referred to according to enterprise's public opinion evaluation Mark, it is for statistical analysis to the network public opinion data, obtain the statistical value of each enterprise's public opinion evaluation index.The enterprise carriage It is as shown in table 3 by evaluation index.
Step 4.2.2：Author's personal information and other users according to network public opinion data is to the network public opinion data The quantity of forwarding or comment, calculates the confidence level of network public opinion data.Specific method is：If the work of network public opinion data The registion time of person is less than 1 week or the overall network public opinion data number of publication is less than 50, then the author of the network public opinion data Confidence level is relatively low, value 0.3；Otherwise, value 0.7；Sentiment analysis is carried out for the content of network public opinion data, if negative Face content and forward and comment in emotion tendentiousness of text it is inconsistent, then the emotion confidence level of the network public opinion data compared with It is low, value 0.3；Otherwise, value 0.9.Using the product of author's confidence level and emotion confidence level as the network public opinion data Confidence level.
Step 4.3：It is for statistical analysis to the network public opinion data according to enterprise's public opinion evaluation index, obtain each The statistical value of enterprise's public opinion evaluation index, as shown in table 7.
7 enterprise's public opinion evaluation index Score Lists of table
Serial number Enterprise's public opinion evaluation index Value range
1 Exposure rate of the enterprise in microblogging 2656/3000
2 Microblogging ratio in front in microblogging 1439/3000
3 Negative microblogging ratio in microblogging 347/3000
4 With the relevant front microblog topic ratio of enterprise 7/39
5 With the relevant negative microblog topic ratio of enterprise 4/39
6 The ratio of working strength involved in microblogging 33/3000
7 The ratio of the microblogging of Enterprise Public Relations event 271/3000
8 With the ratio of enterprise relevant forum's note 0
9 With the ratio of enterprise relevant positive forum's note 0
10 With the ratio of enterprise relevant negative forum's note 0
11 The negative money order receipt to be signed and returned to the sender ratio of social benefit about employee 0
12 Negative money order receipt to be signed and returned to the sender ratio about training and Growth Development 0
13 About the negative money order receipt to be signed and returned to the sender ratio for promoting space 0
14 Negative money order receipt to be signed and returned to the sender ratio about job stability 0
15 Negative money order receipt to be signed and returned to the sender ratio about work climate 0
Step 4.4：The score for each enterprise's public opinion evaluation index that step 4.3 is obtained, is multiplied by what step 4.2 obtained The confidence level of network public opinion data obtains each enterprise's public opinion evaluation index score of enterprise to be evaluated.
Step 5: by corporate reputation evaluation model, the corporate reputation evaluation score value of enterprise to be evaluated is obtained.
On the basis of step 2 to step 4 operates, by the corporate reputation evaluation model as shown in formula (1), obtain The corporate reputation score of enterprise to be evaluated, is indicated with symbol Y.Corporate reputation score Y is higher, indicates that the reputation of enterprise is better.
Wherein, aiIndicate the corresponding coefficient of i-th Internet news evaluation index, aiFor artificial setting value, i ∈ [1,13]； AiIndicate i-th Internet news evaluation index score；bjIndicate the corresponding coefficient of jXiang enterprises talent's input-occupancy-output analysis, bjIt is artificial Setting value, j ∈ [1,11]；BjIndicate jXiang enterprise talent input-occupancy-output analysis scores； ckIndicate kXiang enterprises public opinion evaluation index Corresponding coefficient, ckFor artificial setting value, k ∈ [1,15]；CkIndicate kXiang enterprises public opinion evaluation index score.
Wherein：
[a1,a2,…,a13]=[0.01,0.11,0.02,0.27,0.03,0.28,0.03,0.13,0.11,0.08, 0.08,0.11,0.12,0.08,0.07]；
[b1,b2,…,b13]=[0.03,0.15,0.01,0.04,0.02,0.05,0.14,0.06,0.04,0.11, 0.13]；
[c1,c2,…,c13]=[0.03,0.05,0.07,0.06,0.17,0.21,0.18,0.03,0.02,0.04, 0.10,0.02,0.17,0.11,0.05]。
Above-mentioned parameter is brought into score of the enterprise in corresponding evaluation index in formula (1), Y=can be obtained 0.784.Therefore, corporate reputation to be evaluated evaluation is scored at 0.784 in the present embodiment.
Step 6: according to corporate reputation score, the reputation of enterprise to be evaluated is evaluated.
On the basis of step 5 operates, reputation evaluation is carried out to enterprise to be evaluated according to table 4.
4 corporate reputation evaluation table of table
Serial number Corporate reputation score Y Reputation is evaluated
1 Y≥1.50 A
2 Y∈[1.00,1.50) B
3 Y∈[0.50,1) C
4 Y<0.50 D
Operation through the above steps, you can the reputation for obtaining enterprise to be evaluated is evaluated as C.

Claims (2)

1. the corporate reputation evaluation method under a kind of network environment, it is characterised in that：Its concrete operation step is：
Step 1: determining enterprise to be evaluated；
Step 2: obtaining net matchmaker's image index score of enterprise to be evaluated；
On the basis of step 1 operates, news data related with enterprise to be evaluated is obtained from webpage, in conjunction with Internet news Evaluation index item obtains enterprise network matchmaker's image index score；Specially：
Step 2.1：According to enterprise name to be evaluated, principal products of business and leader's information, enterprise's keyword is determined, by described Enterprise's keyword obtains news data related with enterprise to be evaluated from webpage；The news related with enterprise to be evaluated The quantity of data indicates with symbol m, m >=100；
Step 2.2：It is made a summary to news data described in step 2.1 using autoabstract technology, obtains summary data；
Step 2.3：According to Internet news evaluation index, the summary data obtained to step 2.2 is for statistical analysis, obtains each The score of Internet news evaluation index；The Internet news evaluation index is as shown in table 1；
1 Internet news evaluation index table of table
Step 3: obtaining the talent input index score of enterprise to be evaluated；
On the basis of step 1 operates, is obtained from webpage and once took office in or now took office in the resume number of enterprise personnel to be evaluated According in conjunction with enterprise's talent's input-occupancy-output analysis, obtaining enterprise talent input-occupancy-output analysis score；Specially：
Step 3.1：According to enterprise name to be evaluated, enterprise name keyword is determined, by the enterprise name keyword, from net Personnel's resume data related with enterprise to be evaluated are obtained in page；The number of personnel's resume data related with enterprise to be evaluated Amount indicates with symbol n, n >=100；
Step 3.2：According to enterprise's talent's input-occupancy-output analysis, pair personnel's resume data related with enterprise to be evaluated carry out statistical Analysis, obtains each enterprise talent input-occupancy-output analysis score；The enterprise talent input-occupancy-output analysis is specifically as shown in table 2；
2 enterprise talent input-occupancy-output analysis table of table
Serial number Enterprise's talent's input-occupancy-output analysis Value range 1 Employee's education composition ratio [0,1] 2 Employee's previous graduate college composition [0,1] 3 Male to female ratio in employee [0,1] 4 Employee's average operation time limit [0,1] 5 Employee's average age Positive real number 6 The average salary of employee Positive real number 7 The ratio of direct competitive opponent enterprise is gone in labor turnover [0,1] 8 The patent number that enterprise possesses Positive integer 9 The copyright number that enterprise possesses Positive integer 10 The national major project number that enterprise participates in Positive integer 11 The national science fund number of Enterprise Application Positive integer
Step 4: obtaining enterprise's public opinion evaluation index score of enterprise to be evaluated；
On the basis of step 1 operates, public opinion data related with enterprise to be evaluated are obtained from webpage, in conjunction with enterprise's public opinion The confidence level of evaluation index and network public opinion data obtains enterprise's public opinion evaluation index score；The public opinion data include micro- Rich and comment；Concrete operations are：
Step 4.1：By enterprise's keyword described in step 2.1, network public opinion related with enterprise to be evaluated is obtained from webpage Data；The quantity of the network public opinion data related with enterprise to be evaluated indicates with symbol p, p >=100；
Step 4.2：The confidence level of the network public opinion data is calculated using the confidence level computing technique of comment, obtains institute State the confidence level of network public opinion data；
Step 4.3：It is for statistical analysis to the network public opinion data according to enterprise's public opinion evaluation index, obtain each enterprise The statistical value of public opinion evaluation index；Enterprise's public opinion evaluation index is as shown in table 3；
3 enterprise's public opinion evaluation index table of table
Serial number Enterprise's public opinion evaluation index Value range 1 Exposure rate of the enterprise in microblogging [0,1] 2 Microblogging ratio in front in microblogging [0,1] 3 Negative microblogging ratio in microblogging [0,1] 4 With the relevant front microblog topic ratio of enterprise [0,1] 5 With the relevant negative microblog topic ratio of enterprise [0,1] 6 The ratio of working strength involved in microblogging [0,1] 7 The ratio of the microblogging of Enterprise Public Relations event [0,1] 8 With the ratio of enterprise relevant forum's note [0,1] 9 With the ratio of enterprise relevant positive forum's note [0,1] 10 With the ratio of enterprise relevant negative forum's note [0,1] 11 The negative money order receipt to be signed and returned to the sender ratio of social benefit about employee [0,1] 12 Negative money order receipt to be signed and returned to the sender ratio about training and Growth Development [0,1] 13 About the negative money order receipt to be signed and returned to the sender ratio for promoting space [0,1] 14 Negative money order receipt to be signed and returned to the sender ratio about job stability [0,1] 15 Negative money order receipt to be signed and returned to the sender ratio about work climate [0,1]
Step 4.4：The score for each enterprise's public opinion evaluation index that step 4.3 is obtained is multiplied by the network carriage that step 4.2 obtains By the confidence level of data, each enterprise's public opinion evaluation index score of enterprise to be evaluated is obtained；
Step 5: by corporate reputation evaluation model, the corporate reputation evaluation score value of enterprise to be evaluated is obtained；
On the basis of step 2 to step 4 operates, by the corporate reputation evaluation model as shown in formula (1), obtain to be evaluated The corporate reputation score of valence enterprise, is indicated with symbol Y；Corporate reputation score Y is higher, indicates that the reputation of enterprise is better；
Wherein, aiIndicate the corresponding coefficient of i-th Internet news evaluation index, aiFor artificial setting value, i ∈ [1,13]；AiIt indicates I-th Internet news evaluation index score；bjIndicate the corresponding coefficient of jXiang enterprises talent's input-occupancy-output analysis, bjTo be manually set Value, j ∈ [1,11]；BjIndicate jXiang enterprise talent input-occupancy-output analysis scores；ckIndicate that kXiang enterprises public opinion evaluation index is corresponding Coefficient, ckFor artificial setting value, k ∈ [1,15]；CkIndicate kXiang enterprises public opinion evaluation index score；
Step 6: according to corporate reputation score, the reputation of enterprise to be evaluated is evaluated；
On the basis of step 5 operates, reputation evaluation is carried out to enterprise to be evaluated according to table 4；
4 corporate reputation evaluation table of table
Serial number Corporate reputation score Y Reputation is evaluated 1 Y≥1.50 A 2 Y∈[1.00,1.50) B 3 Y∈[0.50,1) C 4 Y < 0.50 D
Operation through the above steps, you can obtain the reputation evaluation of enterprise to be evaluated.
2. the corporate reputation evaluation method under a kind of network environment as described in claim 1, it is characterised in that：In step 4.2 The confidence level computing technique using comment calculates the confidence level of the network public opinion data, obtains the network carriage Concrete operation step by the confidence level of data is：
Step 4.2.1：Duplicate removal processing is carried out to the network public opinion data；Specially：Using in natural language processing technique Local salted hash Salted carries out Hash calculation to network public opinion data and judges if the cryptographic Hash of two network public opinion data is identical To repeat speech, weed out；
Step 4.2.2：Author's personal information and other users according to network public opinion data is to the network public opinion data forwarding Or the quantity of comment, the confidence level of network public opinion data is calculated；Specific method is：If the author's of network public opinion data Registion time is less than 1 week or the overall network public opinion data number of publication is less than 50, then the author of the network public opinion data is credible Spend it is relatively low, value be [0,0.3]；Otherwise, value is [0.7,1]；Sentiment analysis is carried out for the content of network public opinion data, if Emotion tendentiousness of text for negative content and in forwarding and commenting on is inconsistent, then the emotion of the network public opinion data is credible Spend relatively low, value [0,0.3]；Otherwise, value [0.7,1]；Using the product of author's confidence level and emotion confidence level as the net The confidence level of network public opinion data.
CN201810336472.2A 2018-04-13 2018-04-13 A kind of enterprises under network environment reputation evaluation method Pending CN108764617A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810336472.2A CN108764617A (en) 2018-04-13 2018-04-13 A kind of enterprises under network environment reputation evaluation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810336472.2A CN108764617A (en) 2018-04-13 2018-04-13 A kind of enterprises under network environment reputation evaluation method

Publications (1)

Publication Number Publication Date
CN108764617A true CN108764617A (en) 2018-11-06

Family

ID=64010856

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810336472.2A Pending CN108764617A (en) 2018-04-13 2018-04-13 A kind of enterprises under network environment reputation evaluation method

Country Status (1)

Country Link
CN (1) CN108764617A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110648070A (en) * 2019-09-26 2020-01-03 精硕科技（北京）股份有限公司 Enterprise image evaluation method, enterprise image evaluation device and readable storage medium
CN112559743A (en) * 2020-12-09 2021-03-26 深圳市网联安瑞网络科技有限公司 Method, device, equipment and storage medium for calculating support degree of government and enterprise network
US20220350809A1 (en) * 2021-04-29 2022-11-03 Data Vault Holdings, Inc. Method and system for compiling and utiliziing company data to advance equality, diversity, and inclusion

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104299067A (en) * 2014-08-21 2015-01-21 太仓中科信息技术研究院 Quantifiable enterprise development situation evaluation model
CN105740446A (en) * 2016-02-02 2016-07-06 河南九博科技股份有限公司 Enterprise information integration method and apparatus used for honesty and credit evaluation in recruitment website
CN105787073A (en) * 2016-03-02 2016-07-20 上海源庐加佳信息科技有限公司 Enterprise credit assessment method based on big data mining technology
CN106294532A (en) * 2016-05-18 2017-01-04 广东电网有限责任公司信息中心 The image appraisal algorithm analyzed based on microblog emotional
CN107122432A (en) * 2017-04-18 2017-09-01 广东数相智能科技有限公司 CSR analysis method, device and system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104299067A (en) * 2014-08-21 2015-01-21 太仓中科信息技术研究院 Quantifiable enterprise development situation evaluation model
CN105740446A (en) * 2016-02-02 2016-07-06 河南九博科技股份有限公司 Enterprise information integration method and apparatus used for honesty and credit evaluation in recruitment website
CN105787073A (en) * 2016-03-02 2016-07-20 上海源庐加佳信息科技有限公司 Enterprise credit assessment method based on big data mining technology
CN106294532A (en) * 2016-05-18 2017-01-04 广东电网有限责任公司信息中心 The image appraisal algorithm analyzed based on microblog emotional
CN107122432A (en) * 2017-04-18 2017-09-01 广东数相智能科技有限公司 CSR analysis method, device and system

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110648070A (en) * 2019-09-26 2020-01-03 精硕科技（北京）股份有限公司 Enterprise image evaluation method, enterprise image evaluation device and readable storage medium
CN112559743A (en) * 2020-12-09 2021-03-26 深圳市网联安瑞网络科技有限公司 Method, device, equipment and storage medium for calculating support degree of government and enterprise network
CN112559743B (en) * 2020-12-09 2024-02-13 深圳市网联安瑞网络科技有限公司 Method, device, equipment and storage medium for calculating government and enterprise network support
US20220350809A1 (en) * 2021-04-29 2022-11-03 Data Vault Holdings, Inc. Method and system for compiling and utiliziing company data to advance equality, diversity, and inclusion

Similar Documents

Publication Publication Date Title
Brush et al. The gender gap in venture capital-progress, problems, and perspectives
Islam et al. Use of social media in marketing of library and information services in Bangladesh
CN103150333B (en) Opinion leader identification method in microblog media
CN103699521B (en) Text analyzing method and device
WO2021004550A1 (en) Method, apparatus, and device for generating esg scoring system, and readable storage medium
US8965883B2 (en) Ranking user generated web content
Christensen et al. Universities as agencies: Reputation and professionalization
Williams et al. Mirror on the field: Gender, authorship, and research methods in higher education’s leading journals
JP2014514642A5 (en)
CN108764617A (en) A kind of enterprises under network environment reputation evaluation method
Veselý Conducting large-N surveys on policy work in bureaucracies: Some methodological challenges and implications from the Czech Republic
Debao et al. Analysis of big data job requirements based on K-means text clustering in China
Ronda-Pupo et al. The evolutions of the rich get richer and the fit get richer phenomena in scholarly networks: The case of the strategic management journal
CN106886579A (en) Real-time streaming textual hierarchy monitoring method and device
CN109636337A (en) A kind of talent&#39;s base construction method and electronic equipment based on big data
Bolkan et al. Prolific scholarship in communication studies: Five years in review
Chiu et al. Undesirable output in efficiency and productivity: Example of the G20 countries
Mondal et al. The h-Index: Understanding its predictors, significance, and criticism
CN107590742B (en) Behavior-based social network user attribute value inversion method
Cardoso et al. Topic Prominence of Tourism and Hospitality Scientific Research: The Case of Switzerland
Ginting Organizational culture: An overview and bibliometric analysis
Gray Plotting the contours of an emerging field: an essay and research note concerning British academic social and environmental accountants
CN109102338A (en) Himalaya Art skill Index Assessment method
Claudio-Quiroga et al. Persistence and trends in CO2 emissions in Africa: is Chinese FDI behind these features?
Timofte et al. A bibliometric analysis of creative accounting research

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20181106

WD01 Invention patent application deemed withdrawn after publication