CN109242250A - A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model - Google Patents
A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model Download PDFInfo
- Publication number
- CN109242250A CN109242250A CN201810877992.4A CN201810877992A CN109242250A CN 109242250 A CN109242250 A CN 109242250A CN 201810877992 A CN201810877992 A CN 201810877992A CN 109242250 A CN109242250 A CN 109242250A
- Authority
- CN
- China
- Prior art keywords
- cloud
- behavior
- grade
- attribute
- user
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 85
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 238000011156 evaluation Methods 0.000 claims abstract description 56
- 239000011159 matrix material Substances 0.000 claims abstract description 32
- 230000008569 process Effects 0.000 claims abstract description 11
- 239000013598 vector Substances 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000009193 crawling Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims 1
- 238000012546 transfer Methods 0.000 claims 1
- 238000012854 evaluation process Methods 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 description 90
- 230000003542 behavioural effect Effects 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 13
- 238000013528 artificial neural network Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 239000000428 dust Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000004587 chromatography analysis Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 210000003813 thumb Anatomy 0.000 description 2
- 241000475481 Nebula Species 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000008571 general function Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本发明属于网络数据处理技术领域，公开了一种基于模糊熵权法与云模型的用户行为可信度检测方法,建立行为属性云，建立等级云，通过关联公式计算m个行为云对n个等级云的隶属度，据此得到隶属度矩阵；根据模糊熵权法得到隶属度矩阵元素每一个属性的权重；将隶属度矩阵与权重向量相乘得到最后的评定结果本发明提出的评价系统考虑到计算仅仅围绕云的3个数字特征，没有涉及到更复杂的参数，计算过程简单。发明则是根据属性云与等级云之间的联系确定评价矩阵，云模型本身就是反映事物不确定本质，与行为的随机性、不确定性相契合，评价过程没有主观因素的参与，评价结果更加合理可信。
The invention belongs to the technical field of network data processing, and discloses a user behavior credibility detection method based on a fuzzy entropy weight method and a cloud model. The membership degree of the grade cloud is obtained, and the membership degree matrix is obtained accordingly; the weight of each attribute of the membership degree matrix element is obtained according to the fuzzy entropy weight method; the final evaluation result is obtained by multiplying the membership degree matrix and the weight vector. The evaluation system proposed by the present invention considers To the calculation only revolves around the 3 digital features of the cloud, no more complex parameters are involved, and the calculation process is simple. The invention is to determine the evaluation matrix according to the relationship between the attribute cloud and the grade cloud. The cloud model itself reflects the uncertain nature of things, which is consistent with the randomness and uncertainty of behavior. There is no subjective factor involved in the evaluation process, and the evaluation results are more accurate. reasonably credible.
Description
Technical field
The invention belongs to network data processing technique more particularly to a kind of use based on Based on Entropy method and cloud model
Family behavior reliability detecting method.
Background technique
Currently, the prior art commonly used in the trade is such that
Traditionally, to the management of user primarily directed to subscriber identity information level, including digital certificate, identity card, life
Object feature etc..It is relatively rare to researching and analysing for networks congestion control.Common method has user's row based on fuzzy theory
For trust model, but when constructing weight vectors, the method for use is usually to pass through method of expertise or AHP level
On the one hand analytic approach constructs weight vectors using analysis expert method, so that evaluation procedure supervisor's participation is big, evaluation result is inadequate
It is objective；When on the other hand, using analytic hierarchy process (AHP), the number of plies is not only divided, it is also contemplated that the mutually interconnection between level and level
System, when behavior property is more, complexity is very big.In addition, usual method is that selection is existing when building evaluations matrix is established
Suitable membership function, and these existing subordinating degree functions are the function templates general for a certain classification, it can accurately not
Reflect certain specific behavior ATTRIBUTE INDEXs for the subjection degree of evaluation result.There are also the user behaviors based on BP neural network
Prediction model, such prediction model be on the basis of historical behavior data, by training parameter, reach Trustworthy user behaviour and
The effect of prediction.Evaluation process is sensitive to initial weight value, and needs a large amount of sample data, in the case of data volume is few,
The assessment result that the model obtains is just inaccurate.In addition, such model is also more multiple to the determination of the weight of behavior property index
It is miscellaneous.
In conclusion problem of the existing technology is:
Although traditional trust model can solve the problem of user identity authentication, but after user is successfully accessed network,
Its sequence of operations behavior carried out does not monitor in real time, this just allows malicious user that can forge a normal identity access net
Various attack is carried out after network, and network is allowed to fall among risk.
Existing Trustworthy user behaviour method such as chromatographic analysis model, BP neural network analysis model, fuzzy theory analysis
Method etc. has ignored interaction between each hierarchical elements of user behavior and relation of interdependence and feedback relationship, and assessed
The degree of awareness of the journey dependent on the professional knowledge of expert and to evaluation system, and for weight distribution subjectivity, result in and comment
Estimate that result is unreasonable, influences to establish a credible controllable healthy network environment indirectly.
These assessment model and methods all do not consider the spies such as randomness, the ambiguity of user behavior itself in the prior art
Point has not been able to the uncertain as the principal element for influencing evaluation result of behavior how to influence network peace to user behavior
The understanding of full property is not deep enough.
Solve the difficulty and meaning of above-mentioned technical problem:
The uncertainties such as randomness, the ambiguity of user behavior are an important factor for influencing network security, therefore, to network
While user models, it is necessary to it takes into account and reflects peculiar property possessed by user behavior, and cloud of the present invention
Model theory just perfectly carrys out uncertain be depicted of user behavior, and participates in credible evaluation calculating process, is to realize net
The vital step that network user can manage；Meanwhile the weight of user behavior attribute is assigned using the thought of Based on Entropy method,
Reduce the subjectivity of evaluation process；Since entire evaluation process is calculated without reference to complicated parameter, so that assessment result is more
Accelerate speed, provides calculating guarantee for the real-time of network behavior supervision.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of user's row based on Based on Entropy method and cloud model
For reliability detecting method.
The invention is realized in this way a kind of user's behavior confidence level detection side based on Based on Entropy method and cloud model
Method, the user's behavior confidence level detection method based on Based on Entropy method and cloud model
It establishes behavior property cloud: m behavior property is divided by behavioural characteristic to the behavioral data of collection；According to each
The sample data of behavior property replaces population mean using sample average, it is expected that, sample variance replaces the side of population variance
Method obtains entropy and super entropy, restores the numerical characteristic of cloud；Obtain m behavior property cloud；
It establishes grade cloud: providing n opinion rating, determine the numerical characteristic of each grade cloud, obtain n grade cloud；
M behavior cloud is calculated to the degree of membership of n grade cloud by incidence formula, obtains subordinated-degree matrix accordingly；According to
Based on Entropy method obtains the weight of each attribute of subordinated-degree matrix element；
The evaluation result that subordinated-degree matrix is multiplied to the end with weight vectors.
Further, it establishes in grade cloud, specifically includes: providing n opinion rating, if a grade interval range is [rmin,
rmax], according toHe=0.02 determines the numerical characteristic of each grade cloud, obtains n
A grade cloud；
Wherein, rmin, rmaxThe respectively lower boundary of grade interval and coboundary, Ex, En, HeThe respectively expectation of grade cloud,
Entropy, super entropy.
Further, pass through incidence formulaM behavior cloud is calculated to the person in servitude of n grade cloud
Category degree, obtains subordinated-degree matrix accordingly；It is used according to element of the Based on Entropy method to subordinated-degree matrixObtain the weight of each attribute.
Wherein, m is the number of behavior property, and n is the number of grade classification, rijIt is the element for constituting subordinated-degree matrix R, xi
It is behavior property water dust, Ex, En, it is expectation and the entropy of grade cloud, pijIndicate element r in subordinated-degree matrixijIts column is accounted for
According to specific gravity, EiIt is the entropy for i-th of behavior property that the comentropy calculation formula proposed using Shannon is obtained, E 'iWhat is taken is entropy
Inverse, show that entropy is inversely proportional with attribute weight, wiIt is according to entropy assessment, the weight of i-th of behavior property of calculating.
Another object of the present invention is to provide a kind of computer programs.
Another object of the present invention is to provide a kind of information data processing terminals.
Another object of the present invention is to provide a kind of network user's real-time monitoring and controlling terminals, for realizing described
User's behavior confidence level detection method based on Based on Entropy method and cloud model, is also used to network user's lifecycle management.
Another object of the present invention is to provide a kind of trustworthy user behavior Evaluation Platforms, for realizing described based on mould
The user's behavior confidence level detection method for pasting entropy assessment and cloud model, is also used to each entity trusts data under heterogeneous network environment
Intercommunication carries out trust value transmitting.
Another object of the present invention is to provide the online social activities described in a kind of realize based on Based on Entropy method and cloud model
Networks congestion control credible evaluation system.
Another object of the present invention is to provide the user behaviors described in a kind of realize based on Based on Entropy method and cloud model
The computer program of reliability detecting method.
Another object of the present invention is to provide the user behaviors described in a kind of realize based on Based on Entropy method and cloud model
The information data processing terminal of reliability detecting method.Another object of the present invention is to provide one kind based on Based on Entropy method with
The user's behavior confidence level detection system of cloud model includes:
Behavioral data acquisition module, by crawling user behavior data or special data offer website collection on website
Target data；
Data preprocessing module filters out significant behavioral data, and press for carrying out denoising to target data
Several attributes are divided into according to behavioural characteristic；
Backward cloud generator module passes through each behavior property reverse for describing the uncertainty of behavioral data
Cloud generator forms attribute cloud；
Normal Cloud Generator module for dividing reliability rating, and passes through positive cloud according to each rate range and occurs
Device forms grade cloud.
Evaluation module, for the modeling to behavioral data；Then it further according to the relationship between attribute cloud and grade cloud, obtains
Subordinated-degree matrix obtains the weight of attribute, obtains trustworthy user behavior value using entropy assessment to the processing of subordinated-degree matrix element.
Another object of the present invention is to provide the user behaviors described in a kind of carrying based on Based on Entropy method and cloud model
The network data processing platform of confidence level detection system.
In conclusion advantages of the present invention and good effect are as follows:
It is described as follows:
Trustworthy user behavior assessment models based on fuzzy theory, but when constructing weight vectors, the method for use is logical
It is often on the one hand weight vectors are constructed using analysis expert method by method of expertise or AHP analytic hierarchy process (AHP), so that evaluation
Process hosts' participation is big, and evaluation result is not objective enough；When on the other hand, using analytic hierarchy process (AHP), the number of plies is not only divided, also
Consider connecting each other between level and level, when behavior property is more, complexity is very big.In addition, evaluating square in building
When battle array is established, usual method is to select existing suitable membership function, and these existing subordinating degree functions are for certain one kind
Other general function template, can not accurately reflect that certain specific behavior ATTRIBUTE INDEXs are subordinate to journey for evaluation result
Degree；
Users' behavior model based on BP neural network, such prediction model are on the basis of historical behavior data
On, by training parameter, achieve the effect that Trustworthy user behaviour and prediction.Evaluation process is sensitive to initial weight value, initial value
If distribution is wrong, global outcome will affect, and need a large amount of sample data, in the case of data volume is few, the model
Obtained assessment result is just inaccurate.In addition, such model is also complex to the determination of the weight of behavior property index.
It is proposed by the present invention based on cloud model for the limitation that two kinds of front trustworthy user behavior assessment models have
Comprehensive evaluation system is subordinated-degree matrix to be obtained, both by behavior by calculating the relationship between behavior cloud and grade cloud
Uncertain and ambiguity is described by behavior cloud, but by behavior property to the membership of different grades by behavior cloud and
Incidence relation between grade cloud reflects, and such mode is more in line with objective fact.Further, since the present invention using
' entropy assessment ' self-adjusted block attribute weight assigns different power to evaluation result contribution difference according to each behavior property accordingly
Weight, further reduces subjectivity.The accuracy of the evaluation result of raising.Simultaneously as cloud model is mainly by 3 numbers
Feature forms, easy to operate mainly around 3 digital feature calculations in evaluation procedure, avoids and utilizes fuzzy overall evaluation mould
When type, BP neural network analysis model, the time-consuming as caused by delaminating process and training parameter is big, and evaluation procedure is complicated to ask
Topic.In addition, with the rapid development of Internet, oneself warp of network becomes the main channel that people obtain information.Social networks is existing
A key areas in modern internet.With the development of mobile internet, social networks is more next to people's lives etc.
It is more important, it is that a very important message propagates platform, for people when enjoying virtual world bring convenience, there is also information
The security risks such as leakage, virus link, emotion fraud, therefore carrying out assessment to the trust of user in online social networks is very
An important analysis directions.The present invention acquires behavioral data of the user in social platform from the Behavior trustworthiness of user,
The trust evaluation result of these attributes objects is synthesized to obtain the comprehensive letter of user by cloud model and Fuzzy comprehensive evaluation system
Appoint value, provides a kind of new thinking for the assessment that online social network user is trusted.
Detailed description of the invention
Fig. 1 is the user's behavior confidence level detection method provided in an embodiment of the present invention based on Based on Entropy method and cloud model
Flow chart.
Fig. 2 is the standard cloud exemplary diagram provided in an embodiment of the present invention generated based on cloud models theory Normal Cloud Generator.
Fig. 3 is the grade cloud exemplary diagram provided in an embodiment of the present invention that user's confidence level is described using cloud models theory.
Fig. 4 is the user's behavior confidence level detection system provided in an embodiment of the present invention based on Based on Entropy method and cloud model
Schematic diagram.
In figure: 1, behavioral data acquisition module；2, data preprocessing module；3, backward cloud generator module；4, positive cloud
Generator module；5, evaluation module.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Although traditional trust model can solve the problem of user identity authentication, but after user is successfully accessed network,
Its sequence of operations behavior carried out does not monitor in real time, this just allows malicious user that can forge a normal identity access net
Various attack is carried out after network, and network is allowed to fall among risk.
Existing Trustworthy user behaviour method such as chromatographic analysis model, BP neural network analysis model, fuzzy theory analysis
Method etc. has ignored interaction between each hierarchical elements of user behavior and relation of interdependence and feedback relationship, and assessed
The degree of awareness of the journey dependent on the professional knowledge of expert and to evaluation system, and for weight distribution subjectivity, result in and comment
Estimate that result is unreasonable, influences to establish a credible controllable healthy network environment indirectly.
The invention will be further described combined with specific embodiments below.
Cloud model mainly be made of three expectation, entropy, super entropy numerical characteristics, Normal Cloud Generator be input cloud this three
A numerical characteristic and water dust number obtain water dust and its degree of membership relationship, and backward cloud generator is then the sample point root according to input
Learn the mean value of the inside and the calculation formula of variance and degree of membership according to statistics to go back three numerical characteristics of primitive nebula.
Fig. 1 is the user's behavior confidence level detection method provided in an embodiment of the present invention based on Based on Entropy method and cloud model
Flow chart.
Step 1. establishes behavior property cloud.M behavior property is divided by its behavioural characteristic to the behavioral data of collection.Root
According to the sample data of each behavior property, population mean is replaced using sample average, it is expected that, sample variance replaces overall
The method of variance obtains entropy and super entropy, restores the numerical characteristic of cloud.To obtain m behavior property cloud.
Step 2. establishes grade cloud.Traditional grade determines that method is usually directly according to number of levels, by a closed zone
Between be divided into corresponding section number, but this method leads to that there are level boundaries since the subjectivity of interval division is too big
Fuzzy problem.Therefore in order to reduce the ambiguities of level boundaries, the present invention establishes grade cloud appropriate to expand each etc.
Grade section.N opinion rating is provided in advance, if a grade interval range is [rmin, rmax], according toHe=0.02 determines the numerical characteristic of each grade cloud, and then obtains n etc.
Grade cloud.
Step 3. passes through incidence formulaM behavior cloud is calculated to the person in servitude of n grade cloud
Category degree, obtains subordinated-degree matrix accordingly.It is used according to element of the Based on Entropy method to subordinated-degree matrixObtain the weight of each attribute.Root
Obtaining the weight of m behavior property its basic thought according to ' entropy assessment ' is: attribute i is smaller for the entropy of opinion rating j, description line
Contribution for the evaluation of attribute In Grade is bigger, then the weight distributed should be bigger.Entropy is bigger, illustrates that uncertainty is bigger, useful
Information is fewer, and corresponding weight is smaller.
The evaluation result that subordinated-degree matrix is multiplied to the end by step 4. with weight vectors.
Fig. 2 is the standard cloud exemplary diagram provided in an embodiment of the present invention generated based on cloud models theory Normal Cloud Generator.
Fig. 3 is the grade cloud exemplary diagram provided in an embodiment of the present invention that user's confidence level is described using cloud models theory.
Fig. 4, the user's behavior confidence level detection system provided in an embodiment of the present invention based on Based on Entropy method and cloud model,
Include:
Behavioral data acquisition module 1, by crawling user behavior data or special data offer website receipts on website
Collect target data.
Data preprocessing module 2, due to that may include the field useless to research in the data of offer, because needing logarithm
According to denoising is carried out, significant behavioral data is filtered out, and be divided into several attributes according to behavioural characteristic.
Backward cloud generator module 3 passes through each behavior property inverse for the uncertain feature for describing behavioral data
Attribute cloud is formed to cloud generator.
Normal Cloud Generator module 4 divides reliability rating, and passes through Normal Cloud Generator shape according to each rate range
At grade cloud.
Evaluation module 5 completes the modeling to behavior by above-mentioned 4 modules.Then further according to attribute cloud and grade cloud
Between relationship, obtain subordinated-degree matrix, using entropy assessment to subordinated-degree matrix element processing, obtain the weight of attribute, in turn
Obtain trustworthy user behavior value.
The present invention is described further combined with specific embodiments below.
For traditional Trustworthy user behaviour method, that there are evaluation procedures is more subjective, and weight distribution is unreasonable, and neglects
Having omited user behavior, there are the limitations of randomness and the feature of ambiguity, and the invention proposes one kind based on cloud model and to obscure
The use that comprehensive evaluation combines
Family behavior credible evaluation method, using the uncertainty of the uncertainty description user behavior of cloud, by the not true of user
The qualitative principal element as influence evaluation result, utilizes ' Based on Entropy method ' objectively to determine attribute weight.And it is applied
Into the user behavior evaluation of social networks.Specific step is as follows:
(1) user behavior data is collected, and data are pre-processed, set of factors is established according to user behavior attribute.And
Backward cloud generator is utilized according to the behavioral data of input, obtains three numerical characteristics of behavior cloud.User behavior in the invention
Data source inhttps://www.kaggle.com/datasetsWebsite is collected by the website from Facebook society
Hand over the publication item number on platform about 10 users in 2015-2016, comment item number, the information for thumbing up number.Specific number
According to as follows:
Validity in order to better illustrate the present invention, has extracted 1 normal users and an abnormal user is tested, number
According to being respectively:
User 1 issues content=65 and thumbs up=195 comments=378
User 7 issues content=2 and thumbs up=432 comments=708
And using this as 3 attributes of user behavior, the evaluation result of user is divided into: it is extremely insincere, low it is credible, generally may be used
Letter, credible, the corresponding grade 1-4 of height.The grade of each evaluation index of user is as follows:
Due to each level attributed incremental relationship, the present invention integrated using A each evaluation index fall into it is different grades of
Value, i.e. A=v1+3×v2+6×v3+9×v4(v1-v4Successively indicate grade 1-4).
By backward cloud generator using the data of these three evaluation indexes as input, 3 user behavior attribute clouds are obtained.
It being computed, the numerical characteristic of 3 attribute clouds of user 1 is respectively as follows: (5.42,4.79,1.49), (16.25,12.48,3.77),
(32,15.04,3.13)。
(2) according to user behavior data value divided rank range, the rate range divided herein is [0,0.2], when taking
When to be worth range is [0,0.05], it is in ' extremely credible ' state, when [0.05,0.10], is in ' low credible ', [0.10,0.15],
In ' general credible ' state, when [0.15,0.2], in ' high credible ' state.The specific details that divides is as follows:
| Interval value | Grade | Corresponding description |
| [0,0.05] | 1 | It is extremely insincere |
| [0.05,0.10] | 2 | It is low credible |
| [0.10,0.15] | 3 | It is general credible |
| [0.15,0.2] | 4 | It is high credible |
Therefore according to formulaHe=0.02, obtain 4 grade clouds, such as Fig. 2
It is shown.
(3) different behavior properties is different to the contribution of evaluation result, and it is user behavior category that this model, which utilizes ' entropy assessment ',
Property assign weight.It is computed the weight W=[0.08,0.72,0.20] of these three attributes.
(4) according to formulaBetween the 3 attribute clouds and 4 grade clouds for calculating user 1
Membership, obtained fuzzy relation matrix is
(5) fuzzy relation matrix is multiplied with weight vectors, obtains evaluation result V to the end1=W × R=[0.0067,
0.0061,0.0094,0.0124].Pass through A=v again1+3×v2+6×v3+9×v4This trust vector is synthesized, is obtained to the end
Evaluation result A=0.191.Therefore user 1 is in ' high credible ' state.
It is 0.079 with the confidence values that same step calculates user 2.Therefore user 2 is in ' low credible ' state.
Below with reference to concrete analysis, the invention will be further described.
Evaluation system proposed by the present invention is not involved with more multiple in view of calculating only only around 3 numerical characteristics of cloud
Miscellaneous parameter, calculating process are simple.
Compared with Field Using Fuzzy Comprehensive Assessment, since it is when establishing weight vectors, method of expertise and AHP layers are generally used
Fractional analysis, have stronger subjectivity, due to evaluation procedure influenced by the emotion of people or human-subject test exist limitation
Property, and analytic hierarchy process (AHP) does not reflect that the uncertain essence of things, evaluation result can have biggish error.And the present invention is then
According to the determining evaluations matrix that contacts between attribute cloud and grade cloud, cloud model inherently reflects the uncertain essence of things, with
Randomness, the uncertainty of behavior are mutually agreed with, and evaluation procedure does not have the participation of subjective factor, and evaluation result is more rationally credible.
Compared with BP neural network analysis model, inputted due to needing to establish multistage, and the network number of plies, neuron number
No corresponding theoretical direction is selected, and the weight of every level-one determines and a large amount of sample data is needed to be trained to obtain, and learns
Speed is slow, even a simple question, being generally also required to several hundred times or even thousands of times study could restrain, therefore utilize
For BP neural network analysis model to Trustworthy user behaviour, the evaluation procedure being related to is complicated, computationally intensive.Furthermore the present invention couple
The network user carries out credible evaluation, and the behavioral data amount that each user generates is irregular, and more or less, therefore the present invention is not
It is suitble to utilize BP neural network analysis model, and uses evaluation model of the invention, then any user behavior can be commented effectively
Estimate, and process is simple, calculation amount is small.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one
A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)
Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center
Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access
The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie
Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid
State Disk (SSD)) etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810877992.4A CN109242250A (en) | 2018-08-03 | 2018-08-03 | A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810877992.4A CN109242250A (en) | 2018-08-03 | 2018-08-03 | A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109242250A true CN109242250A (en) | 2019-01-18 |
Family
ID=65070379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810877992.4A Pending CN109242250A (en) | 2018-08-03 | 2018-08-03 | A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109242250A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222961A (en) * | 2019-05-24 | 2019-09-10 | 南京师范大学 | A kind of Power Generation market forces evaluation method based on entropy assessment and cloud model |
| CN110347477A (en) * | 2019-07-02 | 2019-10-18 | 北京邮电大学 | Service adaptation dispositions method and device under a kind of cloud environment |
| CN110598281A (en) * | 2019-08-28 | 2019-12-20 | 桂林理工大学 | Prediction and analysis method of karst collapse in normal cloud model based on entropy weight method |
| CN111598749A (en) * | 2020-05-07 | 2020-08-28 | 重庆第二师范学院 | Optimization method for education development research |
| CN112566219A (en) * | 2020-12-03 | 2021-03-26 | 国网甘肃省电力公司信息通信公司 | Weight determination method for power communication heterogeneous network selection |
| CN113468589A (en) * | 2020-03-30 | 2021-10-01 | 微软技术许可有限责任公司 | Detecting data collection validity |
| CN113660147A (en) * | 2021-10-21 | 2021-11-16 | 成都数默科技有限公司 | IP session sequence periodicity evaluation method based on fuzzy entropy |
| CN114490412A (en) * | 2022-02-14 | 2022-05-13 | 浙江大学 | 3D CAD software performance measurement method and device based on self-reducing inverse cloud generator |
| CN118627125A (en) * | 2024-08-12 | 2024-09-10 | 山东赛宝信息技术咨询有限公司 | Database information leakage detection system based on fuzzy entropy algorithm |
| CN119178738A (en) * | 2024-11-22 | 2024-12-24 | 北京观微科技有限公司 | Soil nutrient evaluation method, device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046453A (en) * | 2015-09-02 | 2015-11-11 | 中铁十一局集团有限公司 | Construction engineering project cluster establishment method introducing cloud model for evaluation and selection |
| CN106407493A (en) * | 2016-03-15 | 2017-02-15 | 中南大学 | Multi-dimensional Gaussian cloud model-based rock burst grade evaluation method |
| CN107909255A (en) * | 2017-11-06 | 2018-04-13 | 天津大学 | A kind of Effect of Curtain Grouting fuzzy synthetic appraisement method based on cloud model |
-
2018
- 2018-08-03 CN CN201810877992.4A patent/CN109242250A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046453A (en) * | 2015-09-02 | 2015-11-11 | 中铁十一局集团有限公司 | Construction engineering project cluster establishment method introducing cloud model for evaluation and selection |
| CN106407493A (en) * | 2016-03-15 | 2017-02-15 | 中南大学 | Multi-dimensional Gaussian cloud model-based rock burst grade evaluation method |
| CN107909255A (en) * | 2017-11-06 | 2018-04-13 | 天津大学 | A kind of Effect of Curtain Grouting fuzzy synthetic appraisement method based on cloud model |
Non-Patent Citations (4)
| Title |
|---|
| 张海侠: "基于熵权—云模型的智慧城市发展潜力评价——以华东13城市为例", 《淮阴工学院学报》 * |
| 徐廷学等: "《导弹武器系统可用度评估、优化与预测方法》", 31 January 2017, 国防工业出版社 * |
| 杨晓东等: "《舰船半解析式惯性导航原理及应用》", 31 May 2014 * |
| 王丹琛: "面向信息系统业务运行的应用安全评估及关键问题研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222961A (en) * | 2019-05-24 | 2019-09-10 | 南京师范大学 | A kind of Power Generation market forces evaluation method based on entropy assessment and cloud model |
| CN110347477B (en) * | 2019-07-02 | 2021-07-09 | 北京邮电大学 | A kind of service adaptive deployment method and device in cloud environment |
| CN110347477A (en) * | 2019-07-02 | 2019-10-18 | 北京邮电大学 | Service adaptation dispositions method and device under a kind of cloud environment |
| CN110598281A (en) * | 2019-08-28 | 2019-12-20 | 桂林理工大学 | Prediction and analysis method of karst collapse in normal cloud model based on entropy weight method |
| CN113468589A (en) * | 2020-03-30 | 2021-10-01 | 微软技术许可有限责任公司 | Detecting data collection validity |
| CN111598749A (en) * | 2020-05-07 | 2020-08-28 | 重庆第二师范学院 | Optimization method for education development research |
| CN112566219A (en) * | 2020-12-03 | 2021-03-26 | 国网甘肃省电力公司信息通信公司 | Weight determination method for power communication heterogeneous network selection |
| CN112566219B (en) * | 2020-12-03 | 2023-05-19 | 国网甘肃省电力公司信息通信公司 | A Weight Determination Method for Power Communication Heterogeneous Network Selection |
| CN113660147A (en) * | 2021-10-21 | 2021-11-16 | 成都数默科技有限公司 | IP session sequence periodicity evaluation method based on fuzzy entropy |
| CN113660147B (en) * | 2021-10-21 | 2022-01-11 | 成都数默科技有限公司 | IP session sequence periodicity evaluation method based on fuzzy entropy |
| CN114490412A (en) * | 2022-02-14 | 2022-05-13 | 浙江大学 | 3D CAD software performance measurement method and device based on self-reducing inverse cloud generator |
| CN118627125A (en) * | 2024-08-12 | 2024-09-10 | 山东赛宝信息技术咨询有限公司 | Database information leakage detection system based on fuzzy entropy algorithm |
| CN118627125B (en) * | 2024-08-12 | 2024-10-18 | 山东赛宝信息技术咨询有限公司 | Database information leakage detection system based on fuzzy entropy algorithm |
| CN119178738A (en) * | 2024-11-22 | 2024-12-24 | 北京观微科技有限公司 | Soil nutrient evaluation method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109242250A (en) | A kind of user's behavior confidence level detection method based on Based on Entropy method and cloud model | |
| US11659050B2 (en) | Discovering signature of electronic social networks | |
| CN108881110B (en) | Security situation assessment and defense strategy combined decision method and system | |
| Zhao et al. | A machine learning based trust evaluation framework for online social networks | |
| TW201734837A (en) | Multi-sampling model training method and device | |
| CN103455842B (en) | Credibility measuring method combining Bayesian algorithm and MapReduce | |
| CN102609612B (en) | Data fusion method for calibration of multi-parameter instruments | |
| CN109117380A (en) | A kind of method for evaluating software quality, device, equipment and readable storage medium storing program for executing | |
| CN110858973A (en) | Method and device for predicting network traffic of cell | |
| CN104599084A (en) | Crowd calculation quality control method and device | |
| CN107895038A (en) | A kind of link prediction relation recommends method and device | |
| CN110443265A (en) | A kind of behavioral value method and apparatus based on corporations | |
| Pérez-Ortiz et al. | Network topological determinants of pathogen spread | |
| CN114444910A (en) | An edge network system health assessment method for power Internet of things | |
| RU2607977C1 (en) | Method of creating model of object | |
| Zeng et al. | Influential simplices mining via simplicial convolutional networks | |
| CN114841598B (en) | Decision-making methods, devices, equipment and program products for operational risks | |
| CN113569162A (en) | Data processing method, device, equipment and storage medium | |
| Dong | Application of Big Data Mining Technology in Blockchain Computing | |
| CN112860973A (en) | Organization health assessment method based on social network analysis | |
| CN117857202B (en) | Multi-dimensional security assessment method for information system | |
| Cheng et al. | Community detection based on directed weighted signed graph convolutional networks | |
| CN110827111A (en) | Real-time credit analysis method and system based on dynamic extension model | |
| CN117196384A (en) | Efficiency evaluation method and system for information system | |
| CN113656692B (en) | Product recommendation method, device, equipment and medium based on knowledge migration algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |