CN106844515A - Computer user's behavior analysis method based on gene expression programming - Google Patents
Computer user's behavior analysis method based on gene expression programming Download PDFInfo
- Publication number
- CN106844515A CN106844515A CN201611239027.1A CN201611239027A CN106844515A CN 106844515 A CN106844515 A CN 106844515A CN 201611239027 A CN201611239027 A CN 201611239027A CN 106844515 A CN106844515 A CN 106844515A
- Authority
- CN
- China
- Prior art keywords
- computer user
- database
- population
- data
- gene expression
- 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
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 230000014509 gene expression Effects 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims description 52
- 238000013500 data storage Methods 0.000 claims description 11
- 230000000977 initiatory effect Effects 0.000 claims description 6
- 108090000623 proteins and genes Proteins 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 108020004999 messenger RNA Proteins 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000005215 recombination Methods 0.000 claims 1
- 230000006798 recombination Effects 0.000 claims 1
- 230000006399 behavior Effects 0.000 description 22
- 238000005516 engineering process Methods 0.000 description 7
- 230000002068 genetic effect Effects 0.000 description 4
- 238000011112 process operation Methods 0.000 description 4
- 238000010206 sensitivity analysis Methods 0.000 description 4
- 230000006978 adaptation Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000010353 genetic engineering Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 241000222120 Candida <Saccharomycetales> Species 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24564—Applying rules; Deductive queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of computer user's behavior analysis method based on gene expression programming, including：Obtain the personal information and progress information of computer user；Obtain corresponding initial data and be saved in database, corresponding computation rule is obtained according to progress information；By initial data and computation rule knot merga pass GEP algorithms, obtain calculating data, data will be calculated and be saved in database；The calculating data that will be obtained are matched with the precondition in knowledge base, obtain inference conclusion, inference conclusion from when existing characteristic is different in database, database is stored in using inference conclusion as characteristic, replaced calculating Data duplication matching with inference conclusion, until inference conclusion is identical with existing characteristic in database；Step 5：Inference conclusion is exported.The present invention is analyzed to current operation behavior, browsed web content of user etc. in real time using the pattern of Multi- Base Coopera-tion, and manual intervention is few, and automaticity and precision are higher.
Description
Technical field
The present invention relates to areas of information technology.It is more particularly related to a kind of based on gene expression programming
Computer user's behavior analysis method.
Background technology
User behavior analysis, refer in the case where website visiting amount master data is obtained, relevant data are counted,
The process of analysis.By user behavior analysis it can be found that user accesses the rule of website, and by these rules and network marketing
Strategy etc. is combined, so that problem that may be present in current network marketing activity is found, and further to correct or making again
Determine net marketing strategy and foundation is provided.
Gene expression programming (Gene Expression Programming, GEP) is Portugal scholar Candida
Ferreira is in 2001 in genetic algorithm (Genetic Algorithm, GA) and genetic programming (Genetic
Programming, GP) on the basis of develop new ideas.With represent individual GA with the linear string of regular length and use length
Represent that individual GP is different with variform non-linear entity, GEP be by individual UVR exposure into regular length linear string (gene
Group or chromosome), the non-linear entity of different length and shape is then converted into, it is achieved thereby that being represented with simple code
Challenge, while be easy to genetic manipulation, and by the new individual produced by genetic manipulation be all grammatically it is effective, no
Need to carry out Effective judgement and treatment to new individual, 2~4 orders of magnitude are improve than GP in speed.
The aspect such as real-time current operation behavior, browsed web content to user is entered in existing user behavior analysis scheme
Being related to for row sensitivity analysis is few, on the one hand, when user behavior data is obtained, due to that ceaselessly can be asked to each application service
Ask, face under huge request of data, the problems such as often cause congestion to collapse.On the other hand, database it is sufficiently complete, it is necessary to
Correction amendment, is unfavorable for the situation of the multiple behavior of analytic statistics at any time, is unfavorable for inquiring about single user behavior, and for big number
Data computing capability according to amount is poor, and data storage capacities are also poor, easily cause system bottleneck.Therefore consider using the intelligence of GEP
A kind of automaticity of analytical technology research and precision computer user's behavior analysis method higher.
The content of the invention
It is an object of the invention to solve the above problems, and provide the advantage that will be described later.
It is a still further object of the present invention to provide a kind of behavioural analysis side of computer user based on gene expression programming
Method, using the pattern of Multi- Base Coopera-tion, carries out sensitiveness point to aspects such as current operation behavior, the browsed web contents of user in real time
Analysis, manual intervention is few, and automaticity and precision are higher, and system operation is more smooth.
In order to realize these purposes of the invention and further advantage, there is provided a kind of based on gene expression programming
Computer user's behavior analysis method, comprises the following steps：
Step one：Obtain the personal information and progress information of computer user；
Step 2：The computation rule in rule base is obtained by progress information correspondence, is obtained according to personal information and progress information
To after corresponding initial data, initial data is saved in database；
Step 3：GEP algorithms in the initial data and computation rule combination knowledge base that will obtain, obtain calculating data,
Data will be calculated and be saved in database；
Step 4：After the calculating data that will be obtained are matched with the precondition in knowledge base, precondition is obtained
Inference conclusion, when inference conclusion from existing characteristic is different in database when, using inference conclusion as new characteristic
Database is stored in, is replaced calculating Data duplication step 4 with inference conclusion, until inference conclusion and existing spy in database
Levy data it is identical when terminate；
Step 5：To as a result be exported with existing characteristic identical inference conclusion in database.
Preferably, calculating data are obtained in step 3 and specifically includes following steps：
S1：Self-defined initiation parameter, initiation parameter includes Population Size N, sub- Population Size M, maximum evaluation number of times
MAX_FE, functor and terminal symbol, mrna length, gene number, mutation probability, slotting string probability, slotting string length and restructuring are general
Rate, the original data definition that will be obtained is initial population Pt={ X1,X2,…,XN, and initial population is calculated according to computation rule
In each individual fitness fi；
S2：By performing selection, variation, slotting string and the restructuring of gene expression programming to population PtIn individuality
M new individual of generation, and M new individual is organized into sub- population Ot, O is calculated according to computation ruletIn each individuality adaptation
The maximum individuality of degree, wherein fitness is optimum individual；
S3：By population PtIn individual and sub- population OtIn the M individual interim population P ' t of composition, and according to calculating
Rule calculates each the individual fitness in interim population P ' t, then deletes in interim population P ' t before fitness maximum
M individual, obtains the population P of new generation being made up of individualityt+1；
S4：S2 to S3 is repeated until evaluating number of times and reaching and terminate after MAX_FE, at the end of the population P of new generation that obtainst+1I.e.
To calculate data.
Preferably, the database of step 2 includes data storage storehouse, knowledge base and rule base, divides in data storage storehouse
Be not stored with initial data, calculating data and characteristic.
Preferably, the knowledge base of step 4 includes that IF-THEN is regular, C=C (P) in IF-THEN rules, wherein, C
It is conclusion, condition premised on P,N is the premise number of every rule, θ={ AND, OR }, and every rule
Confidence level is CF, CF=[0,1].
Preferably, the progress information for being obtained in step one includes real-time process and historical progress；
Whether computation rule in step 2 includes the typing of the time of origin of corresponding real-time process and to performing correspondence
Real-time process judgement.
Preferably, the computation rule in step 2 also including the cycle calculating and in historical progress in the cycle with it is real-time
The statistics of the generation frequency of same process in process.
Preferably, when the personal information of computer user is obtained in step one, calculated with each according in each account
The one-to-one ID correspondence of machine user obtains the personal information of computer user, wherein, each computer user correspondence
One or more accounts.
The present invention at least includes following beneficial effect：
Using the powerful function discovery feature of the Intellectual Analysis Technology of GEP and IF-THEN rule generation work(in the present invention
Can, an intellectuality, integrated, concert rationality expert system are set up, based on gene expression programming to computer user's behavior
It is analyzed, the pattern cooperateed with using multiple databases such as data storage storehouse, knowledge base and rule bases, in real time to the current of user
The aspects such as operation behavior, browsed web content carry out sensitivity analysis and monitoring, and computing capability and storage capacity are powerful, system fortune
Row is more stable, smooth, and manual intervention is few, and automaticity and precision are higher, helps to realize the prison to networks congestion control
Pipe, preferably purifies Internet environment, and especially underage users have to computer user to reduce or avoid the bad network information
Evil influence.
Further advantage of the invention, target and feature embody part by following explanation, and part will also be by this
The research and practice of invention and be understood by the person skilled in the art.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the computer user's behavior analysis method based on gene expression programming of the invention.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings, to make those skilled in the art with reference to specification text
Word can be implemented according to this.
As shown in figure 1, the present invention provides a kind of computer user's behavior analysis method based on gene expression programming, bag
Include following steps：
Step one i.e. 101：Obtain the personal information and progress information of computer user；
Step 2 is 102：The computation rule in rule base is obtained by progress information correspondence, is believed according to personal information and process
After breath obtains corresponding initial data, initial data is saved in database；
Step 3 is 103：GEP algorithms in the initial data and computation rule combination knowledge base that will obtain, are calculated
Data, will calculate data and are saved in database；
Step 4 is 104：After the calculating data that will be obtained are matched with the precondition in knowledge base, premise bar is obtained
The inference conclusion of part, when inference conclusion from existing characteristic is different in database when, using inference conclusion as new feature
Data are stored in database, are replaced calculating Data duplication step 4 with inference conclusion, that is, repeat the process matched with precondition,
Until the new inference conclusion that obtains with terminate when existing characteristic is identical in database；
Step 5 is 105：To as a result be exported with existing characteristic identical inference conclusion in database.
User access pattern classification is carried out first with GEP technologies in the present invention, relevant classification model, disaggregated model is set up
In grouped data can preserve offline for calling below, then using preconditions such as the IF-THEN rules in knowledge base, knot
The grouped data for closing different user makes inferences, and improves the real-time of GEP technology applications, reaches the purpose of on-line analysis.Using
The pattern of Multi- Base Coopera-tion, establishes database, rule base and knowledge base etc., current operation behavior real-time to user, browses
The aspects such as web page contents carry out sensitivity analysis and monitoring, and manual intervention is few, and automaticity and precision are higher, and system operation is more
It is smooth.
There are 2 main frames S, C in LAN, wherein, S is service end, and C is client；It is loaded with S including knowledge base, is deposited
Storage database and the database of rule base, scheduling controller and inference machine, the initial data that is stored with respectively in data storage storehouse,
Data and characteristic are calculated, a series of IF-THEN rules are generated by GEP technologies wherein in knowledge base.
Wherein, the relevant data and result of data storage library storage, rule base deposits some relevant mathematical computations
Model, methods and procedures, knowledge base deposit some about expert's property, regular knowledge in computer network field.Reasoning
Using data-driven, positive uncertain inference strategy, the essence of reasoning is that knowledge rule is chained up to process, forms one
Bar or a plurality of inference chain.System according to the measurement result of user, by after word, the treatment of image, by corresponding information characteristics
Preserved in the form of data storage, used for inference machine as the fact that be input into.
Scheduling controller is on the basis of network model, it is established that a kind of many storehouses between, knowledge base and inference machine it
Between collaborative strategy.Because for each specific project, the form of input and output is all fixed, derivation relationship
It is also identical, the difference is that specific content in reasoning process, scheduling controller is each information bank of connection and functional module
Hinge, relies primarily on program meanses to realize.
Embodiment 1
User wishes to supervise the playtime of underage users by service end S, 6~7 points of objects for appreciation can only such as play at night,
In advance on service end S set underage users only have at night 6~7 points could open game, remaining time can not open trip
Play, adult human user is not limited then, and this information is stored in knowledge base as rule, i.e. C=C (P, CF) in IF-THEN rules, its
In, precondition P includes P1 and P2, and meeting P1 or P2 can be it is concluded that the i.e. game process operation of C, and this rule confidence level CF is
1.P1 is teenage, evening 6~7 point and is played that correspondence premise number is 3, and three premises meet simultaneously just can enter game
Cheng Yunhang, P2 are for adult and are played, and correspondence premise number is 2, and two premises meet simultaneously can run game process.
When underage users open client C opening music at 9 points at night on client C, game is then clicked on.
First, service end S obtains input scheduling control after the personal information and real-time process information of user from client C
Device, because underage users show user for underage users in logon account information, real-time login time is 10 points of night,
Real-time process information is music process and game process, and scheduling controller is obtained according to the personal information and real-time process information of user
It is teenage, 10 points of night, music process and game process to corresponding initial data, computation rule is to judge that now each enters
Whether journey is run.
The initial data and computation rule that will be obtained are combined, and by the GEP algorithms in knowledge base, obtaining calculating data is
It is teenage in the operation of 10 music processes of night and game process operation, teenage in 10 music processes operations of night and play
Process do not run, teenage do not run but game process operation and teenage in 10 sounds of night in 10 music processes of night
Happy process is not run and game process does not run, and the precondition in the calculating data that scheduling controller will be obtained and knowledge base is i.e.
Underage users evening 6~7 point could be opened after game matched, and obtain the inference conclusion of precondition for teenage at night
10 music processes of evening run and game process does not run, and this inference conclusion is not present in existing characteristic before, will
This inference conclusion is saved in database as new characteristic, then by teenage in 10 music process operations of night and trip
Play process is not run and could open game with the precondition in knowledge base i.e. underage users b evenings 6~7 point and matched
Afterwards, obtain teenage in the operation of 10 music processes of night and game process does not run this inference conclusion, by this inference conclusion
Found when being saved in database as another new characteristic it is identical with existing characteristic in database, by it is teenage
10 music processes of night run and game process does not run this inference conclusion and exports as a result, now underage users b
Music process can be run but game process can not run on the client C for using, so as to reach during the game to underage users
Between the purpose supervised.Surf time or the browsed web content of underage users can similarly be supervised.
Calculating data are obtained in another technical scheme, in step 3 to specifically include：
S1：Self-defined initiation parameter, initiation parameter includes Population Size N, sub- Population Size M, maximum evaluation number of times
MAX_FE, functor and terminal symbol, mrna length, gene number, mutation probability, slotting string probability, slotting string length and restructuring are general
Rate, the original data definition that will be obtained is initial population Pt={ X1,X2,…,XN, and initial population is calculated according to computation rule
In each individual fitness fi。
S2：By performing selection, variation, slotting string and the restructuring of gene expression programming to population PtIn individuality
M new individual of generation, and M new individual is organized into sub- population Ot, O is calculated according to computation ruletIn each individuality adaptation
The maximum individuality of degree, wherein fitness is optimum individual.
S3：By population PtIn individual and sub- population OtIn the M individual interim population P ' t of composition, and according to calculating
Rule calculates each the individual fitness in interim population P ' t, then deletes in interim population P ' t before fitness maximum
M individual, obtains the population P of new generation being made up of individualityt+1。
S4：S2 to S3 is repeated until evaluating number of times and reaching and terminate after MAX_FE, at the end of the population P of new generation that obtainst+1I.e.
To calculate data.
Wherein, the ambiguity and uncertainty that are introduced for preferably expressing knowledge of fitness, the adaptation of the system
Degree section definition is [0,1], the size of fitness by the research of multidigit marriage counselor and system repeatedly experiment be combined by way of coming
Obtain.
Analysis process based on GEP is an adaptive process for dynamic learning, and the analysis result of many successes can be made
For new analysis rule is present, it is possible to which the analysis experience accumulated according to system abandons long-term invalid rule come automatic, so that
Optimization inference rule storehouse is reached, the purpose of accuracy of analysis is improved.
User behavior complicated or that number of samples is more can be analyzed with reference to this kind of GEP algorithms, first classify and divide again
Analysis, and respectively obtaining the fitness of each class makes overall data relatively reliable, the conclusion for analyzing more versatility, it is adaptable to more
Many crowds.
In another technical scheme, database includes data storage storehouse, knowledge base and rule base, in data storage storehouse
Be stored with initial data, calculating data and characteristic respectively.
In another technical scheme, knowledge base includes the IF-THEN rules generated by GEP algorithms, IF-THEN rules
Middle C=C (P), wherein, C is condition premised on conclusion P,N is every premise number of precondition, θ=
{ AND, OR }, is CF, CF=[0,1] by the credit assignment of every rule.Multiple premises are may include in each IF-THEN rules
Condition, potentially includes multiple premises in each precondition, multiple premises may include various and/or relation each other.
In another technical scheme, the progress information obtained in step one includes real-time process and historical progress；
Whether computation rule in step 2 includes the typing of the time of origin of corresponding real-time process and to performing correspondence
Real-time process judgement.
Subsequently be may determine that or according to real-time with reference to initial data and computation rule including real-time process and historical progress
The time of origin of process judges whether corresponding real-time process performs, and makes the information of collection more complete, enables follow-up reasoning
Historical data support is obtained, it is relatively reliable.
In another technical scheme, the computation rule in step 2 also includes the calculating in cycle and history in the cycle is entered
With the statistics of the generation frequency of same process in real-time process in journey.The analysis to user behavior in longer cycle is capable of achieving, is made
Analysis validity is more permanent reliable, and also only a period of time computer user's behavior can be analyzed, in long or short term can root
Flexibly selected the need for according to user.
In another technical scheme, when the personal information of computer user is obtained in step one, according in each account
The personal information for obtaining computer user corresponding with each computer user one-to-one ID, wherein, each calculating
Corresponding one or more accounts of machine user.Each computer is recognized according to each computer user one-to-one ID
User, it is to avoid cause the personal information and progress information of acquisition imperfect because computer user switches different accounts, make data
It is more complete reliable.
Embodiment 2
There is service end S in LAN, client a, b, c, d etc. are loaded with including knowledge base, characteristic database on S
Database, scheduling controller and inference machine with rule base.
Service end subscriber wishes the uncivil term on a certain network forum by service end S supervision client users,
Uncivil term is such as filtered out, the uncivil term m or n that such as client user delivers does not show, if continuous in 1 hour
Deliver 5 times and speech half an hour is then prohibited in speech of the above containing uncivil term, other are not limited then, and this information exists as rule storage
In knowledge base, i.e., precondition P is that speech contains uncivil term m or n in first IF-THEN rule, meet P can it is concluded that
C is that uncivil term m or n does not show, this rule confidence level CF for delivering is 1.Precondition in second IF-THEN rule
P be it is continuous in same user 1 hour deliver 5 times and speech of the above containing uncivil term, conclusion C is to prohibit speech half an hour.
When certain user delivers the uncivil term containing vocabulary m on network forum for the first time on client C, obtain original
Data are sometime delivering uncivil term m for certain user, and whether computation rule shows for the uncivil term m of interpretation, and opens
The frequency is 1 in beginning timing 1 hour；
By the GEP algorithms in knowledge base, it is that the uncivil term m that this user delivers is shown, this is used to obtain calculating data
The uncivil term m that family is delivered does not show, is that user can not deliver the unliterary of m containing vocabulary or n with the precondition in knowledge base
After bright term is matched, the inference conclusion for obtaining precondition is that the uncivil term m that this user delivers does not show, this reasoning
Conclusion is not present in existing characteristic, and database is saved in using this inference conclusion as new characteristic, then will
The uncivil term m that this user delivers does not show the uncivil term delivered with the precondition in knowledge base i.e. client user
M or n do not shown after being matched, and is obtained the uncivil term m that this user delivers and is not shown this inference conclusion, by this reasoning knot
It is identical with existing characteristic in database by being found when being saved in database as another new characteristic, by this user
The uncivil term m for delivering does not show and exports as a result.
When user delivers uncivil term again, it is only necessary to by 1 reasoning be can obtain this user deliver it is unliterary
The conclusion that bright term m does not show, it is seen that the computer user's behavior analysis method based on gene expression programming of the invention is
One adaptive process of dynamic learning.
When user delivered 5 uncivil terms in 1 hour, delivered according to regular this user of first IF-THEN
Uncivil term m does not show that it is taboo speech half an hour to obtain inference conclusion according to second IF-THEN rule, so as to according to multiple
The aspects such as regular preferably real-time current operation behavior, browsed web content to user carry out sensitivity analysis and supervision.Phase
Between, when user switches other accounts delivers uncivil term, because each user one ID of correspondence is multiple accounts
Between it is interrelated, in step one obtain computer user personal information when, statistical information including the letter in multiple accounts
Breath, even if therefore user switch other accounts and deliver uncivil term the data of frequency statistics can be also counted according to computation rule
In.
Although embodiment of the present invention is disclosed as above, it is not restricted to listed in specification and implementation method
With, it can be applied to various suitable the field of the invention completely, for those skilled in the art, can be easily
Other modification is realized, therefore under the universal limited without departing substantially from claim and equivalency range, the present invention is not limited
In specific details and shown here as the legend with description.
Claims (7)
1. a kind of computer user's behavior analysis method based on gene expression programming, it is characterised in that comprise the following steps：
Step one：Obtain the personal information and progress information of computer user；
Step 2：The computation rule in rule base is obtained by progress information correspondence, obtains right according to personal information and progress information
After the initial data answered, initial data is saved in database；
Step 3：GEP algorithms in the initial data and computation rule combination knowledge base that will obtain, obtain calculating data, will count
The evidence that counts is saved in database；
Step 4：After the calculating data that will be obtained are matched with the precondition in knowledge base, the reasoning of precondition is obtained
Conclusion, when inference conclusion from existing characteristic is different in database when, preserved inference conclusion as new characteristic
In database, replaced calculating Data duplication step 4 with inference conclusion, until inference conclusion and existing characteristic in database
According to it is identical when terminate；
Step 5：To as a result be exported with existing characteristic identical inference conclusion in database.
2. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
Calculating data are obtained in step 3 and specifically includes following steps：
S1：Self-defined initiation parameter, initiation parameter includes Population Size N, sub- Population Size M, maximum evaluation number of times MAX_
FE, functor and terminal symbol, mrna length, gene number, mutation probability, slotting string probability, slotting string length and recombination probability, will
The original data definition of acquisition is initial population Pt={ X1,X2,…,XN, and according to computation rule calculate initial population in each
Individual fitness fi；
S2：By performing selection, variation, slotting string and the restructuring of gene expression programming to population PtIn individual generation M
Individual new individual, and M new individual is organized into sub- population Ot, O is calculated according to computation ruletIn each individual fitness, its
The maximum individuality of middle fitness is optimum individual；
S3：By population PtIn individual and sub- population OtIn the M individual interim population P ' t of composition, and according to computation rule
Each the individual fitness in interim population P ' t is calculated, then delete fitness maximum in interim population P ' t preceding M
Individuality, obtains the population P of new generation being made up of individualityt+1；
S4：S2 to S3 is repeated until evaluating number of times and reaching and terminate after MAX_FE, at the end of the population P of new generation that obtainst+1As count
Count evidence.
3. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
The database of step 2 include data storage storehouse, knowledge base and rule base, the initial data that is stored with respectively in data storage storehouse,
Calculate data and characteristic.
4. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
The knowledge base of step 4 includes that IF-THEN is regular, C=C (P) in IF-THEN rules, wherein, C is conclusion, bar premised on P
Part,N is the premise number of every rule, θ={ AND, OR }, and confidence level per rule is CF, CF=[0,
1]。
5. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
The progress information obtained in step one includes real-time process and historical progress；
Whether computation rule in step 2 includes the typing of the time of origin of corresponding real-time process and to performing corresponding reality
The judgement of Shi Jincheng.
6. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
The calculating also including the cycle of computation rule in step 2 and in historical progress in the cycle with same process in real-time process
There is the statistics of the frequency.
7. computer user's behavior analysis method of gene expression programming is based on as claimed in claim 1, it is characterised in that
When the personal information of computer user is obtained in step one, used correspondingly with each computer user according in each account
Family mark correspondence obtains the personal information of computer user, wherein, corresponding one or more accounts of each computer user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611239027.1A CN106844515B (en) | 2016-12-28 | 2016-12-28 | Computer user behavior analysis method based on gene expression programming |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611239027.1A CN106844515B (en) | 2016-12-28 | 2016-12-28 | Computer user behavior analysis method based on gene expression programming |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106844515A true CN106844515A (en) | 2017-06-13 |
| CN106844515B CN106844515B (en) | 2020-05-05 |
Family
ID=59114390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611239027.1A Expired - Fee Related CN106844515B (en) | 2016-12-28 | 2016-12-28 | Computer user behavior analysis method based on gene expression programming |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106844515B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610772A (en) * | 2017-08-23 | 2018-01-19 | 广西师范学院 | A kind of thyroid nodule CT image diagnostic system design methods |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060200436A1 (en) * | 2005-03-07 | 2006-09-07 | Chi Zhou | Gene expression programming with enhanced preservation of attributes contributing to fitness |
| US20080307399A1 (en) * | 2007-06-05 | 2008-12-11 | Motorola, Inc. | Gene expression programming based on hidden markov models |
| CN104951847A (en) * | 2014-12-31 | 2015-09-30 | 广西师范学院 | Rainfall forecast method based on kernel principal component analysis and gene expression programming |
| CN105912887A (en) * | 2016-03-31 | 2016-08-31 | 安徽农业大学 | Classifying method of crop data based on fuzzy C mean value utilizing improved gene expression programming |
| CN105930648A (en) * | 2016-04-15 | 2016-09-07 | 郑州轻工业学院 | Gene expression programming (GEP) bidirectional prediction-based short elliptic arc fitting method |
| CN106250583A (en) * | 2016-07-15 | 2016-12-21 | 西安电子科技大学 | Dynamic job shop scheduling rule optimization methods based on double population gene expression programmings |
-
2016
- 2016-12-28 CN CN201611239027.1A patent/CN106844515B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060200436A1 (en) * | 2005-03-07 | 2006-09-07 | Chi Zhou | Gene expression programming with enhanced preservation of attributes contributing to fitness |
| US20080307399A1 (en) * | 2007-06-05 | 2008-12-11 | Motorola, Inc. | Gene expression programming based on hidden markov models |
| CN104951847A (en) * | 2014-12-31 | 2015-09-30 | 广西师范学院 | Rainfall forecast method based on kernel principal component analysis and gene expression programming |
| CN105912887A (en) * | 2016-03-31 | 2016-08-31 | 安徽农业大学 | Classifying method of crop data based on fuzzy C mean value utilizing improved gene expression programming |
| CN105930648A (en) * | 2016-04-15 | 2016-09-07 | 郑州轻工业学院 | Gene expression programming (GEP) bidirectional prediction-based short elliptic arc fitting method |
| CN106250583A (en) * | 2016-07-15 | 2016-12-21 | 西安电子科技大学 | Dynamic job shop scheduling rule optimization methods based on double population gene expression programmings |
Non-Patent Citations (1)
| Title |
|---|
| 李秀喜等: "基于GEP的化工过程故障诊断知识抽取", 《2009中国过程系统工程年会暨中国MES年会论文集》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610772A (en) * | 2017-08-23 | 2018-01-19 | 广西师范学院 | A kind of thyroid nodule CT image diagnostic system design methods |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106844515B (en) | 2020-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alihodzic et al. | Improved bat algorithm applied to multilevel image thresholding | |
| Saputra et al. | Analysis resilient algorithm on artificial neural network backpropagation | |
| CN106570597A (en) | Content popularity prediction method based on depth learning under SDN architecture | |
| CN112153426B (en) | Content account management method and device, computer equipment and storage medium | |
| CN110188272A (en) | A kind of community's question and answer web site tags recommended method based on user context | |
| CN106777402B (en) | A kind of image retrieval text method based on sparse neural network | |
| Wang et al. | Echo state networks regulated by local intrinsic plasticity rules for regression | |
| CN109902823A (en) | A kind of model training method and equipment based on generation confrontation network | |
| Biswas et al. | Classification of indian dance forms using pre-trained model-vgg | |
| Moroz et al. | Hybrid sorting-out algorithm COMBI-GA with evolutionary growth of model complexity | |
| CN117473041A (en) | Programming knowledge tracking method based on cognitive strategy | |
| Fil et al. | Minimal spiking neuron for solving multilabel classification tasks | |
| Volna et al. | Training set fuzzification based on histogram to increase the performance of a neural network | |
| CN106844515A (en) | Computer user's behavior analysis method based on gene expression programming | |
| CN116186378A (en) | Package recommendation method, package recommendation device, electronic equipment and storage medium | |
| CN112529637B (en) | Service demand dynamic prediction method and system based on context awareness | |
| van Knippenberg et al. | Evolutionary construction of convolutional neural networks | |
| Husain et al. | Forecasting model of Covid-19 cases using fuzzy time series using persentage change | |
| Petrovic-Lazarevic et al. | Neuro-fuzzy modelling in support of knowledge management in social regulation of access to cigarettes by minors | |
| Cortez et al. | An evolutionary artificial neural network time series forecasting system | |
| Mostafa et al. | Estimate of stochastic model parameter of exchange rate using machine learning techniques | |
| CN106911462B (en) | Wireless router password analysis method based on gene expression programming | |
| Zhao et al. | Terminal replacement prediction based on deep belief networks | |
| Thammasiri et al. | Imbalance classification model for churn prediction | |
| Zhao et al. | A genetic algorithm for reinforcement learning |
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 | ||
| CB02 | Change of applicant information |
Address after: 530001 Nanning Normal University, 175 Mingxiu East Road, Nanning City, Guangxi Zhuang Autonomous Region Applicant after: NANNING NORMAL University Applicant after: GUANGXI TUMOUR PREVENTION AND CONTROL Research Institute Address before: Mingxiu Road East of Nanning city the Guangxi Zhuang Autonomous Region 530001 Guangxi Teachers Education University No. 175 Applicant before: Guangxi Normal University Applicant before: GUANGXI TUMOUR PREVENTION AND CONTROL Research Institute |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200505 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |