CN1808483A - Biologic heuristic calculation based floating population information analysis and alert and alarm method - Google Patents
Biologic heuristic calculation based floating population information analysis and alert and alarm method Download PDFInfo
- Publication number
- CN1808483A CN1808483A CNA2006100201491A CN200610020149A CN1808483A CN 1808483 A CN1808483 A CN 1808483A CN A2006100201491 A CNA2006100201491 A CN A2006100201491A CN 200610020149 A CN200610020149 A CN 200610020149A CN 1808483 A CN1808483 A CN 1808483A
- Authority
- CN
- China
- Prior art keywords
- population
- algorithm
- floating population
- data
- polygenes
- 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
Images
Abstract
The invention provides a floating population information analysis and warning and alarming method based on biological instructive calculation. Wherein, with current floating population information system from department of public safety, extracting need information to calculate floating population data with the multilayer chromosome gene expression programming algorithm and overlap-expressive polygene evolution algorithm in biological instructive calculation filed; and completing data mine and classification and isolated-point analysis to realize the warning and alarm. This invention can take deep analysis to trace and trend of abnormal floating population and different criminal feature, builds a intelligent analysis platform, and can also provide useful information for local peace.
Description
One, technical field
The invention belongs to the information analysis technical field of public safety, be specifically related to a kind of inspire calculating and find and excavate the flow pattern of all kinds of personnel among the floating population and floating population's tendency problem based on biology, find out unusual flowing information and pattern, according to all kinds of criminal features case is carried out the similarity comparison, and realize intelligent analysis method pre-, that report to the police.
Two, background technology
Along with sustainable development of economy, the total amount sharp increase of urban floating population, all kinds of public security relevant with the floating population, crime rate have surpassed city permanent resident population's incidence of cases at present, have constituted main threat to the city especially social security of coastal economy developed regions.Sharp increase like this, the huge high fluidity population cluster of quantity, the high crime rate of its appearance is added characteristics such as data scarcity, investigation difficulty, has proposed stern challenge for current city integrated security administration work.Thereby study the intelligentized floating population's analytical approach of a cover, timely investigation of potential offender and prediction and alarm are just seemed very necessary.
At present for the highest places such as hotel of floating population's occurrence rate, though public security organ takes strict management system to guarantee the public security demand always, as the passenger who goes into to stay at hotels is required to carry out the registration of detailed information such as passenger's identity, but because the registration of information such as these passenger's identity is all manually carried out, public security organ will retrieve it according to case, not only speed is slow, program is loaded down with trivial details, easily make mistakes, and cost height, also easily because of improper storage causes data at random, incomplete, filing has brought difficulty to file maintenance.Secondly, on business peace office pursues and captures an escaped prisoner, hits escaped convict and reconnoitring and predicting the criminal investigation case, mainly be by provide listing inspection as wanted to hotel security department, browse check-in list or passenger's check-in is thin verifies by the front clerk again, the mode that reports step by step that discovers a clue is carried out.Obviously, this mode, on the one hand along with hotel quantity rolls up, the expanded range of deploying to ensure effective monitoring and control of illegal activities, ordering to arrest inspection often sends out less than all hotels, cause the blind spot on deploying to ensure effective monitoring and control of illegal activities, two aspects are owing to order to arrest inspection quantity cumulative year after year, thisly browse information verification by hand by waiter, have the situation of failing to report and misrepresenting deliberately to take place unavoidably, waste time and energy simultaneously, turnover period is long, can not in time provide suspicion object and the clue that to search, give offender's formation time difference easily and hide oneself and stop over or escape and chase to public security organ.
It is reported that in order to address the above problem, existing relative technological products has at present:
1. " the administration of public order at hotels infosystem " of Chengdu Jia De Li Co., Ltd exploitation.This system has proposed administration of public order at hotels infosystem integrative solution, has hotel's management, information acquisition, functions such as warning, supervision and management of deploying to ensure effective monitoring and control of illegal activities, and has realized ladder of management digitizing, work efficiency informationization.Though this system can satisfy the management of hotel to passenger information, can satisfy the characteristics of public security organ again to the real-time of people information processing, but this system does not utilize existing passenger, hotel information to carry out profound level analysis, potential offender is predicted, reports to the police, for cracking of cases is given a clue, auxiliary public security organ carries out the detection of case and the decision-making commander that crime was hit, taken precautions against in public security organ is provided support.
2. " the hotel industry security administration system " of Gao Ke softcom limited of Wuhan Beijing University exploitation.As one of the service sub-system in the whole nation " police management information system ", be the important technical of public security department to floating population's management.This system adopts earlier to call, and typing personnel's passport NO. and room number are finished checking in of lodging personnel, and system is transferred to register information in the public security net by communication middleware and parameter network automatically then.Though this system has solved several problems that long-term puzzlement " hotel industry security administration system " is built, huge as foreground software implementation cost, running cost is uncontrollable, problems such as PC version software operation complexity, especially report late, fail to report for the lodging personnel's at little hotel information in addition the solution effect of problem such as concealed and unreported better.But it can not be moved in personal information to the hotel equally and carry out intelligent analysis, and then realizes that prediction, warning to potential offender, auxiliary public security organ carry out the detection of case and the decision-making commander of public security organ is provided support.
Three, summary of the invention
The objective of the invention is problem, a kind of floating population information analysis and pre-, alarm method that calculates that inspire based on biology is provided at the prior art existence.
Method provided by the invention is to be based upon biological the inspiration to calculate on the data mining technology basis with efficient, accurate and mass data processing ability.The biological inspiration calculated (Bio-inspired Computing) and is meant that various spontaneous phenomenons or process with organic sphere are inspiration, and biological phenomenon or process are simulated and a series of heuristic intelligence computation method that proposes.As evolution algorithm, artificial neural network, cellular automaton etc.Important branch comprised during current biological inspiration was calculated: genetic algorithm, immune algorithm, neural calculating, ant algorithm and gene expression programming.Data mining technology is expansion and the in-depths of people to data management, storage and retrieval technique, it is intended to from real application data a large amount of, incomplete, noisy, fuzzy, at random, extract lie in wherein, people are ignorant in advance but be the information of potentially useful and the process of knowledge.
Floating population information analysis and general plotting pre-, alarm method based on biology inspiration calculating provided by the invention is to utilize the system of the grasp floating population information that has public security organ's foundation now, therefrom extract the information that needs, adopt the multiple biological algorithm that calculates that inspires, finish the excavation classification and the isolated point analysis of data, and, in real time floating population's data stream is carried out in advance, reported to the police according to the isolated point analysis result.The step of this method comprises:
Step by foreground, hotel microcomputer typing lodging passenger information;
Utilize public network the lodging passenger information of typing to be transferred to the step of Communications Processor Module;
Communications Processor Module is compared data transmission to comparison server and blacklist data of database step;
By Communications Processor Module with the step of data maintenance to floating population's database;
It is characterized in that this method also comprises:
Utilize data pick-up conversion load-on module to finish load step from floating population's database to floating population's data warehouse;
Produce the sample data that is used to classify according to floating population's data warehouse, be transferred to simultaneously based on the gene expression programming algoritic module of multi-layer dyeing body with based on the polygenes evolution algorithm module of overlapping expression and carry out computational analysis, and produce the step of corresponding sorter;
Floating population's data of database is transferred to the gene expression programming algorithm classification device of multi-layer dyeing body respectively and sorts out, and produce the step of isolated point tabulation based on the polygenes evolution algorithm sorter of overlapping expression;
Communications Processor Module is transferred to the step that the warning processing module is carried out Realtime Alerts with data transmission after comparing module and blacklist data of database are compared; With
The data that spread out of of comparison server are transferred to successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression judge whether step into isolated point;
To carry out the step of Realtime Alerts to the warning processing module based on the gene expression programming algorithm classification device of multi-layer dyeing body and the information transmission that is judged as isolated point based on the polygenes evolution algorithm sorter of overlapping expression, not, the step that finishes just.
The step that produces corresponding sorter with the gene expression programming algoritic module of multi-layer dyeing body in the said method comprises:
According to user's initial conditions, select the training sample data and produce the step of training set from floating population's data warehouse;
Gene expression programming algorithm parameter according to the multi-layer dyeing body carries out initialized step to population;
According to the gene expression programming algorithm of multi-layer dyeing body, each individuality is to the step of the fitness of training set in the calculating population;
The individuality that the reservation fitness ranks the first is to follow-on step;
Population is carried out the step of basic genetic operator operation;
Successively population is carried out recombinate the at random step of operator and chromosome reorganization operator of gene;
With operating result ordering and copy to follow-on step;
If the result that preferably evolves satisfies accuracy requirement or reaches default maximum generation's number, then carry out next step, otherwise go on foot the step of carrying out cycling to the 3rd;
With the step of optimal adaptation degree individuality as the algorithm classification device.
The step of using polygenes evolution algorithm module based on overlapping expression to produce corresponding sorter in the said method comprises:
According to user's initial conditions, select the training sample data and produce the step of training set from floating population's data warehouse;
According to polygenes evolution algorithm parameter population is carried out initialized step based on overlapping expression;
According to the polygenes evolution algorithm based on overlapping expression, each individuality promptly calls the step of evaluation function to the fitness of training set in the calculating population;
The individuality that the reservation fitness ranks the first is to follow-on step;
Population is carried out the step of basic genetic operator operation;
With operating result ordering and copy to follow-on step;
If the optimized individual fitness satisfies accuracy requirement or reaches default maximum generation's number, then carry out next step, go on foot the step of carrying out cycling otherwise get back to the 3rd;
With the step of optimal adaptation degree individuality as this algorithm classification device.
The step of Realtime Alerts comprises in the said method:
Real-time floating population's data stream that transmission comes to Communications Processor Module is finished the successful information of comparison blacklist database comparison by comparing module, shows step red, brown, orange color tips according to the case-involving order of severity;
Real-time floating population's data stream that transmission comes to Communications Processor Module, finish comparison blacklist database by comparing module and do not compare successful information, use classifying of having trained successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression, be found to be isolated point, promptly show the step of yellow color tips;
Otherwise withdraw from, show the step of the green color tips of not reporting to the police.
The present invention has the following advantages:
1, owing to utilized the moving message breath of biological inspiration computing technique convection current to analyze in the inventive method, thereby find all kinds of personnel's flow pattern and floating population's tendency problem for public security organ, find out unusual flowing information and pattern, and further unusual flow of personnel track and trend, all kinds of criminal feature are carried out profound level analysis, set up intellectualized analysis platform, filled up the blank that real-time intelligent is pre-, report to the police.
2, since in the inventive method the gene expression programming algorithm of the multi-layer dyeing body that utilizes have in the population each individuality and form by a plurality of chromosomes; Different chromosome has kinds of characters collection and length; A plurality of chromosomes are divided into many levels, the upper strata chromogene can call the characteristics of the result of calculation of lower floor's chromogene, thereby compare with polygenes GEP algorithm with traditional single-gene GEP, average evolution generation number only (is seen Peng Jing, Tang Changjie, Li Chuan for its 29 ~ 81%, .M-GEP recklessly for army building: based on the evolutionary programming algorithm of multi-layer dyeing body gene expression programming, Chinese journal of computers, 2005,28 (9) .1459-1466.).
3, be used for hereditary individuality by a plurality of genomic constitutions because the polygenes evolution algorithm based on overlapping expression that utilizes in the inventive method has, the fragment of constitutivegene can repeat to express under certain condition; Because fragment can repeat to express, its expression of space efficient is better than other algorithms; Need not gene or chromosome content are made the characteristics of constraint, thereby all must make certain requirement for restriction to the gene form with traditional GP and GEP algorithm, owing to reduced restriction, its efficiency of algorithm can increase substantially, experiment shows under equal conditions, based on the speed of the polygenes evolution algorithm of overlapping expression 2.8 to 9.7 times of GEP, and compare with GEP, (see Jing Peng based on the polygenes evolution algorithm EAOGE of overlapping expression having a more substantial increase aspect the success ratio of function discover, Chang-jie Tang, Jing Zhang, and Chang-an Yuan.Evolutionary Algorithm Based on Overlapped Gene Expression, ICNC 2005, and LNCS 3612, pp.194-204,2005.).
4, inspire computing technique to be applied to case biology and detect work, thereby compare with traditional investigation pattern, on the one hand can be quick, accurately and in bulk produce analysis result, and collect evidence by inquiry and solve a case rapidly according to analysis result, hitting all kinds of illegal activities for public security organ grasps the opportunity in advance, the initiative of assurance work, can be prevention on the other hand and hit all kinds of criminal offences, controlling regional public security overall situation provides the information of usefulness, strengthen early warning, susceptibility, construction suggestion for decision-making commander provider tropism and directiveness makes it more effectively to take precautions against and fight crime.
5, because the biology of the inventive method utilization inspires calculating to carry out deep excavation to floating population's data, thereby make cunning offender also be unable to escape the analysis and judgement of this method, greatly improve public security organ's case handling efficiency.
6, use the inventive method can reduce the cost of handling a case greatly.At present, the cost of on average handling a case of arresting a runaway convict is 10,000 yuan, and according to Ministry of Public Security's statistics, it is more than 10 ten thousand that runaway convict's number was arrested in 1 year in the whole nation, and the cost of handling a case is more than 10 hundred million.If adopt this advanced technological means, make the cost of handling a case of arresting each runaway convict reduce by 10%, the whole nation then can reduce in 1 year handles a case cost more than 100,000,000 yuan.
7, the inventive method has the scalability of height, not only can be used for the analysis to hotel floating population's information, also can be used for the analysis of floating population's information in places such as market, warehouse, airport and bank, applied range.
Four, description of drawings
Fig. 1 is the floating population's analytic system instance graph that makes up according to the inventive method;
Fig. 2 is that the inventive method is to floating population information analysis overview flow chart pre-to realize, that report to the police;
Fig. 3 is for producing the process flow diagram of corresponding sorter based on the gene expression programming algoritic module of multi-layer dyeing body in the inventive method;
Fig. 4 is for producing the process flow diagram of corresponding sorter based on the polygenes evolution algorithm module of overlapping expression in the inventive method.
Five, embodiment
Also the invention will be further described to provide embodiment below in conjunction with accompanying drawing; but be pointed out that following examples can not be interpreted as limiting the scope of the invention, the person skilled in the art of this area foregoing according to the present invention still belongs to protection scope of the present invention to some nonessential improvement and the adjustment that the present invention makes.
The floating population's analytic system that makes up according to the method for the invention as shown in Figure 1.This system can be divided into foreground, transmission, three parts in backstage.Wherein the foreground comprises the typing terminal 1 that is installed in each foreground, hotel, and this typing terminal 1 adopts common computer, but networking; Hop comprises the networked devices of foreground typing terminal, outer net networked devices 2 as ADSL, center, backstage, as router or three-tier switch, the outer net communication server 3 and the Intranet communication server 5, physical isolator and firewall box 4 and Intranet networked devices, promptly switch 12; Back partition comprises: database server 6, data warehouse server 7, training server 8, comparison server 9, Alarm Server 10 and management platform 11.
Floating population's data that foreground, hotel typing terminal 1 is gathered are saved in database server 6 by hop 2 ~ 5,12; Then the content conversion of database server 6 is extracted and be loaded into data warehouse server 7; Training server 8 utilizes the content of data warehouse 7 according to the instruction of management platform 11, trains sorter, and sorter is analyzed data-base content again, obtains the isolated point tabulation and shows the user.
The process that data are reported to the police is: at first the sorter that will have been trained by training server 8 copies to comparison server 9; Comparing server 9 then compares to each bar data of being submitted to by the communication server 3,5 according to the blacklist data in sorter and the database, if the data that discovery meets then will write down and relevant comparison situation and alert levels information transmission arrives Alarm Server 10; Give corresponding alarm by Alarm Server 10 according to alert levels at last.
The alert levels information of utilizing native system to provide, personnel concerning the case, wanted circular personnel or the personnel's that deploy to ensure effective monitoring and control of illegal activities of great runaway convict, weight, spy, major case in the blacklist data whereabouts both can have in time been found, be convenient in time implement to arrest, can divide it alert levels of isolated point into by native system again, find the complete different personnel of some lodging characteristic and common passenger, as utilizing the hotel to implement the clique of swindle, the user can classify it as suspect object investigation immediately, thereby gives a clue for rapid clear up a criminal case.Native system can also improve the data acquisition quality by assisted user, because project is imperfect or the data of the serious deviation of content (as move in date be 10 years before) can be divided into isolated point by native system equally, the user can utilize it to find the quality problems of foreground, hotel logging data and adjusted in time, correct.
(1) overall procedure of the inventive method
Floating population information analysis that inspiration is calculated based on biology provided by the invention, as to be used for said system and overall procedure pre-, alarm method, as shown in Figure 2, concrete steps are as follows:
1, the step by foreground, hotel microcomputer typing lodging passenger information, lodging passenger information comprise the administrative division at personnel's name, sex, age, identification card number, inhabitation hotel, affiliated local police station, inhabitation hotel numbering and title, move in time, time departure, room number, live people, photo etc. together.
2, utilize public network the lodging passenger information of typing to be transferred to the step of Communications Processor Module.
3, by Communications Processor Module with the step of data maintenance to floating population's database.
4, utilize data pick-up conversion load-on module to finish load step from floating population's database to floating population's data warehouse.Database and data warehouse all can adopt oracle database, and data ETL process can realize that by general ETL software also can finish by custom program, purpose is to remove the imperfect and illegal data of project, then data load is arrived data warehouse.
5, produce the sample data be used to classify according to floating population's data warehouse, the sample data field comprises: the numbering and the title at the administrative division at sex, age, identification card number, inhabitation hotel, affiliated local police station, inhabitation hotel, move in time, time departure, room number etc.Be transferred to simultaneously based on the gene expression programming algoritic module of multi-layer dyeing body with based on the polygenes evolution algorithm module of overlapping expression and carry out computational analysis, and produce the step of corresponding sorter.
6, floating population's data of database is transferred to respectively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression and sorts out, and produce the step of isolated point tabulation.The isolated point tabulation can be the prevention and control of user's public security and cracking of cases is given a clue.
Simultaneously, Communications Processor Module step that data transmission is compared to comparison server and blacklist data of database.
7, Communications Processor Module is transferred to the step that the warning processing module is carried out Realtime Alerts with data transmission after comparing module and blacklist data of database are compared success, if do not have success, then
8, will compare data that server spreads out of is transferred to successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression and judges whether step into isolated point.
9, will carry out the step of Realtime Alerts to the warning processing module based on the gene expression programming algorithm classification device of multi-layer dyeing body and the information transmission that is judged as isolated point based on the polygenes evolution algorithm sorter of overlapping expression, not, the step that finishes just.
The step of Realtime Alerts comprises again:
1) real-time floating population's data stream that transmission comes to Communications Processor Module is finished the successful information of comparison blacklist database comparison by comparing module, shows step red, brown, orange color tips according to the case-involving order of severity.
2) real-time floating population's data stream that transmission comes to Communications Processor Module, finish comparison blacklist database by comparing module and do not compare successful information, use classifying of having trained successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression, be found to be isolated point, promptly show the step of yellow color tips.
3) step that otherwise withdraw from, shows the green color tips of not reporting to the police.
The warning of 5 kinds of different colours that present embodiment is used, it is expressed as respectively:
● red (especially big) represents especially big cases.As the A of Ministry of Public Security level runaway convict, or the case-involving member of particularly serious case.
● the case-involving member of important cases such as palm fibre (greatly) representative is killed a person, robbery.
● orange (in) the general case wanted circular personnel of expression or the personnel that deploy to ensure effective monitoring and control of illegal activities.
● yellow (little) abnormal information, the i.e. isolated point that obtains by intelligent analysis method analysis and excavation.
● green (safety) is promptly without any warning.
(2) produce the flow process of corresponding sorter with the gene expression programming algoritic module of multi-layer dyeing body
The flow process that produces corresponding sorter with the gene expression programming algoritic module of multi-layer dyeing body as shown in Figure 3, concrete steps comprise:
1,, selects the training sample data and produce the step of training set from floating population's data warehouse according to user's initial conditions.The sample data field can according to user input from the numbering and the title at the administrative division at sex, age, identification card number, inhabitation hotel, affiliated local police station, inhabitation hotel, move in time, time departure, the room number etc. and select.
2, import the gene expression programming algorithm parameter of multi-layer dyeing body according to the user: chromosome quantity, gene dosage, population size, single-point recombination fraction, multiple spot recombination fraction, aberration rate, transport ratio, root transport ratio, gene transport ratio, genetic recombination rate, gene be recombination fraction, chromosome recombination fraction, precision, maximum generation's number at random, and population is carried out initialized step.
3, according to the gene expression programming algorithm of multi-layer dyeing body, each individuality is to the step of the fitness of training set in the calculating population, and the fitness formula is as follows:
Wherein K is the fitness standard factor, and subscript i shows i individual j in the population not and represents j use-case in the fitness use-case, C
tThe number of expression fitness use-case, C
(i, j)Represent the result that obtains after i j test case of individual substitution, T
j(j=1,2 ..., C
t) target function value of expression in the test case.
4, keep individuality that fitness ranks the first to follow-on step.
5, to population makes a variation, basic genetic operators such as displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination are operated step.The probability selection algorithm that the basic genetic operator adopts when selecting the effect in the population individual is the roulette algorithm.The formula that the roulette algorithm adopts is as follows:
Wherein α is non-0 positive number, even guarantee that fitness is 0, individual I
kStill has certain probability of choosing.The fitness score of f (x) expression x individuality.
6, successively population is carried out recombinate the at random step of operator and chromosome reorganization operator of gene, wherein recombinate the at random calculation procedure of operator of gene is as follows:
1) calculating need be by the back algebraically of this operator generation;
2) individuality of selecting two needs to recombinate at random according to the mode of roulette;
3) with the gene be unit, produce the chromosome recombinable site at random;
4) all genes behind the recombinable site of two individualities of exchange;
5) produce two offspring's individualities;
6) repeat to go on foot 2～step 5, specify back algebraically up to reaching the step 1.
The calculation procedure of chromosome reorganization operator is as follows:
1) calculating need be by the back algebraically of this operator generation;
2) individuality of selecting two needs to recombinate at random according to the mode of roulette;
3) whether two all individual genes of reorganization are selected to exchange at random;
4) generate two offspring's individualities with selection result;
5) repeat to go on foot 2～step 5, specify back algebraically up to reaching the step 1.
7, with operating result ordering and copy to follow-on step.
If 8 results that preferably evolve satisfy accuracy requirement or reach default maximum generation's number, then carry out next step, otherwise the step of carrying out cycling to the 3rd step.
9, with the step of optimal adaptation degree individuality as the algorithm classification device.
(3) use the flow process that produces corresponding sorter based on the polygenes evolution algorithm module of overlapping expression
Use flow process that polygenes evolution algorithm module based on overlapping expression produces corresponding sorter as shown in Figure 4, concrete steps comprise:
1,, selects the training sample data and produce the step of training set from floating population's data warehouse according to user's initial conditions; The sample data field can according to user input from the numbering and the title at the administrative division at sex, age, identification card number, inhabitation hotel, affiliated local police station, inhabitation hotel, move in time, time departure, the room number etc. and select.
2, import the polygenes evolution algorithm parameter of overlapping expression according to the user: gene dosage, population size, single-point recombination fraction, multiple spot recombination fraction, aberration rate, transport ratio, root transport ratio, gene transport ratio, genetic recombination rate, precision, maximum generation's number carry out initialized step to population.
3, according to the polygenes evolution algorithm based on overlapping expression, each individuality promptly calls the step of evaluation function to the fitness of training set in the calculating population.Concrete formula is identical with the gene expression programming algorithm of multi-layer dyeing body.
4, keep individuality that fitness ranks the first to follow-on step.
5, to population makes a variation, basic genetic operators such as displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination are operated step.The probability selection algorithm that the basic genetic operator adopts when selecting the effect in the population individual is the probability selection algorithm of roulette algorithm or population density.The roulette algorithm is identical with the gene expression programming algorithm of multi-layer dyeing body, and the probability selection algorithm formula of population density is as follows:
Wherein α is non-0 positive number, even guarantee that fitness is 0, individual I
kStill has certain probability of choosing.The fitness score of f (x) expression x individuality.
6, with operating result ordering and copy to follow-on step;
If 7 optimized individual fitness satisfy accuracy requirement or reach default maximum generation's number, then carry out next step, go on foot the step of carrying out cycling otherwise get back to the 3rd;
8, with the step of optimal adaptation degree individuality as this algorithm classification device.
The specific algorithm details can also list of references " Peng Jing, Tang Changjie, Li Chuan; .M-GEP recklessly for army building: based on the evolutionary programming algorithm of multi-layer dyeing body gene expression programming, Chinese journal of computers, 2005; 28 (9) .1459-1466. " and " JingPeng; Chang-jie Tang, Jing Zhang, and Chang-an Yuan.Evolutionary AlgorithmBased on Overlapped Gene Expression; ICNC 2005; LNCS 3612, pp.194-204,2005. "
Claims (10)
1, a kind of floating population information analysis that inspiration is calculated based on biology and pre-, alarm method comprise:
Step by foreground, hotel microcomputer typing lodging passenger information;
Utilize public network the lodging passenger information of typing to be transferred to the step of Communications Processor Module;
Communications Processor Module is compared data transmission to comparison server and blacklist data of database step;
By Communications Processor Module with the step of data maintenance to floating population's database;
It is characterized in that this method also comprises:
Utilize data pick-up conversion load-on module to finish load step from floating population's database to floating population's data warehouse;
Produce the sample data that is used to classify according to floating population's data warehouse, be transferred to simultaneously based on the gene expression programming algoritic module of multi-layer dyeing body with based on the polygenes evolution algorithm module of overlapping expression and carry out computational analysis, and produce the step of corresponding sorter;
Floating population's data of database is transferred to respectively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression sorts out, and produce the step of isolated point tabulation;
Communications Processor Module is transferred to the step that the warning processing module is carried out Realtime Alerts with data transmission after comparing module and blacklist data of database are compared; With
The data that spread out of of comparison server are transferred to successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression judge whether step into isolated point;
To carry out the step of Realtime Alerts to the warning processing module based on the gene expression programming algorithm classification device of multi-layer dyeing body and the information transmission that is judged as isolated point based on the polygenes evolution algorithm sorter of overlapping expression, not, the step that finishes just.
2, floating population information analysis and pre-, the alarm method that inspire to calculate based on biology according to claim 1 is characterized in that the step that gene expression programming algoritic module based on the multi-layer dyeing body produces corresponding sorter comprises:
According to user's initial conditions, select the training sample data and produce the step of training set from floating population's data warehouse;
According to gene expression programming algorithm parameter population is carried out initialized step based on the multi-layer dyeing body;
According to the gene expression programming algorithm based on the multi-layer dyeing body, each individuality is to the step of the fitness of training set in the calculating population;
The individuality that the reservation fitness ranks the first is to follow-on step;
Population is carried out the step of basic genetic operator operation;
Successively to population carry out gene recombinate at random calculate in the step of chromosome reorganization operator;
With operating result ordering and copy to follow-on step;
If the result that preferably evolves satisfies accuracy requirement or reaches default maximum generation's number, then carry out next step, otherwise go on foot the step of carrying out cycling to the 3rd;
With the step of optimal adaptation degree individuality as the algorithm classification device.
3, floating population information analysis and pre-, the alarm method that inspire to calculate based on biology according to claim 1 is characterized in that the step that polygenes evolution algorithm module based on overlapping expression produces corresponding sorter comprises:
According to user's initial conditions, select the training sample data and produce the step of training set from floating population's data warehouse;
According to polygenes evolution algorithm parameter population is carried out initialized step based on overlapping expression;
According to the polygenes evolution algorithm based on overlapping expression, each individuality promptly calls the step of evaluation function to the fitness of training set in the calculating population;
The individuality that the reservation fitness ranks the first is to follow-on step;
Population is carried out the step of basic genetic operator operation;
With operating result ordering and copy to follow-on step;
If the optimized individual fitness satisfies accuracy requirement or reaches default maximum generation's number, then carry out next step, go on foot the step of carrying out cycling otherwise get back to the 3rd;
With the step of optimal adaptation degree individuality as this algorithm classification device.
4, according to claim 1 or the 2 or 3 described floating population information analysis that inspiration is calculated based on biology and pre-, alarm methods, it is characterized in that the step of Realtime Alerts comprises:
Real-time floating population's data stream that transmission comes to Communications Processor Module is finished the successful information of comparison blacklist database comparison by comparing module, shows step red, brown, orange color tips according to the case-involving order of severity;
Real-time floating population's data stream that transmission comes to Communications Processor Module, finish comparison blacklist database by comparing module and do not compare successful information, use classifying of having trained successively based on the gene expression programming algorithm classification device of multi-layer dyeing body with based on the polygenes evolution algorithm sorter of overlapping expression, be found to be isolated point, promptly show the step of yellow color tips;
Otherwise withdraw from, show the step of the green color tips of not reporting to the police.
5, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 2, it is characterized in that when the gene expression programming algorithm classification device that produces based on the multi-layer dyeing body basic genetic operator operation that population is carried out has variation, displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination.
6, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 4, it is characterized in that when the gene expression programming algorithm classification device that produces based on the multi-layer dyeing body basic genetic operator operation that population is carried out has variation, displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination.
7, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 3, it is characterized in that in the polygenes evolution algorithm sorter process that produces based on overlapping expression the basic genetic operator operation that population is carried out has variation, displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination.
8, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 4, it is characterized in that in the polygenes evolution algorithm sorter process that produces based on overlapping expression the basic genetic operator operation that population is carried out has variation, displacement, root displacement, single-point reorganization, 2 reorganization, gene displacement, genetic recombination.
9, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 3, it is characterized in that in the polygenes evolution algorithm sorter process that produces based on overlapping expression the probability selection algorithm that the basic genetic operator adopts when selecting the effect in the population individual is the probability selection algorithm of roulette algorithm or population density.
10, floating population information analysis and pre-, the alarm method that calculates that inspire based on biology according to claim 4, it is characterized in that in the polygenes evolution algorithm sorter process that produces based on overlapping expression the probability selection algorithm that the basic genetic operator adopts when selecting the effect in the population individual is the probability selection algorithm of roulette algorithm or population density.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100201491A CN1808483A (en) | 2006-01-16 | 2006-01-16 | Biologic heuristic calculation based floating population information analysis and alert and alarm method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100201491A CN1808483A (en) | 2006-01-16 | 2006-01-16 | Biologic heuristic calculation based floating population information analysis and alert and alarm method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1808483A true CN1808483A (en) | 2006-07-26 |
Family
ID=36840374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006100201491A Pending CN1808483A (en) | 2006-01-16 | 2006-01-16 | Biologic heuristic calculation based floating population information analysis and alert and alarm method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1808483A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317771A (en) * | 2009-04-28 | 2015-01-28 | 遗传融资(巴巴多斯)有限公司 | Class-based distributed evolutionary algorithm for asset management and trading |
| CN106156461A (en) * | 2015-03-27 | 2016-11-23 | 北京市天元网络技术股份有限公司 | A kind of analysis method and system carrying out movement of population based on mobile communications network |
| CN109840472A (en) * | 2018-12-24 | 2019-06-04 | 航天信息股份有限公司 | Foreigner's passport information checking method and system |
| US10430429B2 (en) | 2015-09-01 | 2019-10-01 | Cognizant Technology Solutions U.S. Corporation | Data mining management server |
-
2006
- 2006-01-16 CN CNA2006100201491A patent/CN1808483A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317771A (en) * | 2009-04-28 | 2015-01-28 | 遗传融资(巴巴多斯)有限公司 | Class-based distributed evolutionary algorithm for asset management and trading |
| CN104317771B (en) * | 2009-04-28 | 2017-12-01 | 思腾科技（巴巴多斯）有限公司 | Class-based distributed evolutionary system for asset management and transaction |
| CN106156461A (en) * | 2015-03-27 | 2016-11-23 | 北京市天元网络技术股份有限公司 | A kind of analysis method and system carrying out movement of population based on mobile communications network |
| US10430429B2 (en) | 2015-09-01 | 2019-10-01 | Cognizant Technology Solutions U.S. Corporation | Data mining management server |
| US11151147B1 (en) | 2015-09-01 | 2021-10-19 | Cognizant Technology Solutions U.S. Corporation | Data mining management server |
| CN109840472A (en) * | 2018-12-24 | 2019-06-04 | 航天信息股份有限公司 | Foreigner's passport information checking method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111988285B (en) | Network attack tracing method based on behavior portrait | |
| Madhuri et al. | A new multi-level semi-supervised learning approach for network intrusion detection system based on the ‘goa’ | |
| CN108718310B (en) | Deep learning-based multilevel attack feature extraction and malicious behavior identification method | |
| Wang et al. | Delving into data: Effectively substitute training for black-box attack | |
| Hu et al. | Online adaboost-based parameterized methods for dynamic distributed network intrusion detection | |
| CN106375345B (en) | It is a kind of based on the Malware domain name detection method being periodically detected and system | |
| CN102479298B (en) | Program identification method and device based on machine learning | |
| CN101060443A (en) | An improved adaptive boosting algorithm based Internet intrusion detection method | |
| CN107846392A (en) | A kind of intrusion detection algorithm based on improvement coorinated training ADBN | |
| CN107370752B (en) | Efficient remote control Trojan detection method | |
| CN1889573A (en) | Active decoy method and system | |
| CN105208037A (en) | DoS/DDoS attack detecting and filtering method based on light-weight intrusion detection | |
| CN111126820B (en) | Method and system for preventing electricity stealing | |
| CN108322445A (en) | A kind of network inbreak detection method based on transfer learning and integrated study | |
| CN109299741A (en) | A kind of network attack kind identification method based on multilayer detection | |
| CN106973047A (en) | A kind of anomalous traffic detection method and device | |
| CN1808483A (en) | Biologic heuristic calculation based floating population information analysis and alert and alarm method | |
| Meddeb et al. | Anomaly-based behavioral detection in mobile Ad-Hoc networks | |
| CN113269228A (en) | Method, device and system for training graph network classification model and electronic equipment | |
| Bharathi et al. | A survey on crime data analysis of data mining using clustering techniques | |
| CN106447109A (en) | Community public security patrolling route optimization design method based on deep studying algorithm | |
| Feng et al. | Hrs: A hybrid framework for malware detection | |
| Matel et al. | Optimization of network intrusion detection system using genetic algorithm with improved feature selection technique | |
| CN106557983B (en) | Microblog junk user detection method based on fuzzy multi-class SVM | |
| JP2004312083A (en) | Learning data generating apparatus, intrusion detection system, and its program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20060726 |