CN106447041A - Group intelligent matching method based on self-adaptive genetic algorithm - Google Patents
Group intelligent matching method based on self-adaptive genetic algorithm Download PDFInfo
- Publication number
- CN106447041A CN106447041A CN201610894371.8A CN201610894371A CN106447041A CN 106447041 A CN106447041 A CN 106447041A CN 201610894371 A CN201610894371 A CN 201610894371A CN 106447041 A CN106447041 A CN 106447041A
- Authority
- CN
- China
- Prior art keywords
- group
- user
- chromosome
- days
- genetic algorithm
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The invention provides a group intelligent matching method based on a self-adaptive genetic algorithm. The group intelligent matching method comprises an initialization matching method when a group is established for the first time and a group secondary matching method when the group members change. The method is based on the self-adaptive genetic algorithm. The optimal grouping result is obtained by the steps of initialization of the population sequences, calculation of the adaptive value of the population, generation of the new population through selection, crossover and variation and termination condition judgment. The method is applied to grouping management of course learning software users, and the active user resources are reasonably and optimally allocated so that the active rate of the group and the user experience can be obviously enhanced.
Description
Technical field
The present invention relates to mobile internet technical field, it particularly relates to a kind of little based on self-adapted genetic algorithm
Group intelligent Matching method.
Background technology
Popularization with mobile terminal devices such as smart mobile phone, panel computers and the rise of mobile Internet, traditional
Course learning mode is gradually covered by public attention of course learning application program (Application, APP).User first passes through
Mobile terminal is downloaded course learning application program and is completed to register, and carries out phase by logging in this course learning application program afterwards
Close the study of course.
Existing course learning APP is many to carry out group match by course or user's registration time to user, however, due to
The course learning progress of each user and liveness are different, lead to the group having very active, some groups are not very active
Phenomenon.Be in be not enliven very much group Consumer's Experience poor, learning initiative is also significantly affected.
Accordingly, it would be desirable to improve current user grouping rule, seek a kind of group intelligent Matching method, using existing subscriber
Sample passes through to become further initialization coupling and the Optimum Matching bringing the group of completing, and active user resources are rationalized
Optimum allocation so that group enliven rate and Consumer's Experience is obviously improved.
Self-adapted genetic algorithm (Genetic Algorithm) is natural selection and the something lost of simulation Darwinian evolutionism
Pass the computation model of the biological evolution process learning mechanism, be a kind of method by simulating natural evolution process searches optimal solution.
Genetic algorithm is may a population (population) of potential disaggregation to start from the problem that represents, and population then by
Individuality (individual) composition of some encoding through gene (gene).Each individuality is actually chromosome
(chromosome) carry the entity of feature.As the main carriers of inhereditary material, i.e. the set of multiple genes, in it for chromosome
Portion's performance (i.e. genotype) is certain assortment of genes, and it determines the external presentation of the shape of individuality.Therefore, need at the beginning
The mapping realized from phenotype to genotype is coding work.Work due to copying gene code is very complicated, and we often enter
Row simplifies, such as binary coding, after just producing for population, according to the principle of the survival of the fittest and the survival of the fittest, by generation
(generation) develop and produce the approximate solution become better and better, in every generation, according to individual fitness in Problem Areas
(fitness) size selects (selection) individual, and the genetic operator (genetic by means of natural genetics
Operators) it is combined intersecting (crossover) and variation (mutation), produce the population representing new disaggregation.This
Individual process will lead to the same rear life of kind of images of a group of characters natural evolution to be adaptive to environment for population than former generation, in last reign of a dynasty population
Excellent individual process decodes (decoding), can be used as problem approximate optimal solution.
The basic operation process of genetic algorithm is as follows:
A) initialize:Setting evolutionary generation counter t=0, the maximum evolutionary generation T of setting, random M individual conduct of generation
Initial population P (0).
B) individual evaluation:Calculate each individual fitness in colony P (t).
, wherein represent integer set
C) Selecting operation:Selection opertor is acted on colony.The purpose selecting is that the individuality optimizing is genetic directly to down
A generation or intersected by pairing and produce new individuality and be genetic to the next generation again.Selection operation is built upon the adaptation of individual in population
On the basis of degree assessment.
D) crossing operation:Crossover operator is acted on colony.
E) mutation operator:Mutation operator is acted on colony, is on some locus to the individual string in colony
Genic value changes.
Colony P (t) obtains colony P (t+1) of future generation after selection, intersection, mutation operator.
F) end condition judges:If t=T, using in evolutionary process obtained have maximum adaptation degree individual as
Excellent solution output, terminates calculating.
Content of the invention
The purpose of the present invention is to carry out group to user in internet product based on liveness based on self-adapted genetic algorithm
Coupling classification, with reasonable distribution any active ues resource, obtains Optimum Matching result, lifts Consumer's Experience.
The present invention proposes a kind of group based on self-adapted genetic algorithm initialization matching process, comprises the steps:
Step 1):By user according to certain rule packet;
Step 2):Each user being grouped is classified according to user activity grade, each activity level classification
Include N number of user;
Step 3):By above-mentioned steps 2) in N number of user be divided into a chromosome according to every M user, by self adaptation
Genetic algorithm obtains optimum initialization matched packet result so that the user activity of each activity level classification keeps one
Cause.
Preferably, step 3) in optimum initialization matched packet result obtained by self-adapted genetic algorithm be specially:
Step 3.1):Initialization colony sequence;
N number of user in active classification is divided into a chromosome P (1,2 ... ..., M) according to every M, defines K=N/
M initial population, then the liveness of i-th user be, the liveness average of user, wherein, i, K, N, M are positive integer;
Step 3.2):Calculate the adaptive value of population;
The standard deviation of this chromosome=, wherein=, then adaptive value is, adaptive value is bigger, then more excellent, and adaptive value is less, then
Poorer;
Step 3.3):Select;
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, selects
The then selection that probability is more than random number enters filial generation, and select probability PiChange according to adaptive value carries out adaptive polo placement such as
Under:Pi=
Step 3.4):Intersect;
Select 2 chromosomes from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is less than handed over
Fork rate is just intersected, and the length then along chromosome randomly chooses a position, and all of position after this position is entered
Row exchanges.And crossover probability PcIt is calculated as follows:
Pc==, wherein
Step 3.5):Variation;
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than to become
Different probability just enters row variation, then just randomly chooses a position along the length of chromosome, and the stochastic transformation this position
Become another numeral.Mutation probability PmIt is calculated as below:
Pm=
Step 3.6):According to above-mentioned step 3.3) -3.5) form new population afterwards;
Step 3.7):Judge whether to terminate;
The liveness average of t-th subsequence g is,
==
As iterations j<When X time, then directly terminate iteration,
Otherwise iteration X time, wherein X are positive integer.
Preferably, step 1) described in necessarily rule be user's course application APP study schedule;Step 2) described in user
Liveness x is defined as:Log within full 7 days, the x=login of nearly 7 days described APP number of days/7, be discontented with login in 7 days, x=logs in institute
State APP number of days/registration number of days.
After completing group's initialization coupling according to the method described above, when in group, user changes, the present invention also carries
Go out a kind of group based on self-adapted genetic algorithm matching process again, comprised the steps:Step 4):According to upgrading rule
Then, filter out upgrading user;Step 5):According to collapsing rule, filter out degrading user;Step 6):Calculated based on Adaptive Genetic
The group of each grade is reorganized by method, forms Optimum Matching group.
Preferably, step 4) in upgrading rule be specially:The 1-7 days upgrading rules, accumulative number of days of checking card is 3, at once
Upgrading;The upgrading rule of more than 8 days, liveness reaches a high grade, at once upgrades;Check card within continuous 7 days, at once upgrade;Step
Collapsing rule in rapid 5) is specially:The 1-7 days collapsing rules, accumulative number of days of not checking card is 3, at once degrades;The fall of more than 8 days
Level rule, enlivening grade is lowermost level, at once degrades;Do not check card within continuous 7 days, at once degrade.
Preferably, step 6) in based on self-adapted genetic algorithm, the group of each grade is reorganized, form optimum
Join group to be specially:
Step 6.1):Initialization colony sequence;
Filter out Y described upgrading user and degrading user from the group being made up of M user after, remaining M-Y use
The group of family composition is referred to as incomplete group;By the user group of incomplete group entirely as chromosome a part, to incompleteness
User group and upgrading user carry out random number, and each Customs Assigned Number in wherein incomplete user group is the same, according to
Above-mentioned rule obtains chromosome (1,2 ... ... Y), and wherein M, Y are integer;
Step 6.2):Adaptive value calculates;
Calculate the adaptive value of each chromosome, the standard deviation of this chromosome=wherein=, then adaptive value is, wherein, adapts to
Value is bigger, then more excellent, adaptive value is less, then poorer;
Step 6.3):Selection operation;
Calculate the select probability P of each chromosome according to the adaptive value of each chromosomei, concrete formula:
Pi=
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, selects
The then selection that probability is more than random number enters filial generation;
Step 6.4):Crossover operation;
Select the sequence being available for intersecting, need the intersected group ensureing to have equal amount in two chromosomes just can carry out
Crossover operation, described crossover operation step is as follows:
Step 6.4.1):Randomly choose two Cross reaction bodies,
Step 6.4.2):Judge whether two chromosomes have intersection possible, if it is not, continuing to select, if it has, then
Enter next step.
Step 6.4.3):Adaptive value according to Cross reaction body calculates crossover probability
Pc==
Step 6.4.4):Generate the random number of (0,1), intersected if numerical value is less than crossover probability, otherwise
Continue to generate random number, until its numerical value is less than crossover probability;
Step 6.5):Mutation operation;
Adaptive value according to mutated chromosome calculates mutation probability:
Pm=
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than to become
Different rate just enters row variation, selects minimum individuality in series of variation to enter row variation, in the numeral equal with minimum individuality number with
Machine selects variation numeral;
Step 6.6):According to above-mentioned step 6.3) -6.5) form new population afterwards;
Step 6.7):Judge whether end condition;
The liveness average of t-th subsequence g is,
==
As iterations j<When X time, then directly terminate iteration,
Otherwise iteration X time, wherein X are positive integer.
Preferably, the user's number in each alive packets above-mentioned is 10 people, i.e. M=10;Iterations is 100 times, i.e. X
=100.
Brief description
Fig. 1 is the basic operation process flow diagram flow chart of self-adapted genetic algorithm.
Fig. 2 is a kind of group's initialization matching process flow chart based on self-adapted genetic algorithm of the present invention.
Fig. 3 is a kind of group based on self-adapted genetic algorithm optimum restructuring matching process flow chart of the present invention.
Specific embodiment
With reference to specific embodiment to the present invention based on group's intelligent Matching method of self-adapted genetic algorithm do into
One step describes in detail, but protection scope of the present invention is not limited to this.
Embodiment one:Set up group when first time, when initialization group coupling, using based on self-adapted genetic algorithm
Group initialization matching process.As shown in figure 1, the method is broadly divided into three big steps.
Step 1) user is classified according to curricular advancement;
Step 2) it is divided into A, B, C, D tetra- class according still further to user activity in the classification of each curricular advancement;
Step 3) user in each active classification is one group according to 10 people, ensures that the liveness of each group keeps as far as possible
Unanimously, optimum initialization matched packet result is obtained by self-adapted genetic algorithm.
Define user activity x:Log within full 7 days, x=login APP number of days/7 of nearly 7 days, be discontented with login in 7 days, x=
Log in APP number of days/registration number of days.
All users at present are carried out by the group's initialization matching process based on self-adapted genetic algorithm according to liveness
Join formation group, panel size's upper limit is 10 people it is ensured that the liveness average of group is minimum with the difference of user's average.Step 3) in
Obtain optimum initialization matched packet result based on self-adapted genetic algorithm to be specially:
Step 3.1):Initialization colony sequence;
N number of user of same course is divided into a chromosome P (1,2,3,4,5,6,7,8,9,10) according to every 10, is formed
K=N/10 initial population, then the liveness of i-th user be, the liveness average of user, wherein
Step 3.2):Calculate the adaptive value of population;
Calculate the adaptive value of each chromosome in initial population:
The standard deviation of this chromosome=, wherein=, then adaptive value is, adaptive value is bigger, then more excellent, and adaptive value is less, then
Poorer.
Step 3.3):Select;
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, selects
The then selection that probability is more than random number enters filial generation, and select probability PiChange according to adaptive value carries out adaptive polo placement such as
Under:Pi=
Step 3.4):Intersect;
Select 2 chromosomes from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is less than handed over
Fork rate is just intersected, and the length then along chromosome randomly chooses a position, and all of position after this position is entered
Row exchanges.And crossover probability PcIt is calculated as follows:
Pc==, wherein
Step 3.5):Variation;
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than to become
Different probability just enters row variation, then just randomly chooses a position along the length of chromosome, and the stochastic transformation this position
Become another numeral.Mutation probability PmIt is calculated as below
Pm=
Step 3.6):Generate new population;
According to the above-mentioned selection to population, intersect and make a variation and generate new population,
Step 3.7):Judge whether to terminate;
The liveness average of t-th subsequence g is,
==
As iterations j<When 100 times, then directly terminate iteration.
Otherwise iteration 100 times.
Embodiment two:After completing user's initialization coupling, daily need to upgrade user and degradation in internet product
User filters out, and then uses the group's matching process again based on self-adapted genetic algorithm.As shown in Fig. 2 the method is main
It is divided into three big steps.
Step 4):According to upgrading rule, filter out upgrading user
Step 5):According to collapsing rule, filter out degrading user
Step 6):Based on self-adapted genetic algorithm, the group of each grade is reorganized, form Optimum Matching
Group.
Wherein, collapsing rule is specially:The 1-7 days collapsing rules, accumulative number of days of not checking card is 3, at once degrades;8 days with
On collapsing rule, enliven grade be D level, at once degrade;Do not check card within continuous 7 days, at once degrade.
Upgrading rule is specially:The 1-7 days upgrading rules, accumulative number of days of checking card is 3, at once upgrades;The upgrading of more than 8 days
Rule, liveness reaches a high grade, at once upgrades;Check card within continuous 7 days, at once upgrade.
By promotedemote rule, the user of upgrade or downgrade will be needed in former group to filter out, then its original place is little
Group becomes incomplete group (less than 10 people) it is known that the customer group of incomplete small group of users composition, optimum little again through merging formation
Group configuration is so that the liveness standard deviation of optimum group and the difference of the liveness average of customer group are minimum.Step 6) based on adaptive
Answer genetic algorithm to reorganize the group of each grade, form Optimum Matching group and be specially:
Step 6.1):Initialization colony sequence
Before initialization colony, need clearly to initialize rule, specific as follows:
Former incomplete small group of users colony can not be broken, that is, this small group of users colony can only integrally be selected,
Intersect, mutation operation.
In order to meet rule, incomplete user group is carried out random number with upgrading user by us, wherein incomplete customer group
Each Customs Assigned Number in body is the same, and it is P (1,2,3,4,5) that there are 5 users in for example incomplete group, numbers rear 5 users
Numbering be the same such as P (1,1,1,1,1),
By said process, organize into groups formation individual chromosome at random by sequence number for one group according to 10 people, formed after the completion of marshalling
Initialization colony, and chromosome form is as follows
P (1,1,1,1,1,2,2,3,5,7)
Step 6.2):Adaptive value calculates
Calculate the adaptive value of each chromosome, the standard deviation of this chromosome=wherein=, then adaptive value is, wherein
Adaptive value is bigger, then more excellent, and adaptive value is less, then poorer.
Step 6.3):Selection operation;
Calculate the select probability P of each chromosome according to the adaptive value of each chromosomei, concrete formula:
Pi=, wherein N is user's number.
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, selects
The then selection that probability is more than random number enters filial generation.
Step 6.4):Crossover operation
Select the sequence being available for intersecting, need the intersected group ensureing to have equal amount in two chromosomes just can carry out
Crossover operation.Such as chromosome A (1,1,1,1,1,2,2,4,4,6) and chromosome B (3,3,3,3,3,3,3,3,5,5) this two
What chromosome was available for intersecting is (2,2) (4,4) and (5,5) are intersected, or (3,3,3,3,3,3,3,3) with (1,1,1,1,
1,2,2,6) intersected, and (3,3,3,3,3,3,3,3) and (1,1,1,1,1) are unequal due to number, then cannot complete to hand over
Fork operation, therefore crossover operation should following steps be carried out:
Step 6.4.1):Randomly choose two Cross reaction bodies,
Step 6.4.2):Judge whether two chromosomes have intersection possible, if it is not, continuing to select, if it has, then
Enter next step.
Step 6.4.3):Adaptive value according to Cross reaction body calculates crossover probability
Pc==
Step 6.4.4):Generate the random number of (0,1), intersected if numerical value is less than crossover probability, otherwise
Continue to generate random number, until its numerical value is less than crossover probability.
Crossover operation is carried out according to above-mentioned crossover rule.
Step 6.5):Mutation operation;
Adaptive value according to mutated chromosome calculates mutation probability
Pm=
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than to become
Different rate just enters row variation, selects minimum individuality in series of variation to enter row variation, in the numeral equal with minimum individuality number with
Machine selects variation numeral.
Step 6.6):Generate new population
Form new population according to after three above-mentioned steps.
Step 6.7):Judge whether end condition
The liveness average of t-th subsequence g is,
==
As iterations j<When 100 times, then directly terminate iteration,
Otherwise iteration 100 times.
Self-adapted genetic algorithm being applied to based on group's intelligent Matching method of self-adapted genetic algorithm in the present invention
Group initialization coupling and group restructuring coupling in, compared with prior art, the present invention's it is a technical advantage that:The present invention can
So that liveness is consistent in similar group, the group that effectively prevent is very active, and some groups are not very active
Phenomenon, active user resources have been carried out rationalization optimum allocation, after applying self-adapted genetic algorithm, group active
Rate and Consumer's Experience are obtained for and are obviously improved.
It is understood that the embodiment of above principle being intended to be merely illustrative of the present and the exemplary enforcement adopting
Mode, but the invention is not limited in this.For those skilled in the art, in the essence without departing from the present invention
In the case of god and essence, various modifications and improvement can be made, these modifications and improvement are also considered as protection scope of the present invention.
Claims (7)
1. a kind of based on self-adapted genetic algorithm group initialization matching process it is characterised in that:Comprise the steps:
Step 1):By user according to certain rule packet;
Step 2):Each user being grouped is classified according to user activity grade, is wrapped in each activity level classification
Include N number of user;
Step 3):By above-mentioned steps 2) in N number of user be divided into a chromosome according to every M user, by Adaptive Genetic
Algorithm obtains optimum initialization matched packet result so that the user activity of each activity level classification is consistent.
2. group's initialization matching process based on self-adapted genetic algorithm as claimed in claim 1 it is characterised in that:Step
3) obtain optimum initialization matched packet result by self-adapted genetic algorithm in be specially:
Step 3.1):Initialization colony sequence;
Every group of N number of user is divided into a chromosome P (1,2 ... ..., M) according to every M, defines K=N/M initial kind
Group, then the liveness of i-th user is, the liveness average of user, and wherein, i, K, N, M are positive integer;
Step 3.2):Calculate the adaptive value of population;
The standard deviation of this chromosome=, wherein=, then adaptive value is, adaptive value is bigger, then more excellent, and adaptive value is less, then poorer;
Step 3.3):Select;
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, select probability
Enter filial generation more than the then selection of random number, and select probability PiIt is as follows that change according to adaptive value carries out adaptive polo placement:Pi=
Step 3.4):Intersect;
Select 2 chromosomes from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is less than crossing-over rate
Just intersected, the length then along chromosome randomly chooses a position, and all of position after this position is carried out mutually
Change.And crossover probability PcIt is calculated as follows:
Pc==, wherein
Step 3.5):Variation;
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than to make a variation generally
Rate just enters row variation, then just randomly chooses a position along the length of chromosome, and the stochastic transformation of this position is become another
One numeral.Mutation probability PmIt is calculated as below:
Pm=
Step 3.6):According to above-mentioned step 3.3) -3.5) form new population afterwards;
Step 3.7):Judge whether to terminate;
The liveness average of t-th subsequence g is,
==
As iterations j<When X time, then directly terminate iteration,
Otherwise iteration X time, wherein X are positive integer.
3. group's initialization matching process based on self-adapted genetic algorithm as claimed in claim 2 it is characterised in that:Step
1) described in, necessarily rule is user's course application APP study schedule;Step 2) described in user activity x be defined as:Full 7 days
Log in, the x=login of nearly 7 days described APP number of days/7, be discontented with login in 7 days, x=logs in described APP number of days/registration number of days.
4. a kind of group's initialization matching process based on self-adapted genetic algorithm as described in one of claim 1-3, it is special
Levy and be:After completing group's initialization coupling, when in group, user changes, need group is mated again;
Comprise the steps:
Step 4):According to upgrading rule, filter out upgrading user;
Step 5):According to collapsing rule, filter out degrading user;
Step 6):Based on self-adapted genetic algorithm, the group of each grade is reorganized, form Optimum Matching group.
5. as claimed in claim 4 a kind of group based on self-adapted genetic algorithm again matching process it is characterised in that:Institute
State step 4) in described upgrading rule be specially:The 1-7 days upgrading rules, accumulative number of days of checking card is 3, at once upgrades;8 days with
On upgrading rule, liveness reaches a high grade, at once upgrades;Check card within continuous 7 days, at once upgrade;
Described step 5) in described collapsing rule be specially:The 1-7 days collapsing rules, accumulative number of days of not checking card is 3, at once drops
Level;The collapsing rule of more than 8 days, enlivening grade is lowermost level, at once degrades;Do not check card within continuous 7 days, at once degrade.
6. a kind of group based on the self-adapted genetic algorithm matching process again as described in claim 4 or 5, its feature exists
In:Step 6) described in based on self-adapted genetic algorithm, the group of each grade is reorganized, form Optimum Matching group tool
Body is:
Step 6.1):Initialization colony sequence;
Filter out Y described upgrading user and degrading user from the group being made up of M user after, remaining M-Y user's group
The group becoming is referred to as incomplete group;By the user group of incomplete group entirely as chromosome a part, to incomplete user
Colony and upgrading user carry out random number, and each Customs Assigned Number in wherein incomplete user group is the same, according to above-mentioned
Rule obtains chromosome (1,2 ... ... Y), and wherein M, Y are integer;
Step 6.2):Adaptive value calculates;
Calculate the adaptive value of each chromosome, the standard deviation of this chromosome=wherein=, then adaptive value, wherein, adaptive value is bigger,
Then more excellent, adaptive value is less, then poorer;
Step 6.3):Selection operation;
Calculate the select probability P of each chromosome according to the adaptive value of each chromosomei, concrete formula:
Pi=
Selection course is the mode according to roulette, and the random number generating 0-1 at random is compared with select probability, select probability
Enter filial generation more than the then selection of random number;
Step 6.4):Crossover operation;
Select the sequence being available for intersecting, need to ensure that the intersected group having equal amount in two chromosomes just can be intersected
Operation, described crossover operation step is as follows:
Step 6.4.1):Randomly choose two Cross reaction bodies,
Step 6.4.2):Judge whether two chromosomes have intersection possible, if it is not, continuing to select, if it has, then entering
Next step.
Step 6.4.3):Adaptive value according to Cross reaction body calculates crossover probability:
Pc==
Step 6.4.4):Generate the random number of (0,1), intersected if numerical value is less than crossover probability, otherwise continue
Generate random number, until its numerical value is less than crossover probability;
Step 6.5):Mutation operation;
Adaptive value according to mutated chromosome calculates mutation probability:
Pm=
Select 1 chromosome from colony, generate its value random number between 0 to 1 simultaneously, if numerical value is higher than aberration rate
Just enter row variation, select minimum individuality in series of variation to enter row variation, select at random in the numeral equal with minimum individuality number
Select variation numeral;
Step 6.6):According to above-mentioned step 6.3) -6.5) form new population afterwards;
Step 6.7):Judge whether end condition;
The liveness average of t-th subsequence g is,
==
As iterations j<When X time, then directly terminate iteration,
Otherwise iteration X time, wherein X are positive integer.
7. the group's initialization matching process based on self-adapted genetic algorithm as described in one of claim 1-6, its feature exists
In:M=10, X=100.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610894371.8A CN106447041A (en) | 2016-10-13 | 2016-10-13 | Group intelligent matching method based on self-adaptive genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610894371.8A CN106447041A (en) | 2016-10-13 | 2016-10-13 | Group intelligent matching method based on self-adaptive genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106447041A true CN106447041A (en) | 2017-02-22 |
Family
ID=58174518
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610894371.8A Withdrawn CN106447041A (en) | 2016-10-13 | 2016-10-13 | Group intelligent matching method based on self-adaptive genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106447041A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021129530A1 (en) * | 2019-12-26 | 2021-07-01 | 百果园技术(新加坡)有限公司 | Virtual item display method and apparatus, computer device and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102184345A (en) * | 2011-06-27 | 2011-09-14 | 山东地纬计算机软件有限公司 | Test-paper generation method based on genetic algorithm |
US8069131B1 (en) * | 2004-05-21 | 2011-11-29 | Ray Anthony Luechtefeld | Method, artificially intelligent system and networked complex for facilitating group interactions |
CN103379020A (en) * | 2012-04-24 | 2013-10-30 | 苏州引角信息科技有限公司 | Method and system for massively sending emails |
CN103905246A (en) * | 2014-03-06 | 2014-07-02 | 西安电子科技大学 | Link prediction method based on grouping genetic algorithm |
CN105205130A (en) * | 2015-09-15 | 2015-12-30 | 广东工业大学 | Method of improving accuracy of recommendation system |
-
2016
- 2016-10-13 CN CN201610894371.8A patent/CN106447041A/en not_active Withdrawn
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8069131B1 (en) * | 2004-05-21 | 2011-11-29 | Ray Anthony Luechtefeld | Method, artificially intelligent system and networked complex for facilitating group interactions |
CN102184345A (en) * | 2011-06-27 | 2011-09-14 | 山东地纬计算机软件有限公司 | Test-paper generation method based on genetic algorithm |
CN103379020A (en) * | 2012-04-24 | 2013-10-30 | 苏州引角信息科技有限公司 | Method and system for massively sending emails |
CN103905246A (en) * | 2014-03-06 | 2014-07-02 | 西安电子科技大学 | Link prediction method based on grouping genetic algorithm |
CN105205130A (en) * | 2015-09-15 | 2015-12-30 | 广东工业大学 | Method of improving accuracy of recommendation system |
Non-Patent Citations (3)
Title |
---|
DR. DEVASENATHIPATHI N MUDALIAR ET AL.: "Contemplating Crossover Operators of Genetic Algorithm for Student Group Formation Problem", 《IJETAE EXPLORING RESEARCH AND INNOVATIONS》 * |
张春玲: "基于遗传算法的协作学习自动分组", 《华南农业大学信息学院》 * |
李峰 等: "基于遗传算法的分组算法", 《长春工业大学学报(自然科学版)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021129530A1 (en) * | 2019-12-26 | 2021-07-01 | 百果园技术(新加坡)有限公司 | Virtual item display method and apparatus, computer device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Fang et al. | A hybrid of real coded genetic algorithm and artificial fish swarm algorithm for short-term optimal hydrothermal scheduling | |
Linder et al. | Reconstructing patterns of reticulate evolution in plants | |
Eiben et al. | What is an evolutionary algorithm? | |
Man et al. | Genetic algorithms: concepts and designs | |
CN107122843A (en) | A kind of traveling salesman problem method for solving based on improved adaptive GA-IAGA | |
CN101118611A (en) | Business process model resource configuring optimizing method based on inheritance algorithm | |
Hutt et al. | Synapsing variable-length crossover: Meaningful crossover for variable-length genomes | |
CN115543556A (en) | Adaptive symbolic regression method based on multitask genetic programming algorithm | |
Cury et al. | Simulation of bacterial populations with SLiM | |
CN106447041A (en) | Group intelligent matching method based on self-adaptive genetic algorithm | |
CN107257307A (en) | A kind of parallelization genetic algorithm for solving multiple terminals collaboration network access method based on Spark | |
Alba et al. | Evolutionary algorithms | |
CN111352650A (en) | Software modularization multi-objective optimization method and system based on INSGA-II | |
CN116050685A (en) | Path planning method based on improved multi-factor optimization | |
CN110533186B (en) | Method, device, equipment and readable storage medium for evaluating crowdsourcing pricing system | |
Gwak et al. | Novel dynamic diversity controlling EAs for coevolving optimal negotiation strategies | |
CN103974388B (en) | A kind of wireless sensor network data fusion method and device | |
CN111859807A (en) | Initial pressure optimizing method, device, equipment and storage medium for steam turbine | |
CN107301040B (en) | Software product line product derivation method based on subtree decomposition | |
CN106951726A (en) | A kind of optimization method and device of gene coded sequence | |
McAvoy | Stochastic selection processes | |
Ünal et al. | Genetic algorithm | |
CN113742251B (en) | Software test path generation method and system based on set evolution | |
Abraham et al. | Shortest path computation in large graphs using bidirectional strategy and genetic algorithms | |
CN105978816A (en) | Multicast tree optimization method based on genetic framework |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170222 |
|
WW01 | Invention patent application withdrawn after publication |