CN107943818A - A kind of Urban Data service system and method based on Multi-source Information Fusion - Google Patents
A kind of Urban Data service system and method based on Multi-source Information Fusion Download PDFInfo
- Publication number
- CN107943818A CN107943818A CN201710933559.3A CN201710933559A CN107943818A CN 107943818 A CN107943818 A CN 107943818A CN 201710933559 A CN201710933559 A CN 201710933559A CN 107943818 A CN107943818 A CN 107943818A
- Authority
- CN
- China
- Prior art keywords
- information
- module
- mrow
- attribute
- fusion
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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
- G06F16/248—Presentation of query results
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
To realize the trans-sectoral business division data unified fusion of city magnanimity, the target for showing and managing, the present invention proposes a kind of Urban Data service system and Urban Data processing method based on Multi-source Information Fusion, urban multi-source information is pre-processed, realize information cleaning and conversion, multi-source information is tentatively merged by support vector machine classifier, further preliminary fusion results are optimized with reference to genetic algorithm, final fusion results are obtained, and think that user shows by graphical guide interface.The present invention realizes the fusion of urban multi-source information, can fast convergence, restructuring with associate the information of separate sources, so as to effectively serve in urban information management.
Description
Technical field
The present invention relates to a kind of Urban Data service system and data processing method based on Multi-source Information Fusion, belong to
The processing of big data process field, especially smart city correlation big data.
Background technology
With " concept of smart city 1.0 " is different, " the stronger adjusting data in smart city 2.0 " real-time, comprehensive and altogether
Enjoy, therefore " new smart city " usually with serve the people whole process is full-time, Urban Governance effective, data open it is co-melting common
Enjoy, economic development green is increased income, cyberspace clear and bright safely is main target.
And wherein, need to business division data progress trans-sectoral to city magnanimity to realize the target of " data open co-melting shared "
Unified fusion, show and manage, and generates city fuse information resources bank, realize the unified issue of information resources, subscribe on demand and
The functions such as exchange and interdynamic, office can be done towards each operation system in city, each committee, public's offer information resources service is shared
Key foundation tenability.Multi-sources Information Fusion Method provides the business service of information conversion and information fusion processing, fusion
Unified multi-source heterogeneous resource information, for cross-system, cross-cutting information exchange provide dynamically, expansible information format with it is interior
Hold transfer capability.
Information fusion method of the prior art, often lacks to the adaptation during cross-platform multi-source data actual fused
Property adjustment, such as CN200810059244.1 discloses an evidence theory urban transportation stream information based on fuzzy coarse central and melts
Conjunction method, merges the basic probability function of each group evidence based on improved D-S evidence theory composite formula, and produces
The resolution that the evidence of conflict conflicts, judges the confidence level of correlation detector, and relevant detection is selected according to confidence level height
Parameter melts the basic probability function of each group evidence using D-S evidence theory composite formula as fusion results, but at it
Lack robustness during conjunction, when real data is handled, when evidence clashes, the conclusion of mistake can be produced.
The content of the invention
To realize the trans-sectoral business division data unified fusion of city magnanimity, the target for showing and managing, the present invention proposes one
Urban Data service system of the kind based on Multi-source Information Fusion, system include:Information acquisition module, is used for realization to urban multi-source
Believe the collection suffered from;Pretreatment module, is used for realization the pretreatment to information, basis of formation database;Fusion Module, is used for realization
Preliminary fusion to data in basic database;Optimization module, for being optimized to preliminary fusion results, obtains final melt
Close result.
In addition, the invention also discloses a kind of Urban Data method of servicing based on Multi-source Information Fusion, including:Gather city
City's multi-source information;Multi-source information is pre-processed, basis of formation database;Data in basic database are tentatively melted
Close;Preliminary fusion results are optimized, obtain final fusion results;Final fusion results are stored, respond searching for user
Suo Zhiling, therefrom searches matched information and is shown by graphic interface.
Beneficial effects of the present invention are, during being pre-processed to urban multi-source information, realize information cleaning and
Conversion, tentatively merges multi-source information by support vector machine classifier, with reference to genetic algorithm further to preliminary fusion
As a result optimize, obtain final fusion results, and think that user shows by graphical guide interface.The present invention realizes city
The fusion of city's multi-source information, can fast convergence, restructuring with associate the information of separate sources, so as to effectively serve in urban information
Management.
Brief description of the drawings
Fig. 1 is Urban Data serving system architecture figure.
Fig. 2 is Urban Data service system work flow diagram.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right
The present invention is further elaborated, it will be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
It is shown in the drawings now with detailed reference to the embodiment of the present invention, the example of these embodiments.The suffix of element
" module " and " unit " is used herein to conveniently describe, and therefore can convertibly be used, and is distinguished without any
Meaning or function.
Although all elements or unit that form the embodiment of the present invention illustrated as being coupled in discrete component or are grasped
As discrete component or unit, but the present invention may be not necessarily limited to such a embodiment.According to embodiment, in the purpose of the present invention
One or more elements can be selectively bonded to element all in scope and are operating as one or more elements.
The embodiment of the present invention is described in detail below in conjunction with the accompanying drawings.
Fig. 1 is Urban Data serving system architecture figure proposed by the present invention, and system mainly includes information acquisition module, pre- place
Manage module, Fusion Module, four part of optimization module.
Wherein, information acquisition module, is used for realization the collection to urban multi-source information;Pretreatment module, is used for realization pair
The pretreatment of information, basis of formation database;Fusion Module, is used for realization the preliminary fusion to data in basic database;It is excellent
Change module, for being optimized to preliminary fusion results, obtain final fusion results.
Fig. 2 is Urban Data service system work flow diagram proposed by the present invention, and the data of information acquisition module collection are come
Source can include enterprise's basic information, public institution's basic information, public organization's information, legal person's tax information, legal person's labour protection etc.
One or more in urban multi-source information.
Information acquisition module communicates information to pretreatment module, pretreatment module point after urban multi-source information is obtained
It is other to n cities such as enterprise's basic information, public institution's basic information, public organization's information, legal person's tax information, legal person's labour protections
Information source is pre-processed, and attribute discretization is carried out to each information source.
Wherein, pretreatment module includes:Data cleansing module, for carrying out duplicate checking and cleaning to multi-source information;Feature carries
Modulus block, for carrying out feature extraction to multi-source information, the character subset of selection, which reflects, maximally depends on former multi-source information,
There is maximum mutual information with former multi-source information vector;Descretization module, is used for realization the attribute discretization of multi-source information.
Identical with mode of the prior art to the mode of information progress duplicate checking and cleaning, this will not be repeated here.
In one embodiment, descretization module realizes the attribute discretization of multi-source information in the following way:
With vectorial Ai(i=1,2 ... ..., n) represents the attribute of each information source, and A=A1∪A2∪…∪An, then it is vectorial
A is the conditional attribute of system;
The information content that each information source is included in system is mi(i=1,2 ... ..., n) group,It is then total
Basic database is m rows | A |+| B | row, wherein, B is output attribute vector, then before | A | be classified as it is defeated humanized, it is rear | B | be classified as defeated
Go out attribute.
Multi-source information after discretization can be stored in the basic database of system.Afterwards, Urban Data service system
Fusion Module the basic database of foundation is handled using the attribute reduction of fuzzy coarse central, eliminate redundant digit therein
According to establishing metadatabase, and be trained.
Wherein, Fusion Module includes:Redundancy cuts down module, the base for the attribute reduction by fuzzy coarse central to foundation
Plinth database is handled, and is eliminated redundant data therein, is established metadatabase;Training module, for according to metadatabase structure
Initial training sample is made, designs support vector machine classifier, initial training is carried out to metadatabase, realizes that preliminary information merges.
In one embodiment, redundancy abatement module is achieved by the steps of data processing:
1. appointing in conditional attribute collection A and taking finite element of some attribute as attribute reduction set R, R=is set to
{Aj(j=1,2 ... ..., n);
2. take an attribute to appointing in conditional attribute AImportance degree SGF (a, R, B) is calculated, ifMeet SGF (a ', R, B)=max (SGF (a, R, B) | a ∈ A), then a ' is required set, note;
3. to just selected attribute a ', itself and existing attribute of an element dependency degree k in R are calculated, i.e., Wherein, IND () represents equivalence relation, POSIND(a′)IND (b) represents IND
(b) the positive domains of IND (a ');
The attribute b (the corresponding k calculated is 0) with attribute a ' with maximum dependency degree is temporarily rejected from yojan set,
R ' is denoted as, recalculates the importance degree SGF (b, R ', B) of b, if being less than defined threshold value in advance with SGF (a, R, B) difference
δ, then do not reject attribute b;
4. calculate the dependence γ of R and BR(B),Wherein, card () represents collection
The radix of conjunction, if γR(B)=γA(B), then R be the condition that meets yojan attribute, otherwise return 2..
In one embodiment, training module is trained the sample in metadatabase using following steps:
1. determine that kernel function is Radial basis kernel function:
K (x, y) is expressed as the monotonic function of Euclidean distance between any point x to a certain center y in vector space, wherein y
For kernel function center, σ is the width parameter of function, controls the radial effect scope of function.
2. constructing support vector machines, it is S to set classification number, finds out all different classes of combination of two, shared S (S-1)/
2, respectively with the two classification samples into two class problem training sets, it is established that S (S-1)/2 support vector machines, is used
The support vector machines for stating the two class problems that solve tries to achieve G discriminant function respectively, counts S classification respectively in G discriminant function knot
Number of votes obtained in structure, the most classification of number of votes obtained are exactly final judgement classification;
3. building initial training sample set according to metadatabase, preliminary fusion is formed by the support vector machines of above-mentioned construction
As a result.
After obtaining preliminary fusion results, optimization module passes through fitness function and heredity according to preliminary information fusion results
Algorithm picks optimal subset;
Wherein, optimization module includes:
Optimal subset chooses module, and optimal subset is chosen from preliminary fusion results by optimization algorithm;
Optimal subset and metadatabase, are formed new training sample by retraining module, then be trained and circulate until
Untill meeting preset condition.
In one embodiment, optimal subset is chosen the optimization algorithm that module uses and is selected for fitness function and genetic algorithm
Optimal subset is taken, is comprised the following steps that:
1. determining fitness function, fusion accuracy is averaged as fitness function using training set data, i.e.,:
Wherein, T is the number of samples of training set, and θ is parameter vector to be learned, yiFor i-th of sample in training set,
For the weighted average of i-th of sample in training set;
2. carrying out least reduction to data, carrying out minimum using binary coded form about subtracts, wherein, each condition category
The discrete of property is counted out and σ is determined by its value range and required precision;
3. determining control parameter, the control parameter includes genetic algebra, Population Size, crossover probability, mutation probability, essence
Degree.
In one embodiment, it is as follows to perform step for retraining module:
1. optimal subset and metadatabase are formed into new training sample;
2. the fusion results of multi-source information are obtained by support vector machine classifier;
3. calculating each individual fitness and by ranking fitness;
4. genetic operator is acted on colony and produces the next generation, circulation performs, until each individual fitness is big
Untill preset value.
In one embodiment, system further includes:
Information fusion database, for storing final fusion results;Search module, searches for entrance, response for providing
The search instruction of user, searches matched information from information fusion database;Display module, for by search result figure
Change interface and return to user.
In addition, the invention also discloses a kind of Urban Data method of servicing based on Multi-source Information Fusion, including following step
Suddenly:
S101:Urban multi-source letter is gathered to suffer from;
S102:Multi-source information is pre-processed, basis of formation database;
S103:Data in basic database are tentatively merged;
S104:Preliminary fusion results are optimized, obtain final fusion results;
S105:Final fusion results are stored, respond the search instruction of user, matched information is therefrom searched and passes through figure
Shape showing interface.
Wherein, pretreatment is carried out to multi-source information to specifically include:
S1021:Duplicate checking and cleaning are carried out to multi-source information;
S1022:For carrying out feature extraction to multi-source information;
S1023:Attribute discretization is carried out to multi-source information.
In step S1023, carrying out attribute discretization step to multi-source information includes:
With vectorial Ai(i=1,2 ... ..., n) represents the attribute of each information source, and A=A1∪A2∪…∪An, then it is vectorial
A is the conditional attribute of system;
The information content that each information source is included in system is mi(i=1,2 ... ..., n) group,It is then total
Basic database is m rows | A |+| B | row, wherein, B is output attribute vector, then before | A | be classified as it is defeated humanized, it is rear | B | be classified as defeated
Go out attribute.
Data in basic database are carried out with preliminary fusion includes:
S1031:The basic database of foundation is handled by the attribute reduction of fuzzy coarse central, is eliminated therein superfluous
Remainder evidence, establishes metadatabase;
S1032:According to metadatabase construct initial training sample, design support vector machine classifier, to metadatabase into
Row initial training, realizes that preliminary information merges.
In one embodiment, step S1031 is specifically included:
1. appointing in conditional attribute collection A and taking finite element of some attribute as attribute reduction set R, R=is set to
{Aj(j=1,2 ... ..., n);
2. take an attribute to appointing in conditional attribute AImportance degree SGF (a, R, B) is calculated, ifMeet SGF (a ', R, B)=max (SGF (a, R, B) | a ∈ A), then a ' is required set, note R=R ∪ { a ' };
3. to just selected attribute a ', itself and existing attribute of an element dependency degree k in R are calculated, i.e., The attribute b with attribute a ' with maximum dependency degree, (the corresponding k that calculates is
0) temporarily rejected from yojan set, be denoted as R ', recalculate the importance degree SGF (b, R ', B) of b, if with SGF (a, R, B)
Difference is less than defined threshold value δ in advance, then does not reject attribute b;
4. calculate the dependence γ of R and BR(B),Wherein, card () represents collection
The radix of conjunction, if γR(B)=γA(B), then R be the condition that meets yojan attribute, otherwise return 2..
In one embodiment, step S1032 is specifically included:
1. determine that kernel function is Radial basis kernel function:
K (x, y) is expressed as the monotonic function of Euclidean distance between any point x to a certain center y in vector space, wherein y
For kernel function center, σ is the width parameter of function, controls the radial effect scope of function.
2. constructing support vector machines, it is S to set classification number, finds out all different classes of combination of two, shared S (S-1)/
2, respectively with the two classification samples into two class problem training sets, it is established that S (S-1)/2 support vector machines, is used
The support vector machines for stating the two class problems that solve tries to achieve G discriminant function respectively, counts S classification respectively in G discriminant function knot
Number of votes obtained in structure, the most classification of number of votes obtained are exactly final judgement classification;
3. building initial training sample set according to metadatabase, preliminary fusion is formed by the support vector machines of above-mentioned construction
As a result.
Preliminary fusion results are optimized including:
S1041:Optimal subset is chosen from preliminary fusion results by optimization algorithm;
S1042:Optimal subset and metadatabase are formed into new training sample, then is trained and circulates pre- until meeting
If untill condition.
In one embodiment, step S1041 includes:
1. determining fitness function, fusion accuracy is averaged as fitness function using training set data, i.e.,:
Wherein, T is the number of samples of training set, and θ is parameter vector to be learned, yiFor i-th of sample in training set,
For the weighted average of i-th of sample in training set;
2. carrying out least reduction to data, carrying out minimum using binary coded form about subtracts, wherein, each condition category
The discrete of property is counted out and σ is determined by its value range and required precision;
3. determining control parameter, the control parameter includes genetic algebra, Population Size, crossover probability, mutation probability, essence
Degree.
In one embodiment, step S1042 includes:
1. optimal subset and metadatabase are formed into new training sample;
2. the fusion results of multi-source information are obtained by support vector machine classifier;
3. calculating each individual fitness and by ranking fitness;
4. genetic operator is acted on colony and produces the next generation, circulation performs, until each individual fitness is big
Untill preset value.
When user scans for information, substance feature information is extracted from search condition, in information fusion database
In scan for searching matched information and returned by graphic interface.
The present invention solve enterprise's basic information, public institution's basic information, public organization's information, legal person's tax information,
The generation of data prediction, basic probability function in the urban multi-source information fusion process such as legal person's labour protection and combining evidences are public
Existing for formula etc. on multi-source information de-redundancy, basic probability function acquisition and the problems such as fitness function;
By the new method of rough set fuzzy theory, support vector machines and genetic algorithm, can be obtained in urban multi-source data fusion process
There must be the fusion results of larger confidence level.
While there has been shown and described that the specific embodiments of the embodiment of the present invention, but without departing substantially from the embodiment of the present invention
Exemplary embodiment and its broader aspect on the premise of, those skilled in the art obviously can be made based on teaching herein
Change and modifications.Therefore, appended claim is intended to all such exemplary embodiments without departing substantially from the embodiment of the present invention
True spirit and scope change and change be included in its within the scope of.
Claims (12)
1. a kind of Urban Data service system based on Multi-source Information Fusion, including:
Information acquisition module, is used for realization the collection to urban multi-source information;
Pretreatment module, is used for realization the pretreatment to information, basis of formation database;
Fusion Module, is used for realization the preliminary fusion to data in basic database;
Optimization module, for being optimized to preliminary fusion results, obtains final fusion results.
2. urban service system according to claim 1, it is characterised in that the system also includes:
Information fusion database, for storing final fusion results;
Search module, searches for entrance for providing, responds the search instruction of user, searched from information fusion database matched
Information;
Display module, for search result to be returned to user with graphic interface.
3. the urban service system according to claim 1-2, it is characterised in that the multi-source information includes:Enterprise basis
Information, public institution's basic information, public organization's information, legal person's tax information, legal person's labour protection information.
4. the urban service system according to claim 1-2, it is characterised in that the pretreatment module includes:
Data cleansing module, for carrying out duplicate checking and cleaning to multi-source information;
Characteristic extracting module, for carrying out feature extraction to multi-source information, the character subset of selection, which reflects, maximally to be depended on
Former multi-source information, has maximum mutual information with former multi-source information vector;
Descretization module, is used for realization the attribute discretization of multi-source information.
5. urban service system according to claim 4, it is characterised in that category of the descretization module to multi-source information
Property discretization step includes:
With vectorial Ai(i=1,2 ... ..., n) represents the attribute of each information source, and A=A1∪A2∪…∪An, then vector A is to be
The conditional attribute of system;
The information content that each information source is included in system is mi(i=1,2 ... ..., n) group,Then total basic number
Be m rows according to storehouse | A |+| B | row, wherein, B is output attribute vector, then before | A | be classified as it is defeated humanized, it is rear | B | be classified as output and belong to
Property.
6. the urban service system according to claim 1-2, it is characterised in that the Fusion Module includes:
Redundancy cuts down module, and the basic database of foundation is handled for the attribute reduction by fuzzy coarse central, eliminates
Redundant data therein, establishes metadatabase;
Training module, for constructing initial training sample according to metadatabase, designs support vector machine classifier, to metadatabase
Initial training is carried out, realizes that preliminary letter suffers from fusion.
7. urban service system according to claim 6, it is characterised in that the processing step bag of the redundancy abatement module
Include:
1. appointing in conditional attribute collection A and taking finite element of some attribute as attribute reduction set R, R={ A are set toj(j=
1,2 ... ..., n);
2. take an attribute to appointing in conditional attribute AImportance degree SGF (a, R, B) is calculated, ifMeet SGF (a ', R, B)=max (SGF (a, R, B) | a ∈ A), then a ' is required set, note R=R ∪ { a ' };
3. to just selected attribute a ', itself and existing attribute of an element dependency degree k in R are calculated, i.e., Wherein, IND () represents equivalence relation, POSIND(a′)IND (b) represents IND
(b) the positive domains of IND (a ');
The attribute b (the corresponding k calculated is 0) with attribute a ' with maximum dependency degree is temporarily rejected from yojan set, is denoted as
R ', recalculates the importance degree SGF (b, R ', B) of b, if being less than defined threshold value δ in advance with SGF (a, R, B) difference, then
Attribute b is not rejected;
4. calculate the dependence γ of R and BR(B),Wherein, card () represents set
Radix, if γR(B)=γA(B), then R be the condition that meets yojan attribute, otherwise return 2..
8. urban service system according to claim 6, it is characterised in that the processing step of the training module includes:
1. determine that kernel function is Radial basis kernel function:
<mrow>
<mi>K</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<mfrac>
<msup>
<mrow>
<mo>|</mo>
<mrow>
<mi>x</mi>
<mo>-</mo>
<mi>y</mi>
</mrow>
<mo>|</mo>
</mrow>
<mn>2</mn>
</msup>
<mrow>
<mn>2</mn>
<msup>
<mi>&sigma;</mi>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
K (x, y) is expressed as the monotonic function of Euclidean distance between any point x to a certain center y in vector space, and wherein y is core
Function center, σ are the width parameter of function, control the radial effect scope of function;
2. constructing support vector machines, it is S to set classification number, finds out all different classes of combination of two, shares S (S-1)/2,
Respectively with the two classification samples into two class problem training sets, it is established that S (S-1)/2 support vector machines, is asked with above-mentioned
The support vector machines for solving two class problems tries to achieve G discriminant function respectively, counts S classification respectively in G discriminant function structure
Number of votes obtained, the most classification of number of votes obtained is exactly final to judge classification;
3. building initial training sample set according to metadatabase, preliminary fusion knot is formed by the support vector machines of above-mentioned construction
Fruit.
9. the urban service system according to claim 1-2, it is characterised in that the optimization module includes:
Optimal subset chooses module, and optimal subset is chosen from preliminary fusion results by optimization algorithm;
Retraining module, new training sample is formed by optimal subset and metadatabase, then is trained and is circulated until meeting
Untill preset condition.
10. urban service system according to claim 9, it is characterised in that the optimal subset chooses what module used
Optimization algorithm chooses optimal subset for fitness function and genetic algorithm, comprises the following steps that:
1. determining fitness function, fusion accuracy is averaged as fitness function using training set data, i.e.,:
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>&theta;</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mi>T</mi>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>T</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>y</mi>
<mo>^</mo>
</mover>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
Wherein, T is the number of samples of training set, and θ is parameter vector to be learned, yiFor i-th of sample in training set,For instruction
Practice the weighted average for concentrating i-th of sample;
2. carrying out least reduction to data, carrying out minimum using binary coded form about subtracts, wherein, each conditional attribute
It is discrete to count out and σ is determined by its value range and required precision;
3. determining control parameter, the control parameter includes genetic algebra, Population Size, crossover probability, mutation probability, precision.
11. urban service system according to claim 10, it is characterised in that the retraining module performs step such as
Under:
1. optimal subset and metadatabase are formed into new training sample;
2. the fusion results of multi-source information are obtained by support vector machine classifier;
3. calculating each individual fitness and by ranking fitness;
4. genetic operator is acted on into colony and produces the next generation, circulation perform, until each individual fitness be all higher than it is pre-
If untill value.
12. a kind of Urban Data method of servicing based on Multi-source Information Fusion, including:
Gather urban multi-source information;
Multi-source information is pre-processed, basis of formation database;
Data in basic database are tentatively merged;
Preliminary fusion results are optimized, obtain final fusion results;
Final fusion results are stored, respond the search instruction of user, matched information is therefrom searched and passes through graphic interface
Displaying.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710933559.3A CN107943818A (en) | 2017-10-09 | 2017-10-09 | A kind of Urban Data service system and method based on Multi-source Information Fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710933559.3A CN107943818A (en) | 2017-10-09 | 2017-10-09 | A kind of Urban Data service system and method based on Multi-source Information Fusion |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107943818A true CN107943818A (en) | 2018-04-20 |
Family
ID=61936118
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710933559.3A Pending CN107943818A (en) | 2017-10-09 | 2017-10-09 | A kind of Urban Data service system and method based on Multi-source Information Fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107943818A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108597601A (en) * | 2018-04-20 | 2018-09-28 | 山东师范大学 | Diagnosis of chronic obstructive pulmonary disease auxiliary system based on support vector machines and method |
CN108596344A (en) * | 2018-04-17 | 2018-09-28 | 惠州学院 | A kind of complicated panel data learning method based on big data |
CN108958204A (en) * | 2018-08-15 | 2018-12-07 | 天津农学院 | A kind of edible fungus culturing investigating method based on expert system knowledge base |
CN110223161A (en) * | 2019-05-24 | 2019-09-10 | 东方银谷(北京)科技发展有限公司 | Credit estimation method and device based on feature dependency degree |
CN110458743A (en) * | 2019-08-12 | 2019-11-15 | 软通动力信息技术有限公司 | Community governance method, apparatus, equipment and storage medium based on big data analysis |
CN111274301A (en) * | 2020-01-20 | 2020-06-12 | 启迪数华科技有限公司 | Intelligent management method and system based on data assets |
CN112307026A (en) * | 2020-10-29 | 2021-02-02 | 广东海洋大学 | Method for establishing small ship navigation multi-source information real-time database |
CN112598340A (en) * | 2021-03-04 | 2021-04-02 | 成都飞机工业(集团)有限责任公司 | Data model comparison method based on uncertainty support vector machine |
CN114997344A (en) * | 2022-08-04 | 2022-09-02 | 中关村科学城城市大脑股份有限公司 | Multi-source data planning method and system based on urban brain |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101216998A (en) * | 2008-01-11 | 2008-07-09 | 浙江工业大学 | An information amalgamation method of evidence theory urban traffic flow based on fuzzy rough sets |
CN104121804A (en) * | 2014-07-23 | 2014-10-29 | 中北大学 | Self-loading system early failure predicting method based on multi-field information fusion |
CN106022473A (en) * | 2016-05-23 | 2016-10-12 | 大连理工大学 | Construction method for gene regulatory network by combining particle swarm optimization (PSO) with genetic algorithm |
CN106066892A (en) * | 2016-06-20 | 2016-11-02 | 四川上略互动网络技术有限公司 | A kind of travel information data processing method based on multisource data fusion |
-
2017
- 2017-10-09 CN CN201710933559.3A patent/CN107943818A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101216998A (en) * | 2008-01-11 | 2008-07-09 | 浙江工业大学 | An information amalgamation method of evidence theory urban traffic flow based on fuzzy rough sets |
CN104121804A (en) * | 2014-07-23 | 2014-10-29 | 中北大学 | Self-loading system early failure predicting method based on multi-field information fusion |
CN106022473A (en) * | 2016-05-23 | 2016-10-12 | 大连理工大学 | Construction method for gene regulatory network by combining particle swarm optimization (PSO) with genetic algorithm |
CN106066892A (en) * | 2016-06-20 | 2016-11-02 | 四川上略互动网络技术有限公司 | A kind of travel information data processing method based on multisource data fusion |
Non-Patent Citations (1)
Title |
---|
赵春玲 等: "一种多源信息的最优融合方法研究", 《信息与控制》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108596344A (en) * | 2018-04-17 | 2018-09-28 | 惠州学院 | A kind of complicated panel data learning method based on big data |
CN108596344B (en) * | 2018-04-17 | 2022-03-25 | 惠州学院 | Complex panel data learning method based on big data |
CN108597601B (en) * | 2018-04-20 | 2021-06-25 | 山东师范大学 | Support vector machine-based chronic obstructive pulmonary disease diagnosis auxiliary system and method |
CN108597601A (en) * | 2018-04-20 | 2018-09-28 | 山东师范大学 | Diagnosis of chronic obstructive pulmonary disease auxiliary system based on support vector machines and method |
CN108958204A (en) * | 2018-08-15 | 2018-12-07 | 天津农学院 | A kind of edible fungus culturing investigating method based on expert system knowledge base |
CN110223161A (en) * | 2019-05-24 | 2019-09-10 | 东方银谷(北京)科技发展有限公司 | Credit estimation method and device based on feature dependency degree |
CN110458743A (en) * | 2019-08-12 | 2019-11-15 | 软通动力信息技术有限公司 | Community governance method, apparatus, equipment and storage medium based on big data analysis |
CN111274301A (en) * | 2020-01-20 | 2020-06-12 | 启迪数华科技有限公司 | Intelligent management method and system based on data assets |
CN111274301B (en) * | 2020-01-20 | 2023-08-29 | 国云数字科技(重庆)有限公司 | Intelligent management method and system based on data assets |
CN112307026A (en) * | 2020-10-29 | 2021-02-02 | 广东海洋大学 | Method for establishing small ship navigation multi-source information real-time database |
CN112598340A (en) * | 2021-03-04 | 2021-04-02 | 成都飞机工业(集团)有限责任公司 | Data model comparison method based on uncertainty support vector machine |
CN112598340B (en) * | 2021-03-04 | 2021-06-22 | 成都飞机工业(集团)有限责任公司 | Data model comparison method based on uncertainty support vector machine |
CN114997344A (en) * | 2022-08-04 | 2022-09-02 | 中关村科学城城市大脑股份有限公司 | Multi-source data planning method and system based on urban brain |
CN114997344B (en) * | 2022-08-04 | 2022-10-25 | 中关村科学城城市大脑股份有限公司 | Multi-source data planning method and system based on urban brain |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107943818A (en) | A kind of Urban Data service system and method based on Multi-source Information Fusion | |
CN100565523C (en) | A kind of filtering sensitive web page method and system based on multiple Classifiers Combination | |
CN110162593A (en) | A kind of processing of search result, similarity model training method and device | |
CN107908606A (en) | Method and system based on different aforementioned sources automatic report generation | |
CN106067088A (en) | E-bank accesses detection method and the device of behavior | |
CN104331498A (en) | Method for automatically classifying webpage content visited by Internet users | |
CN101408883A (en) | Method for collecting network public feelings viewpoint | |
CN107341183A (en) | A kind of Website classification method based on darknet website comprehensive characteristics | |
CN103838857B (en) | Automatic service combination system and method based on semantics | |
CN102929891B (en) | The method and apparatus of process text | |
CN101256631B (en) | Method and apparatus for character recognition | |
CN111104511A (en) | Method and device for extracting hot topics and storage medium | |
CN111523035B (en) | Recommendation method, device, server and medium for APP browsing content | |
KR102601545B1 (en) | Geographic position point ranking method, ranking model training method and corresponding device | |
CN107977670A (en) | Accident classification stage division, the apparatus and system of decision tree and bayesian algorithm | |
Liu et al. | Learning to rank complex network node based on the self-supervised graph convolution model | |
CN110232133A (en) | A kind of image of clothing search method and system classified based on Fusion Features and style | |
CN111259167B (en) | User request risk identification method and device | |
CN109857952A (en) | A kind of search engine and method for quickly retrieving with classification display | |
CN106933954A (en) | Search engine optimization technology is realized based on Decision Tree Algorithm | |
CN117093686A (en) | Intelligent question-answer matching method, device, terminal and storage medium | |
CN115238774A (en) | Social network malicious account detection method and device based on heterogeneous graph neural network | |
Cao et al. | Fake reviewer group detection in online review systems | |
CN106909626A (en) | Improved Decision Tree Algorithm realizes search engine optimization technology | |
Yang et al. | Optimization of tourism information analysis system based on big data algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180420 |