CN109871488A - A kind of Web service construction method and Web service for merging availability and user preference - Google Patents
A kind of Web service construction method and Web service for merging availability and user preference Download PDFInfo
- Publication number
- CN109871488A CN109871488A CN201910132974.8A CN201910132974A CN109871488A CN 109871488 A CN109871488 A CN 109871488A CN 201910132974 A CN201910132974 A CN 201910132974A CN 109871488 A CN109871488 A CN 109871488A
- Authority
- CN
- China
- Prior art keywords
- services composition
- preference
- service
- attribute
- availability
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 12
- 239000000203 mixture Substances 0.000 claims abstract description 104
- 238000005457 optimization Methods 0.000 claims abstract description 18
- 239000002131 composite material Substances 0.000 claims abstract description 14
- 238000013528 artificial neural network Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000002068 genetic effect Effects 0.000 claims description 4
- 241000238633 Odonata Species 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims 4
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- NUFBIAUZAMHTSP-UHFFFAOYSA-N 3-(n-morpholino)-2-hydroxypropanesulfonic acid Chemical compound OS(=O)(=O)CC(O)CN1CCOCC1 NUFBIAUZAMHTSP-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000872 buffer Substances 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to Web service technology field more particularly to a kind of Web service construction method and Web services for merging availability and user preference.The following steps are included: the classification of step 1, user preference: the quality of service attribute of Web service being classified according to qualitative preference and quantitative preference according to the classification of user preference；Step 2 defines the calling structure of Services Composition according to the call relation between service and calculates the globally available degree of Services Composition, further according to globally available degree, obtains Services Composition trap queuing；Step 3, according to steps 1 and 2 three obtained Services Composition trap queuing, final Services Composition trap queuing is calculated using multi-objective optimization algorithm.The present invention applies qualitative and quantitative preference in composite services and adjusts initial attribute weight with the BP algorithm in neural network in qualitative preference measurement, user preference is enabled to obtain accurate expression.
Description
Technical field
The present invention relates to Web service technology field more particularly to a kind of Web service structures for merging availability and user preference
Construction method and Web service.
Background technique
Web service combination is to combine multiple single Web services according to specific rule and service logic, is obtained
Service with better function.The Web service both can be used as final service and be supplied to user, can also be used as new service publication
Onto network.Web service has function and non-functional (QoS) attribute, because different web services may have same function
Can, so researcher's more attention is its QoS attribute.In the application field of individualized selection and artificial intelligence, in order to prop up
Automatic decision is held, the extraction of user preference is very important, and either provides the user with service or user selects service, all
Need the help of preference.Meanwhile user preference can be divided into qualitative preference and quantitative preference, research shows that it is this classification be it is reasonable and
Effectively.Such as an itinerary, the purpose that user requests service are divided into four parts: (1) selecting flight；(2) predetermined wine
Shop；(3) mode of transportation；(4) means of payment.Different users has different preferences to same request or different requests, so
It is reasonable that service is formulated or selected according to user preference.Particularly, such as first part selects flight, for flight frequency
It can be described with qualitative fashion according to user preference, and flight prices can then be described with quantitative manner.Therefore exist
It is reasonable method that qualitative and quantitative mode is comprehensively considered in user preference.
There is many research about Services Composition problem, the mode for generally speaking constructing Services Composition process is main
There are two classes of centralization combination and distributed combination.The realization of distribution combination is mainly completed by Service Matching, former piece service
Output interface is matched with the input interface of consequent service, which driven by the meaning of one's words.Cheng J etc. is based on
Horn clause and Petri network handle linguistic fuzzy service, and input/output compatibility and the behavior restraint for solving service are compatible
Property, realize automatic service combination.The calculating such as Lecue can any two service interconnected input and the semantic phase of outlet chamber
Be divided into the matching dimensionality of service like angle value, and by obtained value, including it is equal, include, by comprising, intersect etc., taking
Established between business the meaning of one's words connection, by make task target and service object matching to achieve the purpose that Services Composition.Total
For, distribution combination is to want the function of realizing to user to establish abstract process, and process is made of multiple tasks, wraps in process
It is known that number of tasks, each task contained, which needs the logical relation between the function of realizing, task,.Energy is found according to functional requirement
It realizes the specific service of each task, completes Services Composition process, and tend to the Services Composition model split into service
Retrieval and two steps of services selection are respectively used to search the service for meeting user function demand and nonfunction requirement.And it concentrates
Formula combination is directed to established abstract process, is absorbed in and finds the candidate service for being able to achieve abstract service, and from candidate service
The middle service for finding to meet user's nonfunction requirement.It is excellent to propose a lightweight population by taking travel service as an example by Liao etc.
Change algorithm, multiple-objection optimization is carried out to QoS, to formulate the composite services for meeting user preference to user.The exploitation such as Abbassi
One kind automatically configuring algorithm, efficiently solves the np problem of Services Composition, avoid using heuristic (genetic algorithm,
Particle swarm algorithm etc.), improve the reliability and efficiency of Services Composition construction.Wang etc. proposes qualitative preference, quantitative preference
With the multiple target binding model of trust value, optimal service combination is selected using several multi-objective Algorithms.Zhang is proposed
WCP net, and solve the problems, such as that user preference is fine-grained, the weight that WCP is netted is as user preference is adjusted.
Above method perhaps considers user preference or considers network environment, but not by the two
Problem is simultaneously it is considered that lacking complete model does unified description to both of these problems.
Summary of the invention
In view of the above problems, the present invention considers comprehensive functional requirement for considering user and non-functional (QoS) demand simultaneously
User preference changes with the change of service environment, provides a kind of Web service construction method for merging availability and user preference,
Optimal service composite sequence is provided the user with by comprehensively considering multiple Services Composition targets.
To achieve the above object, The technical solution adopted by the invention is as follows:
A kind of Web service construction method merging availability and user preference, comprising the following steps:
The classification of step 1, user preference: according to the classification of user preference by the quality of service attribute of Web service according to fixed
Property preference and quantitative preference classify；Detailed process are as follows:
Step 1-1, the calculating and adjustment of qualitative preference: the qualitative preference of user is defined using WCP-nets model, by user
Qualitative preference is mapped to quantized value space, obtains the relative importance between qualitative attributes preferred different attribute value, re-defines
Qualitative attributes preferred initial weight is linearly calculated the initial value of violation degree, presses according to relative importance and initial weight
Initial value sorts to obtain the initial trap queuing of Services Composition under qualitative preference；Again using the attribute of BP neural network adjustment service
Weight further calculates violation degree and sorts, and the trap queuing of Services Composition is made to meet the qualitative preference of user；
Step 1-2, the calculating of quantitative preference: calculating the slackness of Services Composition by defining the slackness individually serviced,
Services Composition is sorted from small to large according to its slackness, obtains the Services Composition trap queuing under quantitative preference；
Step 2, the calling structure that Services Composition is defined according to the call relation between service simultaneously calculate the complete of Services Composition
Office's availability obtains Services Composition trap queuing further according to globally available degree；
Step 3, according to steps 1 and 2 three obtained Services Composition trap queuing, calculated most using multi-objective optimization algorithm
Whole Services Composition trap queuing.
The specific steps of step 1-1 adjustment attribute weight are as follows:
(1) formula V (sp)=∑ is appliedXwXVX(sp) to the violation degree and attribute weight of all properties of Services Composition sp
Weighted sum, calculates the violation degree initial value V (sp) of the Services Composition, and each Services Composition is arranged according to initial value size
Sequence, wherein VX(sp) the violation degree of the attribute X of Services Composition sp, w are representedXRepresent the weight of attribute X；
(2) according to the Services Composition historical record of user's actual selection, the Services Composition for actually meeting user preference is determined
The sequence of sequence gives each uniform assignment V'(sp of Services Composition sequence according to order of quality)=spi(0≤spi≤ 1) clothes, are determined
The practical sequence of business combination；
(3) normalized V (sp)=N (V (sp)) is defined, as N (V (sp)) ≠ V'(sp) when, using BP neural network tune
Whole attribute weight, and new Services Composition violation degree V (sp) is calculated, until meeting actual user's preference, i.e. V (sp)=V'
(sp)。
In step 1-2, it is assumed that the value V of attribute XiFor one group of ordinal value, i.e. vm<…<v1, wherein m is the number of attribute value
Amount, specifically calculates step are as follows:
(1) n (V is definedi) be attribute X i-th of value number, according to n (Vi) regulation slackness size, i.e. attribute X
Some value it is more, be considered as that its value is more welcome, then its value slackness is smaller；Define n (Vi) maximum value slackness be 1,
Successively slackness increases, and defines the slackness rd of each attributeXi；
(2) the slackness rd'(s individually serviced is defined) be equal to the sum of the slackness of each of which attribute, i.e.,
(3) calculating the slackness of Services Composition S is the sum of the slackness individually serviced:N is clothes
The quantity of service that business combination S includes.
In step 2, Services Composition calling structure is defined as sequence and calls structure and two kinds of parallel calling structure, it is single
The availability of Web service indicates with symbol A (s), the globally available degree symbol A of composite Web servicesgIt indicates, according to list
The globally available degree of the Calculation of Availability Services Composition of a Web service, and Services Composition is calculated based on service call structure
Availability；Specific steps are as follows:
It is called in structure in sequence, uses Ag=A (s1)·A(s2) globally available degree is calculated, wherein A (s1) it is previous clothes
Be engaged in s1Availability, A (s2) it is to be serviced s1The service s that sequence is called2Availability；In parallel calling structure, useGlobally available degree is calculated,
Wherein A (s2) and A (s3) it is by the service s of parallel calling2And s3Availability；w2And w3It is service s respectively2And s3Through step
(1-2) weight adjusted；The availability of Services Composition is calculated, formula is summarized are as follows:
The specific steps of step 3 are as follows:
(1) by steps 1 and 2 three obtained target, i.e., qualitative preference, quantitative preference, globally available degree carry out reasonable
Services Composition building；For qualitative preference, Services Composition collection sorts according to user preference；For quantitative preference, Services Composition collection
It sorts according to the slackness of Services Composition；For globally available degree, Services Composition collection is arranged according to the globally available degree of Services Composition
Sequence；
(2) it is sorted according to the respective Services Composition of three targets, using three targets as the input of multi-objective optimization algorithm
Data are calculated according to multi-objective optimization algorithm, obtain optimal service composite sequence, and wherein multi-objective optimization algorithm uses multiple target
Dragonfly algorithm and genetic algorithm.
A kind of Web service for merging availability and user preference, is constructed using the above method.
The present invention extracts after user files a request, according to the classification standard made in advance from Web service data library
It is qualitative attributes preferred and quantitative attributes preferred, and qualitative attributes preferred weight is adjusted, it calculates quantitative attributes preferred.The service of extraction can
Expenditure attribute goes out the globally available degree of Services Composition according to service call Structure Calculation.Finally using three modules as multiple target
The input of optimization module exports optimal service composite sequence.
Beneficial effects of the present invention:
(1) present invention applies qualitative and quantitative preference in composite services and has used nerve in qualitative preference measurement
BP algorithm in network adjusts initial attribute weight, and with the change of service environment, user preference also has corresponding change, this
When can dynamically adjust attribute weight, enable user preference to obtain accurate expression.
(2) present invention is by one of nonfunctional space of Web service-availability in view of being adjusted in Services Composition according to service
With structure, the globally available degree of Services Composition can be obtained by the availability of single Web service.
(3) present invention has redefined the calling structure of Services Composition, according to user in original service call structure
Preference is improved, and calls structure to be merged into the parallel calling based on weight original parallel calling structure and probability
Structure.
Detailed description of the invention
Fig. 1 is system architecture diagram of the invention；
Fig. 2 is overview flow chart of the invention；
Fig. 3 is the preference categories schematic diagram of user described in embodiment；
Fig. 4 is the WCP net schematic diagram of user described in embodiment；
Fig. 5 is that buffers call structure chart；
Fig. 6 is that Parallel Service calls structure chart.
Specific embodiment
Below by taking user Frank proposes travel service request to website as an example, the present invention is carried out specifically in conjunction with drawing
It is bright.
It is system architecture diagram of the invention shown in Fig. 1, is system flow chart of the invention shown in Fig. 2.
Firstly, classifying according to user preference to Web service attribute.
The attribute of Web service is generally divided into service quality (QoS) attribute and functional attributes in Web service set, service
Quality (QoS) attribute includes response time, handling capacity and availability etc..The method of the present invention is to be divided into the attribute of Web service to determine
Property preference required for attribute required for attribute and quantitative preference.Used partial service attribute such as Fig. 3 in the present embodiment
It is shown (since space limits, only to list part attribute and be used as reference.), to Service Properties according to qualitative preference and fixed in Fig. 3
Attribute required for amount preference is classified.In qualitative preference, the attribute used is in user's itinerary to tourist garment
The preferences of business, including four attributes: hotel, airline, mode of transportation and the means of payment.In quantitative preference, use
Attribute is the corresponding QoS attribute of travel service, comprising: response time, handling capacity, normalization, potentiality and price.
Secondly, defining the qualitative preference of user using WCP-nets (WCP net) model, and using BP neural network adjustment service
The attribute weight of qualitative preference.
The qualitative preference of user is mapped to quantized value space using WCP-nets model by this method, is obtained qualitative attributes preferred
Different attribute value between relative importance.Fig. 4 illustrates the WCP-nets of user Frank, and which depict Frank to travelling
The preference of service, including four attributes: A airline, the hotel B, C mode of transportation and the D means of payment.From the point of view of Frank,
Select three times more important than hotel of airline, five times more important than mode of transportation of hotel, mode of transportation more important than the means of payment two
Times, the weight of four attributes uses w respectivelya、wb、wc、wdIt indicates, by Fig. 4 (a) it can be concluded that the weight relationship of four attributes at this time
Are as follows: wa=3wb, wb=5wcAnd wc=2wd.There are two kinds of aircrafts of a1, a2 under same airline, as shown in Fig. 4 (c), excess-three
Attribute is also each, and there are two attribute values.In Fig. 4 (b), the relative importance of a1 and a2 is 2 in airline, for its excess-three
A attribute, using preference condition, such as under conditions of selecting a1, the relative importance of b1 and b2 are 1；Selection a2's
Under the conditions of, the relative importance of b2 and b1 are 3.For other two attributes, the important relationship between attribute value is referring to Fig. 4
(b)。
Under important relationship between above-mentioned attribute between attribute value, due to wa/wb=3, wb/wc=5 and wc/wd=2,
And total weight needs to meet wa+wb+wc+wd=1, w can be obtaineda=0.6977, wb=0.2326, wc=0.0465, wd=0.0233.
By the attribute value relative importance rank in Fig. 4 (b), the violation degree V of available each attributeX(sp) (i.e. Services Composition sp
Attribute X violation degree), example: selection attribute value a2, b1, c1, d2 composition Services Composition, the violation of its each attribute
Degree is VA(a2b1c1d2)=2, VB(a2b1c1d2)=3, VC(a2b1c1d2)=0, VD(a2b1c1d2)=2.Wherein VA
(a2b1c1d2)=2 referring to for Services Composition a2b1c1d2, the violation degree of this attribute of A airline is 2, because
In Fig. 4 (b), this selection of a2 is worse than a1, and poor degree is 2.The violation degree of tri- attributes of remaining B, C, D is also similarly to calculate
Method.
The violation degree initial value for calculating Services Composition sp, it is qualitative attributes preferred to each of individually servicing with linear method
Violation degree and attribute weight weighted sum.When the true selection of user and initial service integrated mode have deviation, above-mentioned side
Method cannot carry out accurate description to user preference.Therefore this method obtains the violation of Services Composition according to linear method first
Initial value is spent, nonlinear method is recycled to carry out weight adjustment, specific set-up procedure according to user preference are as follows:
(1) by formula V (sp)=∑XwXVX(sp) the violation degree initial value V (sp) for calculating the Services Composition, by each service
Combination is ranked up according to initial value size, wherein VX(sp) the violation degree of the attribute X of Services Composition sp, w are representedXRepresent attribute X
Weight.
(2) consider that user is likely to select to count under initial attribute weight and attribute violation degree in actual conditions
Obtained optimal service combination, then needs to be adjusted initial attribute weight at this time.According to the service of user's actual selection
Historical record is combined, determines the sequence for actually meeting the Services Composition sequence of user preference, gives each service according to order of quality
The uniform assignment V'(sp of composite sequence)=spi(0≤spi≤ 1) the practical sequence of Services Composition, is determined；
(3) normalized V (sp)=N (V (sp)) is defined, as N (V (sp)) ≠ V'(sp) when, using BP neural network tune
Whole attribute weight, and new Services Composition violation degree V (sp) is calculated, until meeting actual user's preference, i.e. V (sp)=V'
(sp)。
Utilize the concrete operations of BP neural network adjustment attribute weight are as follows:
(1) parameter: the number of iterations 50, learning rate 0.05 are set, and maximum frequency of training is 10000, and training precision is
0.00001；
(2) desired output of BP neural network is the practical violation degree V'(sp of the Services Composition of user's actual selection), BP
The input of neural network is the violation degree initial value V (sp) being calculated under initial attribute weight, and BP neural network is according to input
The attribute weight for meeting desired output can be calculated with desired output.
Again, quantitative preference slackness concept is introduced, quantitative preference is calculated.
Assuming that the value V of attribute XiFor one group of ordinal value, i.e. vm<…<v1, wherein m is the quantity of attribute value, specific to calculate
Step are as follows:
(1) n (V is definedi) be attribute X i-th of value number, according to n (Vi) regulation slackness size, i.e. attribute X
Some value it is more, be considered as that its value is more welcome, then its value slackness is smaller；Define n (Vi) maximum value slackness be 1,
Successively slackness increases, and defines the slackness rd of each attributeXi；
(2) the slackness rd'(s individually serviced is defined) be equal to the sum of the slackness of each of which attribute, i.e.,
(3) calculating the slackness of Services Composition S is the sum of the slackness individually serviced:N is clothes
The quantity of service that business combination S includes.
Again, it defines the calling structure of composite services and calculates globally available degree.
During Web service combination, structure is called according to specific Services Composition, to the availability of Services Composition into
Row Modeling Calculation.The availability of single Web service indicates with symbol A (s), the globally available degree symbol of composite Web services
AgIndicate, according to the globally available degree of the Calculation of Availability Services Composition of single Web service, and based on service call structure come
Calculate the availability of Services Composition；Specific steps are as follows:
It is called in structure in sequence, uses Ag=A (s1)·A(s2) globally available degree is calculated, wherein A (s1) it is previous clothes
Be engaged in s1Availability, A (s2) it is to be serviced s1The service s that sequence is called2Availability；In parallel calling structure, useGlobally available degree is calculated,
Wherein A (s2) and A (s3) it is by the service s of parallel calling2And s3Availability；w2And w3It is service s respectively2And s3Through step
(1-2) weight adjusted；The availability of Services Composition is calculated, formula is summarized are as follows:
Finally, choosing optimal service composite sequence using multi-objective optimization algorithm.
It is all closed by above 4 step preference qualitative to user and quantitative and globally available this three parts of degree of Services Composition
The Services Composition of reason models, for further determining that optimum combination service also needs reasonably all to consider to take by this three parts
In business combination.With reference to multiple-objection optimization, this is theoretical, using three parts as target, to generate final optimal service group
It closes, in order to provide user is given.
(1) by above-mentioned 2,3,4 step three obtained targets, i.e., qualitative preference, quantitative preference, globally available degree all carry out
Reasonable Services Composition building.Specifically, for qualitative preference, the Services Composition collection obtained according to user's actual preference is
The Services Composition sequence obtained according to attribute weight adjusted；For quantitative preference, Services Composition collection is according to Services Composition
Slackness sequence；For globally available degree, Services Composition collection sorts according to the globally available degree of Services Composition.
(2) it is sorted according to the respective Services Composition of three targets, using three targets as the input of multi-objective optimization algorithm
Data are calculated according to multi-objective optimization algorithm, finally obtain optimal service composite sequence.
The multi-objective optimization algorithm that the present invention uses has multiple target dragonfly algorithm (MODA) and genetic algorithm (NSGA2), this
Two kinds of algorithms have same effect of optimization for three objective optimizations of the invention.The major parameter of MOPSO and NSGA2 code is set
It sets as follows:
(1) MODA: population quantity is equal to Services Composition quantity, such as: 10,50,100 etc.；Archive size is equal to 100；
Destination number is 3；Dimension is 5.
(2) NSGA2: population quantity is equal to Services Composition quantity；The number of iterations this patent is taken as 100 times；Destination number is
3；Dimension is 30.
Claims (6)
1. a kind of Web service construction method for merging availability and user preference, it is characterised in that the following steps are included:
The classification of step 1, user preference: according to the classification of user preference by the quality of service attribute of Web service according to qualitative inclined
Good and quantitative preference is classified；Detailed process are as follows:
Step 1-1, the calculating and adjustment of qualitative preference: defining the qualitative preference of user using WCP-nets model, and user is qualitative
Preference is mapped to quantized value space, obtains the relative importance between qualitative attributes preferred different attribute value, re-defines qualitative
The initial value of violation degree is linearly calculated according to relative importance and initial weight in attributes preferred initial weight, by initial
Value sequence obtains the initial trap queuing of Services Composition under qualitative preference；Again using the Attribute Weight of BP neural network adjustment service
Weight, further calculates violation degree and sorts, the trap queuing of Services Composition is made to meet the qualitative preference of user；
Step 1-2, the calculating of quantitative preference: the slackness of Services Composition is calculated by defining the slackness individually serviced, will be taken
Business combination is sorted from small to large according to its slackness, obtains the Services Composition trap queuing under quantitative preference；
Step 2, the overall situation for defining the calling structure of Services Composition according to the call relation between service and calculating Services Composition can
Expenditure obtains Services Composition trap queuing further according to globally available degree；
Step 3, according to steps 1 and 2 three obtained Services Composition trap queuing, final clothes are calculated using multi-objective optimization algorithm
Business combination trap queuing.
2. the Web service construction method of fusion availability and user preference according to claim 1, it is characterised in that: step
The specific steps of rapid 1-1 adjustment attribute weight are as follows:
(1) formula V (sp)=∑ is appliedXwXVX(sp) the violation degree to all properties of Services Composition sp and attribute weight weighting
Summation, calculates the violation degree initial value V (sp) of the Services Composition, each Services Composition is ranked up according to initial value size,
Middle VX(sp) the violation degree of the attribute X of Services Composition sp, w are representedXRepresent the weight of attribute X；
(2) according to the Services Composition historical record of user's actual selection, the Services Composition sequence for actually meeting user preference is determined
Sequence, according to order of quality give each uniform assignment V'(sp of Services Composition sequence)=spi(0≤spi≤ 1) service group, is determined
The practical sequence closed；
(3) normalized V (sp)=N (V (sp)) is defined, as N (V (sp)) ≠ V'(sp) when, it is adjusted and is belonged to using BP neural network
Property weight, and new Services Composition violation degree V (sp) is calculated, until meeting actual user's preference, i.e. V (sp)=V'(sp).
3. the Web service construction method of fusion availability and user preference according to claim 1, it is characterised in that: step
In rapid 1-2, it is assumed that the value V of attribute XiFor one group of ordinal value, i.e. vm<…<v1, wherein m is the quantity of attribute value, specific to calculate
Step are as follows:
(1) n (V is definedi) be attribute X i-th of value number, according to n (Vi) regulation slackness size, i.e. certain of attribute X
A value is more, is considered as that its value is more welcome, then its value slackness is smaller；Define n (Vi) maximum value slackness be 1, successively
Slackness increases, and defines the slackness of each attribute
(2) the slackness rd'(s individually serviced is defined) be equal to the sum of the slackness of each of which attribute, i.e.,
(3) calculating the slackness of Services Composition S is the sum of the slackness individually serviced:N is service group
Close the quantity of service that S includes.
4. the Web service construction method of fusion availability and user preference according to claim 1, it is characterised in that: step
In rapid 2, Services Composition calling structure is defined as sequence and calls structure and two kinds of parallel calling structure, single Web service can
Expenditure indicates with symbol A (s), the globally available degree symbol A of composite Web servicesgIt indicates, according to single Web service
The globally available degree of Calculation of Availability Services Composition, and calculate based on service call structure the availability of Services Composition；Specifically
Step are as follows:
It is called in structure in sequence, uses Ag=A (s1)·A(s2) globally available degree is calculated, wherein A (s1) it is previous service s1's
Availability, A (s2) it is to be serviced s1The service s that sequence is called2Availability；In parallel calling structure, useGlobally available degree is calculated,
Wherein A (s2) and A (s3) it is by the service s of parallel calling2And s3Availability；w2And w3It is service s respectively2And s3Through step
(1-2) weight adjusted；The availability of Services Composition is calculated, formula is summarized are as follows:
5. the Web service construction method of fusion availability and user preference according to claim 1, it is characterised in that: step
Rapid 3 specific steps are as follows:
(1) by steps 1 and 2 three obtained target, i.e., qualitative preference, quantitative preference, globally available degree are reasonably serviced
Combination building；For qualitative preference, Services Composition collection sorts according to user preference；For quantitative preference, Services Composition collection according to
The slackness of Services Composition sorts；For globally available degree, Services Composition collection sorts according to the globally available degree of Services Composition；
(2) it is sorted according to the respective Services Composition of three targets, using three targets as the input data of multi-objective optimization algorithm,
It is calculated according to multi-objective optimization algorithm, obtains optimal service composite sequence, wherein multi-objective optimization algorithm uses multiple target dragonfly
Algorithm and genetic algorithm.
6. a kind of Web service for merging availability and user preference, it is characterised in that: using described in any one of claim 1-5
Method building.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910132974.8A CN109871488A (en) | 2019-02-22 | 2019-02-22 | A kind of Web service construction method and Web service for merging availability and user preference |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910132974.8A CN109871488A (en) | 2019-02-22 | 2019-02-22 | A kind of Web service construction method and Web service for merging availability and user preference |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109871488A true CN109871488A (en) | 2019-06-11 |
Family
ID=66919074
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910132974.8A Pending CN109871488A (en) | 2019-02-22 | 2019-02-22 | A kind of Web service construction method and Web service for merging availability and user preference |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871488A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324471A (en) * | 2020-01-22 | 2020-06-23 | 远景智能国际私人投资有限公司 | Service adjusting method, device, equipment and storage medium |
| CN113420779A (en) * | 2021-05-23 | 2021-09-21 | 济南浪潮数据技术有限公司 | Method, device, equipment and readable medium for web request anomaly detection |
| TWI811574B (en) * | 2019-12-26 | 2023-08-11 | 日商樂天集團股份有限公司 | Violation detection system, violation detection method and program product |
| CN117336352A (en) * | 2023-09-27 | 2024-01-02 | 苏州大学 | Qualitative and quantitative mixed cloud service quality assessment method and system |
| CN117354178A (en) * | 2023-09-27 | 2024-01-05 | 苏州大学 | Service flow optimization method and system based on qualitative and quantitative service attributes |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103646066A (en) * | 2013-12-03 | 2014-03-19 | 东南大学 | Method for selecting credible web services based on qualitative quantitative user preference |
| CN103763302A (en) * | 2013-12-16 | 2014-04-30 | 东南大学 | Web service combination generating method |
| CN107070704A (en) * | 2017-03-22 | 2017-08-18 | 东南大学 | A kind of Trusted Web services combined optimization method based on QoS |
-
2019
- 2019-02-22 CN CN201910132974.8A patent/CN109871488A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103646066A (en) * | 2013-12-03 | 2014-03-19 | 东南大学 | Method for selecting credible web services based on qualitative quantitative user preference |
| CN103763302A (en) * | 2013-12-16 | 2014-04-30 | 东南大学 | Web service combination generating method |
| CN107070704A (en) * | 2017-03-22 | 2017-08-18 | 东南大学 | A kind of Trusted Web services combined optimization method based on QoS |
Non-Patent Citations (2)
| Title |
|---|
| HONGBING WANG等人: "WCP-Nets: A Weighted Extension", 《SERVICE-ORIENTED COMPUTER》 * |
| 邹彬: "基于用户偏好和可信计算的Web服务组合研究", 《中国学位论文全文数据库》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI811574B (en) * | 2019-12-26 | 2023-08-11 | 日商樂天集團股份有限公司 | Violation detection system, violation detection method and program product |
| US11947643B2 (en) | 2019-12-26 | 2024-04-02 | Rakuten Group, Inc. | Fraud detection system, fraud detection method, and program |
| CN111324471A (en) * | 2020-01-22 | 2020-06-23 | 远景智能国际私人投资有限公司 | Service adjusting method, device, equipment and storage medium |
| CN113420779A (en) * | 2021-05-23 | 2021-09-21 | 济南浪潮数据技术有限公司 | Method, device, equipment and readable medium for web request anomaly detection |
| CN117336352A (en) * | 2023-09-27 | 2024-01-02 | 苏州大学 | Qualitative and quantitative mixed cloud service quality assessment method and system |
| CN117354178A (en) * | 2023-09-27 | 2024-01-05 | 苏州大学 | Service flow optimization method and system based on qualitative and quantitative service attributes |
| CN117354178B (en) * | 2023-09-27 | 2024-08-27 | 苏州大学 | Service flow optimization method and system based on qualitative and quantitative service attributes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871488A (en) | A kind of Web service construction method and Web service for merging availability and user preference | |
| Liu et al. | A speculative approach to spatial‐temporal efficiency with multi‐objective optimization in a heterogeneous cloud environment | |
| Mansouri et al. | Hybrid task scheduling strategy for cloud computing by modified particle swarm optimization and fuzzy theory | |
| She et al. | QoS-aware cloud service composition: A systematic mapping study from the perspective of computational intelligence | |
| CN103092683B (en) | For data analysis based on didactic scheduling | |
| CN109992404A (en) | PC cluster resource regulating method, device, equipment and medium | |
| CN110489229A (en) | A kind of multiple target method for scheduling task and system | |
| Zhao et al. | Toward SLA-constrained service composition: An approach based on a fuzzy linguistic preference model and an evolutionary algorithm | |
| CN111222916A (en) | Method and device for determining delivery area, model training method and storage medium | |
| Ben Alla et al. | A novel multiclass priority algorithm for task scheduling in cloud computing | |
| CN116501711A (en) | Computing power network task scheduling method based on 'memory computing separation' architecture | |
| Li et al. | Research on QoS service composition based on coevolutionary genetic algorithm | |
| US11741101B2 (en) | Estimating execution time for batch queries | |
| CN103475730B (en) | The web services system of selection that under a kind of cloud environment, user's degree of certainty guides | |
| Benjaafar et al. | Pricing in on-demand and one-way vehicle-sharing networks | |
| Gao et al. | An efficient evolutionary algorithm based on deep reinforcement learning for large-scale sparse multiobjective optimization | |
| CN114912030A (en) | Equity model training method, equity model recommendation method, electronic terminal and computer medium | |
| CN107066328A (en) | The construction method of large-scale data processing platform | |
| CN118211180A (en) | Information service method and platform for intelligent fusion of big data | |
| Xue et al. | Improved K-means algorithm in user behavior analysis | |
| CN107193940A (en) | Big data method for optimization analysis | |
| CN113901728A (en) | Computer second-class assembly line balance optimization method based on migration genetic algorithm | |
| Qiu et al. | An improved gray wolf optimization algorithm solving to functional optimization and engineering design problems | |
| CN111367661A (en) | Cloud task scheduling method, device, equipment and storage medium based on goblet sea squirt group | |
| CN113343715B (en) | Method, device and equipment for automatically generating regular expression and storage medium |
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: 20190611 |