CN102331929A - Service classification and recommendation method based on service combination history - Google Patents
Service classification and recommendation method based on service combination history Download PDFInfo
- Publication number
- CN102331929A CN102331929A CN201110175055A CN201110175055A CN102331929A CN 102331929 A CN102331929 A CN 102331929A CN 201110175055 A CN201110175055 A CN 201110175055A CN 201110175055 A CN201110175055 A CN 201110175055A CN 102331929 A CN102331929 A CN 102331929A
- Authority
- CN
- China
- Prior art keywords
- service
- sos
- node
- services
- classification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of services computing, relates to a service classification and recommendation method based on a bipartite graph and the projection of the bipartite graph. The method comprises a service classification step and a service recommendation step; the service classification step comprises the specific steps: collecting service-oriented software (SOS) information, abstracting a construction relation between the SOS and the service by using the bipartite graph; then projecting the bipartite graph at a SOS dimension to obtain SOS similar graphs, and clustering the SOS similar graphs to obtain the classes of the SOS; the service recommendation step comprises the following specific steps: projecting the bipartite graph at the service dimension to obtain a service combination graph, and analyzing the use mode of the SOS and the service according to the bipartite graphs of the SOS and the service, taking a plurality of services with high operating frequency as the core services, recommending the corresponding service according to different application scenes. The method has the advantages of realizing the automatic classification of the service and adding the classification information for the service, thereby providing convenience for discovering and searching the service and managing the service resources.
Description
Technical field
The invention belongs to the service compute technical field, particularly relate to a kind of classification of service and recommend method based on bigraph (bipartite graph) and projection thereof.
Background technology
Service-oriented calculating (Service-Oriented Computing; SOC) be the novel computation schema that is directed against distributed system; Organize quick application integration problem to bring many facilities for solving to stride in the distribution isomerous environment, become one of the most popular topic of software field.SOC advocates with service and is combined as the development mode that base configuration is used; Cause software system in main shape, the mode of production, the method for operation and use-pattern that great variety has all taken place, software systems are in the collaborative software environment of opening of being made up of the software service entity.Can imagine that following software systems can be formed by the various service dynamic combined that distribute on the network, we will live in " no net not, do not have to serve do not exist " networked Software World in.Software development based on service will become the main mode that industry develops software and adopts.
Along with going deep into of software architecture research, people recognize that gradually software configuration is the key factor of decision software quality.Some researchers are abstracted into network structure (software network) with software systems; Be about to software element (data object, method, module, class, member, subsystem etc.) and be regarded as node; Relation between the element is as undirected (oriented) limit that connects each node, and the research method of taking soft project and complex network correlation theory to combine is the starting point with the analysis software architectural feature; To understand the architectural characteristic of software; Be target for software development and maintenance provides support, come to have obtained some achievements analyzing towards the project software static structure from characterizing, analyze, measure and use four aspects; (Object-Oriented, OO) Development of Software has certain directive significance to these achievements in research to instructing object-oriented.Nowadays (Service-Oriented Software SOS) has become main flow to service-oriented software day by day, and we are necessary to study the structure of SOS, excavate the knowledge that contains in its structure, instruct the exploitation of SOS better.
Service discovery with the service combination attracts lot of domestic and foreign scholar's concern, become the key issue in the SOC field.But present service lacks classified information, makes troubles for the management of service discovery, retrieval and Service Source.If adopt manual mode to classification of service, obviously be worthless when quantity of service is big, how to realize that therefore the automatic classification of serving constitutes the challenge of SOC.Meanwhile, the relation capable of being combined of serving in the Services Combination judges according to the input and output between service that often this principle is too strict.Except the input and output coupling, whether there is a kind of new service matching method so, excavates the relation potential capable of being combined between service.
Summary of the invention
The present invention solves the present service of existing in prior technology to lack classified information, the technical matters of making troubles etc. for the management of service discovery, retrieval and Service Source; A kind of automatic classification of serving of can realizing is provided, and is that classified information is added in service, thereby gives a kind of based on Services Combination historical classification of service and recommend method that the management of service discovery, retrieval and Service Source facilitates.
It is to solve the relation capable of being combined of serving in the existing in prior technology Services Combination to judge according to the input and output between service often that the present invention also has a purpose, and this principle is the technical matters of strictness etc. too; A kind of historical development Experience of reusing is provided,, thereby has provided support for service-oriented software development for the software developer provides service capable of being combined; Simultaneously, utilize the historical information of Services Combination (SOS), start with from structure angle (bigraph (bipartite graph) and projection thereof); The knowledge that contains in the mining structure; Final realize classification of service and and service recommendation, the information that method is utilized is few, and simple and easy to operate a kind of based on Services Combination historical classification of service and recommend method.
Above-mentioned technical matters of the present invention mainly is able to solve through following technical proposals:
A kind of based on Services Combination historical classification of service and recommend method, it is characterized in that comprise classification of service step and service recommendation step, concrete grammar is following:
Described classification of service step comprises:
Step 1.1; Collect the metamessage data of SOS; Comprise: the label of the set of service that the title of SOS, SOS descriptor, SOS use, sign SOS, the time that SOS adds, the developer of SOS, the address of SOS; And the metamessage data that will collect SOS carry out error in data and handle, and the data after these are handled are stored in local data base;
Step 1.2, the data that analyzing step 1.1 obtains are analyzed the constituent relation between SOS and the service, and this constituent relation are represented with bigraph (bipartite graph) said bigraph (bipartite graph) comprises two category nodes, comprise the node of representing SOS and the node of representing service;
Step 1.3 is carried out projection with the bigraph (bipartite graph) that obtains in the step 1.2 in the SOS dimension, obtains corresponding one dimension perspective view---SOS similar diagram;
Step 1.4 utilizes clustering algorithm or community discovery algorithm that the node in the SOS similar diagram in the step 1.3 is carried out cluster, and the SOS node division is become some classes and check to be all kinds of specified class another names in each type behind each SOS;
Step 1.5 on the basis of the SOS classification results that step 1.4 obtains, according to bigraph (bipartite graph) between SOS in the step 1.2 and service, is calculated the SOS quantity that each service is participated in all kinds of, then it is assigned in that maximum type of SOS quantity, accomplishes classification of service;
Described service recommendation step comprises as follows:
Step 2.1 is carried out projection with the bigraph (bipartite graph) that obtains in the step 1.2 in the service dimension, and obtaining corresponding one dimension perspective view is Services Combination figure, the node representative service of said Services Combination figure; Bian represents the relation capable of being combined between the service that this edge two ends node representes; Weights on the limit are represented the intensity of this relation capable of being combined, and these weights can be measured through the cooccurrence relation of service in SOS of this edge two end nodes representative;
Step 2.2; Analyze the use pattern between SOS and service according to the bigraph (bipartite graph) of SOS that obtains in the step 1.2 and service, promptly SOS uses several services, and service is used by several SOS; And analyze these digital distributions, thereby confirm most of SOS use one of the service number more among a small circle [
A,
B], confirm scope [
A,
B] method be: if having
NIndividual SOS is with this
NIndividual SOS sorts by the service number of its use from small to large, gets to come the front
Individual SOS, wherein,
For more than or equal to
Percentage value less than 100% is established and is come
The service number that the SOS of position uses does
Q, so just be provided with [
A,
B] be [1,
Q];
Step 2.3, with frequency of utilization high before
BIndividual service is as kernel service;
Step 2.4, select to carry out following steps:
Step 2.41 if the user does not also select any service, then is the kernel service in its recommendation step 2.3;
Step 2.42 if the user has selected a service, be that its recommendation links to each other with this service according to Services Combination figure in the step 2.1 then, and weights comes top-between them
kService (
kBe the parameter of user's input, if
kGreater than the degree of this service then
kBe the degree of this service, otherwise
kConstant.
kCan get arbitrary integer greater than 0.But it is general
kBe worth greatly more, proposed algorithm required time in recommendation service is long more;
Step 2.43, if the user has selected 2 (comprising 2) above service: all that at first ask these services make up in twos; To each combination, in Services Combination figure, asking with two services in this combination is terminal then, and comprises all paths of other service (service two services in this combination in the selected service); The limit power of asking every paths at last respectively and, and the power of limit by path and to all path descending sorts, with top-
kThe user is recommended in the path, and (what generally recommend all is that the node number exists on the path
BWith interior path).
At above-mentioned a kind of classification of service and recommend method based on Services Combination history; In the said step 1.2, all there is the limit in the node of the said SOS of representative with the node of representative service, the constituent relation between expression SOS and the service; If promptly a SOS is made up of 3 services; There is the limit in the node of then representing this SOS respectively and between 3 nodes of representative service, but does not have the limit between similar node, promptly represents between the node of SOS and between the node of representative service not have the limit.
At above-mentioned a kind of classification of service and recommend method based on Services Combination history, in the said step 1.3, the node of SOS similar diagram is represented SOS software; The similarity relation is represented to exist between the SOS of its two ends in the limit of network, promptly exists similarity that the limit is then arranged between SOS, has no similarity then boundless; Weights on the limit are represented the similarity between these two SOS, and these weights can be measured through metamessage between these two SOS, comprise the similarity of descriptor, shared service, shared label, common developer.
Therefore, the present invention has following advantage: 1. can realize the automatic classification of serving, be that classified information is added in service, thereby facilitate for the management of service discovery, retrieval and Service Source; 2. reuse historical development Experience,, thereby provide support for service-oriented software development for the software developer provides service capable of being combined; Simultaneously, utilize the historical information of Services Combination (SOS), start with from structure angle (bigraph (bipartite graph) and projection thereof); The knowledge that contains in the mining structure; Final realize classification of service and and service recommendation, the information that method is utilized is few, and simple and easy to operate.
Description of drawings
Fig. 1 mashup and API service bigraph (bipartite graph) and perspective view synoptic diagram thereof;
Fig. 2 is that final sorted classification 1 result gives an example in the present embodiment;
Fig. 3 is that final sorted classification 2 results give an example in the present embodiment.
Embodiment
Pass through embodiment below, and combine accompanying drawing, do further bright specifically technical scheme of the present invention.
Embodiment:
Present embodiment is based on ProgrammableWeb; ProgrammableWeb is famous mashup and opening API service catalogue; More than 5,000 mashup and more than 3,000 API service have only been enumerated in February, 2011; And mashup and API are provided some log-on messages of service, comprise their name, URL, supplier, label etc.Mashup is the SOS that is obtained by the API Services Combination, and API is service, therefore meets the requirement of the present invention to data.For the ease of enforcement, the present invention is a carrier with the data on the ProgrammableWeb, and embodiment is provided.
Present embodiment comprises classification of service step and service recommendation step, and concrete grammar is following:
1, classification of service step
Mashup through constituting, can regard the service of a coarsegrain by the API service as.In order to realize the automatic classification of service (mashup, API):
Step 3.1; Use the net instrument of climbing that ProgrammableWeb is gone up title, descriptor, API and the label information that all mashup used from 2005 (when building a station) on January 12nd, 2011 (when instance work is carried out) and climbed, be stored in the local data base.The last data of ProgrammableWeb are all submitted to by the user; There is certain randomness; There are some mistakes in data: there is the repeated registration phenomenon in (1) some mashup, though their titles of some mashup are different, their out of Memory is all the same.Only preserve portion for these mashup in data centralization.(2) some mashup only provides title, but other log-on message disappearance.These mashup will be as experimental data.Simultaneously, the label that is used to identify mashup also exists inconsistent, with a kind of label of implication the multiple different form of expression is arranged, and all represent API like " api ", " Api " and " APIs ", but form is different, some label even misspelling.Use Suffix Stripping Algorithm that label is carried out pre-service, and change into same part of speech, eliminate the inconsistency that exists as far as possible.Finally, data set comprises 5,115 mashup, 750 API and 1,489 label.
Step 3.2, the data that analyzing step 3.1 obtains are analyzed the constituent relation between mashup and the API service, and this constituent relation is represented with bigraph (bipartite graph) (2-mode graph).This bigraph (bipartite graph) comprises two category nodes, promptly represents the node and the node of representing the API service of mahsup.Have the limit between this two category node, expression mashup has used this API service (see figure 1).
Step 3.3 is carried out projection with the bigraph (bipartite graph) that obtains in the step 3.2 in the mashup dimension, obtains corresponding one dimension perspective view---mashup similar diagram (see figure 1).Node in the mashup similar diagram is represented mashup; The similarity relation is represented to exist between the mashup of its two ends in the limit of network, promptly exists similarity that the limit is then arranged between mashup, has no similarity then boundless; Weights on the limit are represented the similarity between these two mashup.Similarity in this example between the mashup is made up of two parts, by the similarity of shared API generation and the similarity that is produced by shared label, is the weighted sum of these two parts.The similarity of these two parts is calculated through Jacard coefficient of similarity (Jacard similarity coefficient), and the similarity that is promptly produced by API (label) is numerically equal to the ratio by the order of the order of the common factor of API (label) and API (label) union.
Step 3.4 is revised the fast algorithm that Newman M E J proposes, and makes it can handle weighted graph, is used for then the mashup similar diagram is carried out cluster, obtains 17 communities.Then that number of individuals in these communities is few several communities remove, and finally obtain 5 classifications.Manual sampling check through to these 5 classification can be known: the mashup in the classification 1 is main relevant with the mobile phone mobile communication; Mashup in the classification 2 is mainly relevant with online office; Mainly some are used with the mashup that goes out line correlation in the classification 3; Mainly be that some mashups relevant with map use in the classification 4; Mainly be that the closely-related mashup of some and social networks uses in the classification 5.Classification 1 is seen Fig. 2 and shown in Figure 3 with classification 2 result, and other classification results is too big because of scale, shows unclear and it is ignored.
Step 3.5; On the basis of the mashup classification results that step 3.4 obtains; According to bigraph (bipartite graph) between mashup in the step 3.2 and API service; Calculate the quantity of each API service mashup of participation in all kinds of, then it is assigned in that maximum type of mashup quantity, thus the classification of realization API service.
2, service recommendation
In order to be mashup developer's recommendation service, the present invention need accomplish following step:
Step 4.1 is carried out projection with the bigraph (bipartite graph) that obtains in the step 3.2 in the API dimension, obtains corresponding one dimension perspective view---API Services Combination figure (see figure 1).The node of API Services Combination figure is represented the API service; The relation capable of being combined between these two services is represented on internodal limit; Weights on the limit are represented the intensity of this relation capable of being combined.The weights on the limit in this example between the API service calculate as follows: if these two API occur in same mashup; We just serve the component of the inverse of number as the similarity of these two API services in this mashup with API among this mashup so, and then that all are such cooccurrence relation adds up just can obtain the intensity of annexation between these two API.
Step 4.2 is analyzed the use pattern between mashup and API service based on the bigraph (bipartite graph) of mashup that obtains in the step 3.2 and API service, analyzes these use patterns and can guidance be provided for we recommend API service and structure Mashup to use.Can find through analyzing; Most of mashup use only used less API service (80.1% mashup use only used 1 with 2 API services), have only 1.1% mashup to use to have used the API service that surpasses 10 (therefore with most of mashup use API serve one of number more among a small circle [
A,
B] be decided to be [1,10], in the present embodiment, confirm [
A,
B] during scope,
Get
), on average each mashup uses and has only used 13.8 API services.This explains that most of mashup application is only formed by a spot of API Services Combination.
Step 4.3, most API service have only been participated in less mashup (1 and 2 mashup have only been participated in 50.7% API service), have only 17.9% API service to participate in the mashup greater than 10, and on average each API serves and only participated in 9.5mashup.This explains that most of API service only participated in a spot of mashup and used.Can be called core API service with participating in 10 maximum API services of mashup, because they become the formation unit that a lot of mashup use.
Step 4.3, select to carry out following steps:
Step 4. 31; If the user does not also select any service; Then recommend core API service, that is: Google Maps, Flickr, YouTube, Twitter, Amazon eCommerce, Facebook, eBay, Microsoft Virtual Earth, Last.fm, Google Search for it;
Step 4. 32 if the user has selected an API service, be that it recommends to link to each other with this API service according to API Services Combination figure in the step 4.1 then, and weights comes top-between them
kService (
kBe the parameter of user's input, if
kGreater than the degree of this service then
kBe the degree of this service, otherwise
kConstant).As with " 23 " serving as the API service of inquiry in advance, the top-5 that then returns API service is: Flickr, del.icio.us, Google Picasa, Google Maps and Riya;
Step 4. 33 if the user has selected a plurality of (>=2) service, is asked all combinations of optional 2 points from these points, and to each combination, asking with these two points is terminal, and comprises in the path of other service, and routine weight value and maximum and step-length exist
B(comprise with interior
B) top-
kIndividual service is recommended.As serving as inquiry API service with " 23 " and " Riya ", in 3 steps of returning all paths in top-2 be: [23, Flickr, Riya] and [23, Google Picasa, Riya].
Specific embodiment described herein only is that the present invention's spirit is illustrated.Person of ordinary skill in the field of the present invention can make various modifications or replenishes or adopt similar mode to substitute described specific embodiment, but can't depart from spirit of the present invention or surmount the defined scope of appended claims.
Claims (3)
1. one kind based on Services Combination historical classification of service and recommend method, it is characterized in that comprise classification of service step and service recommendation step, concrete grammar is following:
Described classification of service step comprises:
Step 1.1; Collect the metamessage data of SOS; Comprise: the label of the set of service that the title of SOS, SOS descriptor, SOS use, sign SOS, the time that SOS adds, the developer of SOS, the address of SOS; And the metamessage data that will collect SOS carry out error in data and handle, and the data after these are handled are stored in local data base;
Step 1.2, the data that analyzing step 1.1 obtains are analyzed the constituent relation between SOS and the service, and this constituent relation are represented with bigraph (bipartite graph) said bigraph (bipartite graph) comprises two category nodes, comprise the node of representing SOS and the node of representing service;
Step 1.3 is carried out projection with the bigraph (bipartite graph) that obtains in the step 1.2 in the SOS dimension, obtains corresponding one dimension perspective view---SOS similar diagram;
Step 1.4 utilizes clustering algorithm or community discovery algorithm that the node in the SOS similar diagram in the step 1.3 is carried out cluster, and the SOS node division is become some classes and check to be all kinds of specified class another names in each type behind each SOS;
Step 1.5 on the basis of the SOS classification results that step 1.4 obtains, according to bigraph (bipartite graph) between SOS in the step 1.2 and service, is calculated the SOS quantity that each service is participated in all kinds of, then it is assigned in that maximum type of SOS quantity, accomplishes classification of service;
Described service recommendation step comprises as follows:
Step 2.1 is carried out projection with the bigraph (bipartite graph) that obtains in the step 1.2 in the service dimension, and obtaining corresponding one dimension perspective view is Services Combination figure, the node representative service of said Services Combination figure; Bian represents the relation capable of being combined between the service that this edge two ends node representes; Weights on the limit are represented the intensity of this relation capable of being combined, and these weights can be measured through the cooccurrence relation of service in SOS of this edge two end nodes representative;
Step 2.2; Analyze the use pattern between SOS and service according to the bigraph (bipartite graph) of SOS that obtains in the step 1.2 and service, promptly SOS uses several services, and service is used by several SOS; And analyze these digital distributions, thereby confirm most of SOS use one of the service number more among a small circle [
A,
B], confirm scope [
A,
B] method be: if having
NIndividual SOS is with this
NIndividual SOS sorts by the service number of its use from small to large, gets to come the front
Individual SOS, wherein,
For more than or equal to
Percentage value less than 100% is established and is come
The service number that the SOS of position uses does
Q, so just be provided with [
A,
B] be [1,
Q];
Step 2.3, with frequency of utilization high before
BIndividual service is as kernel service;
Step 2.4, select to carry out following steps:
Step 2.41 if the user does not also select any service, then is the kernel service in its recommendation step 2.3;
Step 2.42 if the user has selected a service, be that its recommendation links to each other with this service according to Services Combination figure in the step 2.1 then, and weights comes top-between them
kService, wherein,
kBe the parameter of user's input, if
kGreater than the degree of this service then
kBe the degree of this service, otherwise
kIt is constant,
kGet arbitrary integer greater than 0,
kBe worth greatly more, proposed algorithm required time in recommendation service is long more;
Step 2.43, if the user selected more than or equal to the service more than 2: all that at first ask these services make up in twos; To each combination, in Services Combination figure, asking with two services in this combination is terminal then, and comprises all paths of other service (service two services in this combination in the selected service); The limit power of asking every paths at last respectively and, and the power of limit by path and to all path descending sorts, with top-
kThe user is recommended in the path.
2. according to claim 1 a kind of based on Services Combination historical classification of service and recommend method; It is characterized in that; In the said step 1.2, all there is the limit in the node of the said SOS of representative with the node of representative service, the constituent relation between expression SOS and the service; If promptly a SOS is made up of 3 services; There is the limit in the node of then representing this SOS respectively and between 3 nodes of representative service, but does not have the limit between similar node, promptly represents between the node of SOS and between the node of representative service not have the limit.
3. according to claim 1 a kind of based on Services Combination historical classification of service and recommend method, it is characterized in that in the said step 1.3, the node of SOS similar diagram is represented SOS software; The similarity relation is represented to exist between the SOS of its two ends in the limit of network, promptly exists similarity that the limit is then arranged between SOS, has no similarity then boundless; Weights on the limit are represented the similarity between these two SOS, and these weights can be measured through metamessage between these two SOS, comprise the similarity of descriptor, shared service, shared label, common developer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110175055A CN102331929A (en) | 2011-06-27 | 2011-06-27 | Service classification and recommendation method based on service combination history |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110175055A CN102331929A (en) | 2011-06-27 | 2011-06-27 | Service classification and recommendation method based on service combination history |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102331929A true CN102331929A (en) | 2012-01-25 |
Family
ID=45483715
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110175055A Pending CN102331929A (en) | 2011-06-27 | 2011-06-27 | Service classification and recommendation method based on service combination history |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102331929A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609264A (en) * | 2012-02-14 | 2012-07-25 | 深圳市同洲视讯传媒有限公司 | Method and device for generating calling codes by calling application programming interfaces |
| CN102866911A (en) * | 2012-09-12 | 2013-01-09 | 北京航空航天大学 | Mashup application establishing method and device |
| CN103473128A (en) * | 2013-09-12 | 2013-12-25 | 南京大学 | Collaborative filtering method for mashup application recommendation |
| CN104932944A (en) * | 2015-06-15 | 2015-09-23 | 浙江金大科技有限公司 | Cloud computing resource service combination method based on weighted bipartite graph |
| CN105354327A (en) * | 2015-11-26 | 2016-02-24 | 中山大学 | Interface API recommendation method and system based on massive data analysis |
| CN106354844A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Service combination package recommendation system and method based on text mining |
| CN109144498A (en) * | 2018-07-16 | 2019-01-04 | 山东师范大学 | A kind of the API auto recommending method and device of object-oriented instantiation task |
-
2011
- 2011-06-27 CN CN201110175055A patent/CN102331929A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609264A (en) * | 2012-02-14 | 2012-07-25 | 深圳市同洲视讯传媒有限公司 | Method and device for generating calling codes by calling application programming interfaces |
| CN102866911A (en) * | 2012-09-12 | 2013-01-09 | 北京航空航天大学 | Mashup application establishing method and device |
| CN102866911B (en) * | 2012-09-12 | 2015-03-25 | 北京航空航天大学 | Mashup application establishing method and device |
| CN103473128A (en) * | 2013-09-12 | 2013-12-25 | 南京大学 | Collaborative filtering method for mashup application recommendation |
| CN104932944A (en) * | 2015-06-15 | 2015-09-23 | 浙江金大科技有限公司 | Cloud computing resource service combination method based on weighted bipartite graph |
| CN104932944B (en) * | 2015-06-15 | 2018-04-17 | 浙江工商大学 | Cloud computing resources service combining method based on cum rights bigraph (bipartite graph) |
| CN105354327A (en) * | 2015-11-26 | 2016-02-24 | 中山大学 | Interface API recommendation method and system based on massive data analysis |
| CN106354844A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Service combination package recommendation system and method based on text mining |
| CN106354844B (en) * | 2016-08-31 | 2020-08-04 | 上海交通大学 | Service combination package recommendation system and method based on text mining |
| CN109144498A (en) * | 2018-07-16 | 2019-01-04 | 山东师范大学 | A kind of the API auto recommending method and device of object-oriented instantiation task |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | A cocktail approach for travel package recommendation | |
| CN102331929A (en) | Service classification and recommendation method based on service combination history | |
| US7627542B2 (en) | Group identification in large-scaled networks via hierarchical clustering through refraction over edges of networks | |
| CN102591917B (en) | Data processing method and system and related device | |
| CN102750375B (en) | Service and tag recommendation method based on random walk | |
| CN100384134C (en) | Client-end resource search under broadcast-storage network environment and automatic downloading method | |
| CN104995870A (en) | Multi-objective server placement determination | |
| CN105721279B (en) | A kind of the relationship cycle method for digging and system of subscribers to telecommunication network | |
| Gomes et al. | An infrastructure model for smart cities based on big data | |
| CN104077723A (en) | Social network recommending system and social network recommending method | |
| CN105229636A (en) | The social perception local search of context | |
| CN103838857A (en) | Automatic service combination system and method based on semantics | |
| CN113868549B (en) | Advertisement putting optimization method and device, electronic equipment and storage medium | |
| CN111770020A (en) | Method for network perception service combination algorithm based on optimal path selection | |
| CN112702390B (en) | Block chain-based networking method and device for intelligent contract resources | |
| Faber et al. | Keeping in touch with citizens online: Social media usage in Dutch local government | |
| CN112148866A (en) | Online intelligent project declaration resource matching pushing system and method thereof | |
| CN110704612B (en) | Social group discovery method and device and storage medium | |
| CN101442466B (en) | Superpose network and implementing method | |
| Zhang et al. | Context-aware generic service discovery and service composition | |
| CN114443920A (en) | Address information analysis method and device, electronic equipment and computer readable medium | |
| CN103297538A (en) | Aggregation method for resource environmental models | |
| Balakrishnan et al. | Integrated quality of user experience and quality of service approach to service selection in internet of services | |
| CN111930829A (en) | Standard address generation method, device, equipment and medium | |
| Prasetyo et al. | Implementation of service platform for smart city as a service |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120125 |