CN101022377A - Interactive service establishing method based on service relation body - Google Patents
Interactive service establishing method based on service relation body Download PDFInfo
- Publication number
- CN101022377A CN101022377A CN 200710002989 CN200710002989A CN101022377A CN 101022377 A CN101022377 A CN 101022377A CN 200710002989 CN200710002989 CN 200710002989 CN 200710002989 A CN200710002989 A CN 200710002989A CN 101022377 A CN101022377 A CN 101022377A
- Authority
- CN
- China
- Prior art keywords
- service
- theme
- user
- relation
- candidate
- 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
A method for setting up interactive service based on service relation body includes analyzing input parameter semantics in service request after user service request is received, confirming service subject according to analyzed semantics, confirming candidate service in service relation body according to confirmed service subject, returning matched candidate service result and service-ending confirmation to user, using selected candidate service as new service request to return to starting up step if candidate service is selected by user and ending service setting-up process if ending service is selected by user.
Description
Technical field
The present invention relates to a kind of service creating method of comprehensive existing network service, relate in particular to a kind of interactive service establishing method based on service relation body.
Background technology
Along with development of Internet, Web service has obtained the extensive approval of industrial quarters and academia as a kind of professional presentation mode of opening.Therefore yet the function of single Web service is limited, is difficult to satisfy user's requirement, makes up the existing Web service, generates complicated business and become the current research focus.Present Web service provides service generation method that method comprises user-interactive (referring to Sirin, E.; Parsia, B.; Hendler, J.Filtering and Selecting Semantic Web Services with InteractiveComposition Techniques.Intelligent Systems, IEEE, Volume 19, Issue 4, Jul-Aug2004,42-49), said method is in the service generative process, utilize the service matching algorithm to recommend the follow-up service of mating to the user, the user selects follow-up service on demand, finds out the follow-up service that this service may be mated again after carrying out service, according to this, until the whole needs that satisfy the user.The anabolic process of its service is participated in by the user, has been formulated after the whole services stream by the user, unifiedly carries out and calls.
With respect to full automatic service generation method (referring to LI Man, WANG D-Zhi, DUXiao-Yong, WANG Shan, Dynamic Composition of Web Services Based on DomainOntology, Chinese Journal of Computers, Vol 28, No4, April 2005; And Seog-Chan Oh, Byung-Won On, Eric J.Larson, and etc.BF
*: Web Services Discovery andComposition as Graph Search Problem BF
*: Web services discovery andcomposition as graph search problem.IEEE Proceedings of e-Technology, e-Commerce and e-Service, 29 March-1 April 2005,784-786), the user-interactive method is a kind of semi-automatic technique, it can avoid the user that demand is carried out complete, formal description, and adds access customer control, and flexibility ratio and fitness all obtain bigger raising.The legacy interactive service combining method, utilize user interactions to set up the flow process of service, be that the user participates in serving generative process, but provide the class service for information, the user often needs to decide according to the result of preceding step service execution the selection of follow-up service, needs for this reason to realize that the execution and the user that serve select to carry out alternately.
The key of interactive services combined method is, provides candidate service to the user.The service matching algorithm is a key point of interactive web service combination, promptly how to obtain the follow-up service that is complementary with current service, makes the user can find satisfied service in some scopes, and need not searching in a large amount of irrelevant services.
For the service recommendation method, most of research produces control flow based on exchanges data between service, promptly is complementary as match-on criterion with service interface.Service interface comprises the input parameter and the output parameter of service, the core of service interface coupling is specifically: the output of preceding step service preS, if can mate mutually with the following input of step service nextS, these two services can connect so, and promptly nextS can be used as a candidate service.(I.Budak?Arpinar，Ruoyan?Zhang，Boanerges?Aleman-Meza，andetc.Ontology-Driven?Web?Service?Composition?Platform.In?Proceedings?of?theIEEE?International?Conference?on?E-Commerce?Technology，146-152)。Huge but the process of Interface Matching method search services is complicated, and only depend on the coupling of interface often to obtain the unwanted result of some users, cause the user to select difficulty.Utilize service parameter and user's qos requirement to filter, can dwindle follow-up service range, such as the response time to service, the service success rate limits, to dwindle the quantity of follow-up service.But the user need be provided with the qos parameter of complexity, very complicated concerning the user, and often a large amount of follow-up services is incoherent on the function, and filters the not function attribute of only having considered service according to QoS, can not control the function of service.After the queryAddress service, it is output as address, all services that are input as address all are called candidate service, as inquire about nearby hospitals, near the school inquiry, near the government organs of inquiry etc., the input of these services all is address, and user inquiring is certain address, bar, therefore may have no stomach for to these services.In addition, though the service interface matching process has the automation characteristics, promptly go on foot the output parameter of service in the past, data association between the follow-up service service that service describing can be complementary with it automatically, but other influencing factors of service combination can't be taken into account, as user's custom and business logic etc.The quality that provides of web service has been provided these reasons.
Above Interface Matching method is all described based on the interface of service and is carried out, these descriptions all are the descriptions of static semantic, and the static state of service is described the characteristics in the time of can't embodying service operation, promptly can not hold user's actual concern point, this has also caused the demand that the service that generates can not fine applying user, thereby causes the decline of QoS.
Summary of the invention
In view of this, main purpose of the present invention is to provide a kind of interactive service establishing method based on service relation body, can generate complicated service according to user's service request, fully to satisfy user and service supplier demand in many ways.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of interactive service establishing method based on service relation body from the incidence relation of at least three Perspective Analysis services, is created service relation body with the incidence relation that analyzes; This method is further comprising the steps of:
A, receive user's service request after, the semanteme of input parameter in the analysis service request is determined the theme that service relates to according to the semanteme of being resolved;
B, in service relation body, determine the candidate service of coupling according to the service theme of determining in the steps A;
C, with the result of matching candidate service and finish service and confirm to be back to the user, if the user selects candidate service, then with selected candidate service as new service request, return steps A; If the user selects to finish service, then finish the service-creation flow process.
Wherein, the incidence relation of described service is specially:
From the service theme visual angle service is resolved, according to the incidence relation between the service of the foundation of the incidence relation between the service theme;
From the user behavior visual angle service is analyzed, learnt the incidence relation that gets according to user's service usage log;
The incidence relation that forms at the required business rules of user and strategy that provides according to the third party service provider from the business strategy visual angle service is provided.
Wherein, describedly create service relation body with the service theme incidence relation and may further comprise the steps:
A1, current service is carried out semanteme resolve, determine the theme of each current service;
A2, set up body in the field, set up the incidence relation of service theme with the object of determining in the steps A 1 that themes as;
A3, each current service is corresponded to each service theme in the steps A 2 described domain bodies.
Wherein, comprise step after the steps A 3:
A4, incidence relation, the incidence relation between the service between the next theme of study between the visual angle foundation service of user behavior;
A5, the incidence relation between the service theme set up respectively among steps A 2 and the A4 and the service, the incidence relation between the service are integrated the formation service relation body.
Wherein, also comprise step after the steps A 5:
Business rules that the third party service provider is provided and strategy are described with the semantic web rule language, and are stored in the service relation body, filter by business rules and strategy and expand with the result to candidate service.
Wherein, also comprise after the step B:
To all candidate service that obtain among the step B, calculate inheritance or binary crelation chain that each candidate service and service request exist, determine again to relate to the weighting relationship gap that concerns weight between any two candidate service, the weighting relationship gap of per two candidate service of cycle calculations, and according to this all candidate service are sorted.
Wherein, the incidence relation that study gets according to user's service usage log determine be specially:
Carry out the behavioral statistics of service call according to user's inquiry log, between the service and the frequency of calling of theme and service surpass setting threshold, promptly be defined as having incidence relation between the service.
Wherein, described service is described by the element in the field ontology library, and includes dynamic semantics and static semantic;
Determine described in the steps A that service theme is: determine the current theme of service with the service dynamic semantics.
Wherein, determine described in the step B that the candidate service of mating comprises:
B1, according to the theme of input service, determine that all services that this theme comprises are the service search space;
B2, the IO Interface Matching is carried out in the service in the service search space of determining, the service that the match is successful adds in the candidate service tabulation;
B3, according to service relation body, find out with service request in theme have the service of theme and service relation, join in the described candidate service tabulation of step B2.
Wherein, also comprise after the step B3:
B4, call business rules and the strategy that the semantic web rule language describes and carry out reasoning, utilize The reasoning results the candidate service in the described candidate service tabulation of step B3 is filtered and to expand.
The present invention utilizes the relation body of the web service semantics service of setting up, and the service request of input is resolved with web service semantics storehouse, like this can be according to the particular user characteristics, and efficient and convenient when carrying out follow-up service coupling.The service relation body database that the present invention creates fully takes into account between the service, the incidence relation between service and the service theme and the relation between the service theme, and consider business logic and user's use habit, therefore function is very powerful, having more practicality can provide the series of services relevant with its service request theme to the user, and service provides comprehensive.When serving coupling, utilize the semanteme of service request to determine that the user pays close attention to theme, and according to the definite field of serving of this theme, only mate in the field, can filter out with the user and pay close attention to the incoherent content of content, rationally dwindle the search volume of coupling service, when providing accurate service to the user, save the processing resource, promoted treatment effeciency.Owing to considered commercial viewpoint, make the service recommendation process to be pushed away then and control that third party's business development merchant can formulate business rules voluntarily to attract more users by commerce.The user behavior viewpoint can reflect most of user's daily behavior customs simultaneously, therefore can provide the candidate service that more meets user view.
Service creating method collection of the present invention selects service, execution service, the service of creating to be one, can guarantee that the user decides the selection of follow-up service according to service execution result of last step.
Description of drawings
Fig. 1 is the schematic flow sheet of setting up of service relation body of the present invention;
The fragment schematic diagram of Fig. 2 for concerning between the service theme of the present invention;
Fig. 3 is the schematic diagram of service relation body of the present invention;
Fig. 4 is the visioning procedure figure of service relation body of the present invention;
Fig. 5 is the flow chart of service-creation of the present invention.
Embodiment
Before technical scheme of the present invention is described in detail, at first introduce Semantic Web Services and semantic body.
In order to make machine perception and process information automatically, carried out at present with RDF (referring to FrankManola, Eric Miller, Resource Description Framework (RDF) primer, W3CRecommendation 10 February 2004), OWL (referring to: Grigoris Antoniou, Frank vanHarmelen.Web ontology language:OWL.In:Steffen Staab, Rudi Studer eds..Handbook on Ontologies in Information Systems.Springer Verlag, 2003,67-92) be the research of the Web of future generation on basis, with Semantic Web (Sheila A.McIlraith, Tran CaoSon, Honglei Zeng, Semantic Web Services[J], In IEEE Intelligent Systems (Special Issue on the Semantic Web), 2001.46-53) technology combines with the Web service technology, the Semantic Web Services technology that produces makes the service semantics can be by between man-machine, mutual understanding between the machine.
Again semantic body is described.Ontology originates from the Western Philosophy in ancient times, from the eighties of last century the eighties, ontological thought is introduced into artificial intelligence field, and people begin progressively real world to be carried out modeling, from certain field, take out the set of notion and the relationship of the concepts, promptly form the body in this field.Body helps to change different groups or system adopts the situation of different term descriptions to identical concept in the field, thereby is widely used in numerous areas such as digital library, information integration.The present invention is that semanteme is with the auxiliary user interest point of finding in the time of will excavating service operation, and to recommend other services relevant with its service that is just moving to the user, providing of service is more humane, and service quality is improved.
The present invention at first is described the establishment of service relation body.Service relation body of the present invention is to set up on the basis of present existing web service.As shown in Figure 1, the establishment step of service relation body comprises:
| Service S | Theme |
| Inquire address service S1 | Spot |
| Inquiry level of consumption service S2 | Fee&Entertainment |
| Enquiring telephone number service S3 | Telephone&Spot |
| Inquiry hotel condition service S4 | Hotel&Level |
| Inquiry movie news S5 | Cinema&Film |
Table 1
Among the present invention, the foundation of service relation body is improved and whether rationally will directly be influenced follow-up service-creation.Service relation body is one of core for the present invention.Though the service relation body of setting up in the step 103 can directly use, its function is still unsound, and the service relation of establishment neither be very complete, also needs this service relation body is expanded accordingly.
Abovementioned steps has only been set up the direct incidence relation of service theme, and web service is corresponded to each corresponding theme, sets up relation between the web service with this.This relation is elementary incidence relation, and the relation between service is not only the incidence relation of theme.As shown in Figure 4, be the schematic flow sheet that improves service relation body of the present invention.This flow process is to carry out on the basis of Fig. 1, and step 101 is to 103 and aforementioned identical.Promptly after step 103, also need the service relation in the service relation body is further optimized, to generate final service relation body.In step 401, the present invention also analyzes service relation from the user behavior visual angle, and the incidence relation of learning to come according to user's service usage log further improves service relation body, makes the service association relation more plentiful, the user's that more can fit demand.For example by user's service recorder, can find when the user calls the inquire address service, user for certain editorial office of input, select follow-up service bigger for the click volume of inquiry postcode or email address service, this selection had both met statistical law, then will inquire about postcode or email address service and the service association of inquiring about editorial office; And for the user who imports certain sight spot, the follow-up service of selecting often is the inquiry weather service, and be indifferent to postcode or email address, if the click volume of inquiry sight spot service back inquiry weather service has exceeded setting threshold, then will inquire about weather service and set up incidence relation with the service of inquiry sight spot.Has the theme progressive relationship between theme here and the service.Only be the listed example of explanation below, according to the user behavior daily record, over time, if the statistical law that exists the user to click between some service had both thought that it had incidence relation.Certainly, if the incidence relation that user behavior is set up is created in service relation body before, then no longer repeat to create.
The present invention utilizes the method for incidence relation between the user journal study service to realize by the algorithm of propositions such as Agrawal.Agrawal equals the correlation rule problem (R.Agrawal that has at first proposed to excavate between client's transaction data base discipline collection in 1993, T.Imielinski, and A.Swami.Mining association rulesbetween sets of items in large databases.Proceedings of the ACM SIGMODConference on Management ofdata, pp.207-216,1993.), its core methed is based on the theoretical recurrence method of collection frequently, has proposed to excavate the important implementation method of correlation rule.The excavation correlation rule that the present invention utilizes Agrawal to propose is realized the service association relation at user behavior visual angle.At first, be the recording user behavior, promptly add up for user's services selection step.Excavate potential association between the service by machine learning method.Then, find the item collection (Itemset) of all supports greater than minimum support (preset threshold), these collection are called collection (Frequent Itemset) frequently; Use the aforementioned frequency collection that finds to produce the rule of expectation at last.The service association relation of utilizing the user behavior visual angle to obtain has science very much, and these incidence relations are to get by a large amount of user behavior statistics, and are very representative, also meets most user's demand.
Below service-creation mode of the present invention is described.
As shown in Figure 5, service creating method of the present invention comprises:
Step 501, receive user's service request after, the semanteme of analysis service input parameter, the theme that the service determined relates to.For web service, all have static semantic and dynamic semantics, wherein, the semanteme that the static semantic of service embodies in describing for service OWL-S, and the semanteme that the dynamic semantics of service is had when being service operation.The service dynamic semantics can embody the theme that the user pays close attention to, and its levels of precision is higher than the static semantic of service.Therefore the output parameter dynamic semantics often is subjected to the influence of input parameter dynamic semantics, and the dynamic semantics of service and the relation of static semantic have three kinds (seeing Table 2 in detail for example):
1. Si.DS Si.SS, promptly DS is the concrete of SS and deepens (as S1 in the table 2);
2. Si.DS=Si.SS ∪ N, N represent new notion set, and promptly DS gives new semanteme (as S2 in the table 2) to SS;
3. Si.DS=Si.SS promptly describes to a certain extent at service interface, and DS also can identical with SS (as S3 in the table 2).
| Relationship type | Service S | Parameter value | I Ds(S) | I ss(S)10 |
| I i.DS I i.SS | Inquire address service S 1 | The seventh heaven restaurant | Restaurant | Spot |
| I i.DS＝ I i.SSN | Call out switched service S 2 | 62254741 (little Western Paradise cinema phone) | Cinema | TelephoneNumber |
| I i.DS＝I i.SS | Inquiry speciality service S 3 | Small stream middle forest dining room | Restaurant | Restaurant?15 |
Table 2
The present invention analyzes the input parameter result of service, because input parameter is phrase, obtains its keyword and matches notion example in the domain body, determines its corresponding class according to the notion example, and determines theme exactly.Method is as follows, at first utilize ICTCLAS Words partition system (Hua-Ping Zhang, Hong-Kui Yu, De-Yi Xiong and Qun Liu, HHMM-based Chinese Lexical AnalyzerICTCLAS, In Proceedings of 2nd SigHan Workshop, July 2003,184-187) input information is carried out participle, in word segmentation result, extract rearmounted noun then, utilization knows that (Dong Qiang, Hao Changling, Dong Zhendong extract http://www.keenage.com/html/c_index.html based on the Chinese message structure of knowing net to net hownet; Xu Yun, Fan Xiaozhong, Zhang Feng, lexical semantic relatedness computation based on " knowing net ", Beijing Institute of Technology's journal 2005.5) API that provides (Application Programming Interface) can obtain the English of its correspondence and explain, is mapped to corresponding notion in the domain body.For example, domain body as shown in Figure 2, when the user selects the inquire address service, input parameter is " little Western Paradise cinema ", and system obtains its rearmounted noun and is " cinema " its participle, the API that utilizes hownet to provide, the English that can obtain its correspondence is interpreted as Cinema, the URI of this notion example of location in domain body, and the semanteme of service request is resolved and can be realized.It will be appreciated by those skilled in the art that the generation of service request and identify its theme and dynamic semantics is easy to realize by above-mentioned technology.
Step 502 according to the service theme of determining in the step 501, is determined the candidate service coupling in the domain body of correspondence in service relation body.The purpose of determining service theme is to serve when mating more accurate.If user's service request is the occasion inquiry, there is no need to mate in the weather service field.The field of the invention location can avoid a large amount of couplings to calculate, not only saved system resource and promoted efficient, and, service theme determine more to help serving the accuracy that provides.At first determining the search volume of service coupling according to service theme, obtain the father node of service theme, is object with father node and all child nodes thereof, obtains and the related service of its theme (topicOf), as the search volume of coupling service.
For example: if judge service request is Cinema, the theme diffusion can be obtained its amplification theme Entertainment, as Fig. 3, recommends the theme of incidence relation upper level.Can for example also can recommend to the Spot rank according to the relating subject rank of determining to recommend.Consider that present service relation body hierachy number is less, the present invention only recommends a level to be clipped to Entertainment, and do not consider the service clearance in Office territory, only the service that is associated with Entertainment to user's recommendation comprises themes such as Entertainment, Cinema, Bar, Theatre.Because when the user pays close attention to this field of cinema, the possibility of the related service of going again to select to handle official business is minimum.Like this, the services selection that offers the user will be more accurate.Matching algorithm of the present invention mates by the method for IO interface in the associated service of this theme, can reduce the size of coupling service clearance, can improve the satisfaction of matching result simultaneously, obviously is better than traditional semantic matching method based on interface.
Matching candidate service of the present invention also comprises the association service with above-mentioned service theme.These association service promptly are the association service of setting up by the user behavior visual angle, and the present invention obtains with last step service and current theme has the service of direct correlation as recommendation service.For example, the user has carried out the inquire address service, and its input parameter is " a Beijing University of Post ﹠ Telecommunication editorial office ", and it is editorial office (newsroom) that system obtains service theme, theme newroom has association service queryMailCode service, and therefore the coupling service should comprise the service of inquiry postcode.And if user's input parameter is " Fragrance Hill ", it is the mountain that system obtains service theme, and its theme has association service inquiry weather service, and therefore the coupling service should also can comprise the inquiry weather service.
Step 503, preceding step user is called service result, coupling candidate service result and finish service and confirm to be back to the user.Thereby the present invention collects the selection service, the service of carrying out is one, can guarantee that the user decides the selection of follow-up service according to service execution result of last step.
Step 504, judge user's selection result, promptly can be according to preceding step service result according to the user, if in candidate service, select required follow-up service or reached the service request of oneself then finish service-creation, here, comprised the service that meets service request in the candidate service.Concrete select and operation is: if the user selects candidate service, then with this candidate service as new service request, and be back to step 501, again service request is resolved, and continues the service coupling; If the user selects to finish service, then enter step 505.
Step 505 finishes the service-creation flow process.
At present, the user always is desirable to provide the service of the complicated satisfying personalized characteristics of energy, generates for complicated business, and only the description to serve individual is not enough, and the present invention starts with from relationship between services.Mostly the complex services that provides at present is to satisfy this demand by broad covered area, if but coverage rate is too wide, for user's selection, will compare difficulty.The present invention collect complex services generation, call, be implemented as a RUP, the user can select follow-up service according to the execution result of last step service.And, more can fit user's demand of follow-up recommendation service, under the situation that guarantees service coverage, the quantity of recommendation service can be fewer.
Though the service relation body that abovementioned steps is set up has been considered user's demand more all sidedly, set up very practical service association relation, but provide level in order to promote service of the present invention, after step 503, promptly after generating user's candidate service and before returning the user, also comprise the step that business rules filters and expands candidate service.
Because service combination itself has commercial characteristic, and the present invention can provide the service of the abundanter user's request of more fitting to the user by third party's business development merchant, so business strategy also is the key factor that the present invention realizes service-creation.The present invention need analyze service from the business strategy visual angle, and generates corresponding business rules according to business strategy, thereby finishes the optimization to the coupling service, and realizes that some users use the restriction of service.For example for the inquiry queryRestaurant of restaurant service, from the angle of business strategy, the user also can pay close attention to the association service of queryRestaurant such as the brief introduction, discounting activity in restaurant.Again for example, if the user is a minor, will not allow services such as its inquiry adult store information so.For example the user according to the setting of business rules, can directly provide its association service to order the service of plane ticket or bus ticket to the user under the situation of ordering the train ticket serv-fail.The application of business strategy makes coupling service have more hommization, provides the most appropriate the most comprehensive service with the demand by the user as far as possible to it.These association service also are the services that next step most probable of user is inquired about and called from commercial visual angle.These business strategies are described by descriptive language SWRL (Semantic Web Rule Language, semantic web rule language), and triggering is triggered when calling service, thereby above-mentioned service by the business strategy association is back to the user as candidate service.
In addition, though most business rules has been considered user's needs of problems, also there is the phenomenon that has contradiction between business rules The reasoning results and the candidate service.For example business strategy " if the user does not send the authority of information, then can not use to send messenger service (sendMessage ())." with recommendation service (with request service theme related service) " after the user carries out the inquire address service, trigger show near map, send the multimedia message service and be optional state." be exactly a pair of paradox.In the time of operation, if a user who does not send the information authority has called the inquire address service, then Fu Wu The reasoning results will produce as above conflict.The present invention for the conflict of rule adopt mode to service initialization priority, its principle be " unallowed " priority greater than " can do ", " can do " priority be greater than " can do ", i.e. P (disable)＞P (do)＞P (enable).Wherein p represents priority, and the present invention sets priority to the service association relation of conflict.The conflict processing that should be appreciated that this step is not a special case, and the rule of any conflict can be avoided by the setting of priority-level.Below introduce it in detail.
The present invention utilizes rule description language SWRL (the Ian Horrocks based on body, Peter F.Patel, AProposal for an OWL Rules Language[C], Proceedings of WWW2004,2004 NewYork, USA:723-731) describe business strategy, both can clearly express whole business strategies, can carry out business rules reasoning again based on semanteme.Business rules (strategy) can be understood as " if eligible then carry out (if condition then do) ", is described with the rule description language.These rules mainly are to adjust the relationship between services of the coupling service of returning the user, make it more comprehensive under the reconcilable prerequisite.As shown in following table 3, the present invention utilizes business rules to be optimized processing to the coupling service result of returning the user, can finish the filtration and the expansion of coupling service.It will be appreciated by those skilled in the art that business rules can describe by rule description language SWRL, and be stored in service call port of the present invention, to realize acting on service result.Business rules in the table 3 of the present invention is exemplary explanation, is not special case, and the business rules that any service provider provides all can be described by the description of the invention language, and finally can act among the service-creation result of the present invention.After the filtration and expansion of commercial rule, service-creation of the present invention has more practicality.
| Type | Rule | Definition | Example |
| Support the business rules class | Inference | If the inspection condition is very then adds the new fact | Rule 1: if people's age less than 18, then is the teenager |
| Computation | Provide algorithm computation required mathematic(al) representation | Rule 2:Today-birthDate=age | |
| Support the commercial activity class | Constraint | Expression is to the restriction of commercial activity, condition is an expression formula, its result is true or false, and whether decision triggers service | Rule 3:, then can buy half price recreation ground admission ticket if age of user is lower than 12 years old |
| Action Enabler | Condition is service, is true if service execution is successful, triggers corresponding action | Rule 4:, then book a train ticket if do not subscribe to plane ticket |
Table 3
Table 3 has provided the part business strategy, and these business rules are that the practical application situation according to the user is provided with.These business rules are described by rule description language SWRL, and extensibility and flexibility are more intense.Business rules has suppressed the incidence relation between the part service in the service relation body, according to user's condition the service of recommending has been carried out filtering and corresponding the expansion.For example the user is when the service of carrying out entertainment place inquiry, if the age less than 18 years old, then the service that will be associated with the entertainment place filters out as the occasion that unsuitable teenagers such as inquiry bar pass in and out.Above business rules can be set by the service provider according to user's actual demand, as long as express by rule description language SWRL, and store the business rules that these SWRL express the calling interface place of service relation body into, when calling corresponding service, trigger these strategies according to corresponding condition.It will be appreciated by those skilled in the art that the business strategy in the table 2 is exemplary, is not special case, it more or not limitation of the present invention, any business strategy can be expressed by SWRL, and can be to service relation body generation effect of the present invention, thereby provides more humane service for the user.Below to utilizing SWRL to describe business rules and utilizing business rules to mate service and filter and to describe.
The present invention defines five attributes and expresses candidate service (coupling service) state and restriction, and service state is two kinds:
1. (User Service), expresses a user and carried out this service, and run succeeded success, successfully refers to here serve normally return.
2. (User Service), expresses a user and carries out this serv-fail failed, and improper returning may be for unusually.
The user is restricted to three kinds to service
3. (User Service), expresses for this user enable, and this service is executable state.
4. (User Service), expresses for this user disable, and this service is unavailable.
5. (User Service), expresses for this user do, triggers this service of execution automatically.
Business strategy example in the table 3 is expressed as follows with SWRL:
Rule 1:
User(？x)∧Age(？x，？y)∧swrlb:smallerThan(？y，18)→Younger(？x)
Rule 2:
User(？x)∧Today(？y)∧birthDate(？x，？a)∧age(？x，？b)→swrlb:substract(？b，？y，？a)
Rule 3:
User(？x)∧Adult(？x)→BuyAdultCommodity(？a)∧enable(？x，？a)
Rule 4:
User(？x)∧BuyAirTicket(？y)∧failed(？x，？y)→BuyTrainTicket(？a)∧do(？x，？a)
Those skilled in the art are to be understood that, business strategy in the table 3 is exemplary, it or not special case, it more or not limitation of the present invention, any business strategy can be expressed by having added service state attribute SWRL afterwards, and can be, thereby provide more humane service for the user to recommendation service generation effect of the present invention.These rules of utilizing SWRL to express can act in the coupling service of returning the user, to business rules write and the reasoning using method as follows:
1, utilizes increase income body developing instrument Prot é g é 3.2 (http://protege.standord.edu/) and SWRLJessTab plug-in unit (Martin O ' Connor, Holger Knublauch, Samson Tu, and etcSupporting Rule System Interoperability on the Semantic Web with SWRL, Y.Gil etal. (Eds.): ISWC 2005, LNCS 3729, pp.974-986,2005.) can edit service relation body and business rules.
2, the factual aspect of reasoning is active user's information, and user service and the state carried out, with success (User, Service) or failed (User, Service) expression.
3, the instrument The reasoning results is restriction and the expansion to service, comprise available service enable (User, Service), unavailable service disable (User, Service) and service do (User, Service) three service lists of automatically triggering.
4, judge whether there is conflict in the above-mentioned The reasoning results, judge promptly whether certain service appears in the above-mentioned more than one tabulation simultaneously,, think that then there is conflict in this service if having, judge the list priority at its place, the service that has conflict is deleted from the lower tabulation of priority.After handling conflict, in the service result that step 502 obtains, remove the service in the unavailable service list, place the candidate service tabulation anterior service in the available service tabulation, and call the service of automatic triggering.Like this, the present invention has promptly finished restriction and the expansion to the matching candidate service of the The reasoning results of utilizing business rules.
After step 503, in the candidate service that generates the user and when the user returns the coupling service result, meeting of the present invention is sorted to the candidate service that these return according to the semanteme of service request, stands out as much as possible with the service that the user is paid close attention to, and avoids the user to have no way of selecting.The semantic association degree sorted between the present invention utilized and serves.The semantic association degree is the tightness degree that embodies semantic association between the service, and the present invention utilizes in the body semantic distance between Service Instance to express.Semantic distance is in same body, to the inheritance that exists between two different Service Instances or a kind of tolerance of binary crelation chain.The semantic distance of example in the body, study and use more (ZHANG po.The research and implementationof Semantic Based Web Services Discovery.MS.Thesis at present, Beijing:TsingHuaUniversity, 2005).The method of semantic distance comprises: inheritance and binary crelation are thought of as the relations on an equal basis distances (ERD) of equality distance, have distinguished the different weighting relationship gaps (WRD) that concern weight, consider the GCSM distance of the example degree of depth.In service relation body, the tightness degree between the service that the service semantics degree of association is only considered to be caused by thematic relation, so the present invention has expanded the GSCM distance algorithm, multi-threaded inheritance between the adding service.Specific algorithm is as follows:
Input: the URI of service (promptly serve existence as an example in service ontology, URI is its unique sign) is respectively uri1, uri2
Output: associationDegree, semantic association degree between the service
1) T1=getTopic (uril); // obtain all themes according to the semanteme of uril
T2=getTopic (uri2); // obtain all themes according to the semanteme of uri2
T1＝{T11，T12..T1m}T2＝{T21，T22，T23..T2n}
2) initialization D[m] each value of [n] (be a two-dimensional array, be used for writing down the semantic distance of calculating) is 0;
For(i＝0；i＜m；i++){
For(j＝0；j＜n；j++){
If(T1i＝＝T2j){D[i][j]＝1；}else{D[i][j]＝1/getGCSM(T1i，T2j)}
}
// calculate GCSM distance between each theme, if topic is identical, D=1, otherwise D=1/GCSM (T1, T2).
Dcom=//Dcom is all topic semantic relevancy sums
3) Ddiff (T1, T2)=the T1 T2=m+n//Ddiff differences that is the theme
4) associationDegree=Dcom/Ddiff//degree of correlation equals the merchant of same factor and coefficient of variation
5) return associationDegree
6)End.
Wherein, getGCSM () is the GCSM algorithm, algorithmic code is seen document (ZHANG po.page22-24, The research and implementation of Semantic Based Web Services Discovery.MS.Thesis, Beijing:TingsHua University, 2005.12).
The number of times that per two services need be calculated the GCSM distance in the algorithm of the present invention is the product of two service theme quantity, so algorithm complex is that (m * n) (algorithm complex is represented the time complexity of algorithm to O, be the time that its algorithm needs influenced by the product of m and n value), because m, n value are less positive number, so algorithm complex is not high.Because the frequent calculating of semantic distance can cause the whole algorithm decreased performance, can adopt " buffer memory " technology (ZHANG po.The research and implementation of Semantic BasedWeb Services Discovery.page 19-21, MS.Thesis, Beijing:TsingHua University, 2005.12).In order to make service-creation of the present invention have more practicality, can be in advance semantic distance between the every pair of service in the service relation body be carried out calculated off-line and preserves, when obtaining matching result, by semantic distance these coupling services are sorted, and be back to the user.
As shown in table 4, be the comparative result signal of the present invention and traditional IO matching process.Though service generating mode of the present invention only is the test simulation result, it will be appreciated by those skilled in the art that its notional result should be consistent with result of the test owing to adopt establishment mode of the present invention.As known from Table 3, for same " queryAddress (inquire address service) ", the coupling service " rnextService " of IO matching process has 21 recommendation services, but for the user of desire inquiry cinema, the possibility that concern " near lawyer's office ", " near government department ", " near school ", " nearby hospitals " wherein reach occasions such as " near post offices " does not almost have; And, the not ordering of coupling service of IO matching process, the output of coupling service is at random, user for the inquiry particular theater, attention rate near restaurant, near bar, near market may relatively be hanged down, near the movie theatre comprising, and the user may not too pay close attention to.When particularly the magnanimity service is mated (scale of analogue test of the present invention is limited), when the quantity of recommendation service was huger, the service that the user pays close attention to was not easy to find very much.Coupling service output negligible amounts of the present invention, and the high related service of user's attention rate, it is also earlier to sort, and service quality obviously improves.
| preS | Input (Spot) | Output (Address) | nextService (IOMatch) | NextService (the present invention) |
| QueryAddress (inquire address service) | Five road junction cinemas | No. ten, five road junctions road | 1. near near near near near near near near near near post office 13. transmission information, 14. transfer information 11. nearby hospitals 12. of the school government departments 10. lawyer's offices 9. banks 8. parking lots 7. gas stations 6. markets 5. bars 4. cinemas 3. near the restaurant 2. ... (totally 21) | 1. near near near near automatically triggering service of bank (the totally 11) parking lots 11. gas stations 10. near the movie theatre 6. theater 7. transmission information 8. transfer information 9. 4. inquiries of movie news information 2. movie theatre situations 3. pre-ordered movie tickets enter all to consume 5.: show the peripheral map service |
Table 4
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.
Claims (10)
1, a kind of interactive service establishing method based on service relation body is characterized in that, from the incidence relation of at least three Perspective Analysis services, creates service relation body with the incidence relation that analyzes; This method is further comprising the steps of:
A, receive user's service request after, the semanteme of input parameter in the analysis service request is determined the theme that service relates to according to the semanteme of being resolved;
B, in service relation body, determine the candidate service of coupling according to the service theme of determining in the steps A;
C, with the result of matching candidate service and finish service and confirm to be back to the user, if the user selects candidate service, then with selected candidate service as new service request, return steps A; If the user selects to finish service, then finish the service-creation flow process.
2, the interactive service establishing method based on service relation body according to claim 1 is characterized in that, the incidence relation of described service is specially:
From the service theme visual angle service is resolved, according to the incidence relation between the service of the foundation of the incidence relation between the service theme;
From the user behavior visual angle service is analyzed, learnt the incidence relation that gets according to user's service usage log;
The incidence relation that forms at the required business rules of user and strategy that provides according to the third party service provider from the business strategy visual angle service is provided.
3, the interactive service establishing method based on service relation body according to claim 2 is characterized in that, describedly creates service relation body with the service theme incidence relation and may further comprise the steps:
A1, current service is carried out semanteme resolve, determine the theme of each current service;
A2, set up body in the field, set up the incidence relation of service theme with the object of determining in the steps A 1 that themes as;
A3, each current service is corresponded to each service theme in the steps A 2 described domain bodies.
4, the interactive service establishing method based on service relation body according to claim 3 is characterized in that, comprises step after the steps A 3:
A4, incidence relation, the incidence relation between the service between the next theme of study between the visual angle foundation service of user behavior;
A5, the incidence relation between the service theme set up respectively among steps A 2 and the A4 and the service, the incidence relation between the service are integrated the formation service relation body.
5, the interactive service establishing method based on service relation body according to claim 4 is characterized in that, also comprises step after the steps A 5:
Business rules that the third party service provider is provided and strategy are described with the semantic web rule language, and are stored in the service relation body, filter by business rules and strategy and expand with the result to candidate service.
6, the interactive service establishing method based on service relation body according to claim 4 is characterized in that, also comprises after the step B:
To all candidate service that obtain among the step B, calculate inheritance or binary crelation chain that each candidate service and service request exist, determine again to relate to the weighting relationship gap that concerns weight between any two candidate service, the weighting relationship gap of per two candidate service of cycle calculations, and according to this all candidate service are sorted.
7, according to claim 4ly it is characterized in that the determining of the incidence relation that gets according to user's service usage log study is specially in the interactive service establishing method of service relation body:
Carry out the behavioral statistics of service call according to user's inquiry log, between the service and the frequency of calling of theme and service surpass setting threshold, promptly be defined as having incidence relation between the service.
According to the described interactive service establishing method of arbitrary claim in the claim 1 to 7, it is characterized in that 8, described service is described by the element in the field ontology library, and includes dynamic semantics and static semantic based on service relation body;
Determine described in the steps A that service theme is: determine the current theme of service with the service dynamic semantics.
9, the interactive service establishing method based on service relation body according to claim 5 is characterized in that, determines described in the step B that the candidate service of coupling comprises:
B1, according to the theme of input service, determine that all services that this theme comprises are the service search space;
B2, the IO Interface Matching is carried out in the service in the service search space of determining, the service that the match is successful adds in the candidate service tabulation;
B3, according to service relation body, find out with service request in theme have the service of theme and service relation, join in the described candidate service tabulation of step B2.
10, the interactive service establishing method based on service relation body according to claim 5 is characterized in that, also comprises after the step B3:
B4, call business rules and the strategy that the semantic web rule language describes and carry out reasoning, utilize The reasoning results the candidate service in the described candidate service tabulation of step B3 is filtered and to expand.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710002989 CN101022377A (en) | 2007-01-31 | 2007-01-31 | Interactive service establishing method based on service relation body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710002989 CN101022377A (en) | 2007-01-31 | 2007-01-31 | Interactive service establishing method based on service relation body |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101022377A true CN101022377A (en) | 2007-08-22 |
Family
ID=38710034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710002989 Pending CN101022377A (en) | 2007-01-31 | 2007-01-31 | Interactive service establishing method based on service relation body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101022377A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010015131A1 (en) * | 2008-08-05 | 2010-02-11 | 天津大学 | A semantic-based web service relationship network system |
| CN101848416A (en) * | 2010-04-02 | 2010-09-29 | 宇龙计算机通信科技(深圳)有限公司 | Mobile terminal, application server, application push method and application push system |
| CN102043705A (en) * | 2009-10-19 | 2011-05-04 | 阿里巴巴集团控股有限公司 | Statistical method and apparatus for input behavior |
| CN101414296B (en) * | 2007-10-15 | 2012-07-25 | 日电（中国）有限公司 | Self-adapting service recommendation equipment and method, self-adapting service recommendation system and method |
| CN102609468A (en) * | 2012-01-17 | 2012-07-25 | 电子科技大学 | Matching method for spatial information services based on context awareness and user preferences |
| CN101398835B (en) * | 2007-09-30 | 2012-08-29 | 日电（中国）有限公司 | Service selecting system and method, and service enquiring system and method based on natural language |
| CN101398810B (en) * | 2007-09-30 | 2013-05-01 | 日电（中国）有限公司 | Self-adapting service choice device and method thereof, enquiry system and method thereof |
| CN103488673A (en) * | 2012-06-11 | 2014-01-01 | 富士通株式会社 | Method, controller, program and data storage system for performing reconciliation processing |
| CN103581309A (en) * | 2013-10-22 | 2014-02-12 | 华中科技大学 | Dynamic service combination and selection method and system based on demands |
| CN105488151A (en) * | 2015-11-27 | 2016-04-13 | 小米科技有限责任公司 | Reference document recommendation method and apparatus |
| CN107005576A (en) * | 2015-07-24 | 2017-08-01 | 谷歌公司 | The bridge joint matching identification symbol for link identifiers is generated from server log |
| CN107315831A (en) * | 2017-07-10 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN110188538A (en) * | 2019-04-26 | 2019-08-30 | 北京奇安信科技有限公司 | Using the method and device of sandbox cluster detection data |
-
2007
- 2007-01-31 CN CN 200710002989 patent/CN101022377A/en active Pending
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398835B (en) * | 2007-09-30 | 2012-08-29 | 日电（中国）有限公司 | Service selecting system and method, and service enquiring system and method based on natural language |
| CN101398810B (en) * | 2007-09-30 | 2013-05-01 | 日电（中国）有限公司 | Self-adapting service choice device and method thereof, enquiry system and method thereof |
| CN101414296B (en) * | 2007-10-15 | 2012-07-25 | 日电（中国）有限公司 | Self-adapting service recommendation equipment and method, self-adapting service recommendation system and method |
| WO2010015131A1 (en) * | 2008-08-05 | 2010-02-11 | 天津大学 | A semantic-based web service relationship network system |
| CN102043705A (en) * | 2009-10-19 | 2011-05-04 | 阿里巴巴集团控股有限公司 | Statistical method and apparatus for input behavior |
| CN101848416A (en) * | 2010-04-02 | 2010-09-29 | 宇龙计算机通信科技(深圳)有限公司 | Mobile terminal, application server, application push method and application push system |
| CN102609468A (en) * | 2012-01-17 | 2012-07-25 | 电子科技大学 | Matching method for spatial information services based on context awareness and user preferences |
| CN103488673B (en) * | 2012-06-11 | 2016-12-28 | 富士通株式会社 | For performing the method for reconciliation process, controller and data-storage system |
| CN103488673A (en) * | 2012-06-11 | 2014-01-01 | 富士通株式会社 | Method, controller, program and data storage system for performing reconciliation processing |
| CN103581309A (en) * | 2013-10-22 | 2014-02-12 | 华中科技大学 | Dynamic service combination and selection method and system based on demands |
| CN107005576A (en) * | 2015-07-24 | 2017-08-01 | 谷歌公司 | The bridge joint matching identification symbol for link identifiers is generated from server log |
| US10652221B2 (en) | 2015-07-24 | 2020-05-12 | Google Llc | Generating bridge match identifiers for linking identifers from server logs |
| CN107005576B (en) * | 2015-07-24 | 2020-07-24 | 谷歌有限责任公司 | Generating bridging matching identifiers for linking identifiers from server logs |
| CN111953652A (en) * | 2015-07-24 | 2020-11-17 | 谷歌有限责任公司 | Generating bridging matching identifiers for linking identifiers from server logs |
| CN111953652B (en) * | 2015-07-24 | 2021-11-09 | 谷歌有限责任公司 | System and method for generating a bridge matching identifier for a linking identifier |
| US11363006B2 (en) | 2015-07-24 | 2022-06-14 | Google Llc | Generating bridge match identifiers for linking identifiers from server logs |
| CN105488151A (en) * | 2015-11-27 | 2016-04-13 | 小米科技有限责任公司 | Reference document recommendation method and apparatus |
| CN107315831A (en) * | 2017-07-10 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN107315831B (en) * | 2017-07-10 | 2019-06-07 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN110188538A (en) * | 2019-04-26 | 2019-08-30 | 北京奇安信科技有限公司 | Using the method and device of sandbox cluster detection data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101022377A (en) | Interactive service establishing method based on service relation body | |
| US9542503B2 (en) | Estimation of closeness of topics based on graph analytics | |
| US8447640B2 (en) | Device, system and method of handling user requests | |
| US8626692B2 (en) | Knowledge based method and system for local commerce | |
| JP2011504341A (en) | Method and system for using keyword vectors and associated metrics to learn and predict user correlation of target content messages in a mobile environment | |
| CN105677780A (en) | Scalable user intent mining method and system thereof | |
| CN109255586A (en) | A kind of online personalized recommendation method that E-Governance Oriented is handled affairs | |
| CN105608194A (en) | Method for analyzing main characteristics in social media | |
| US11245649B2 (en) | Personalized low latency communication | |
| CN113505311B (en) | Scenic spot interaction recommendation method based on' potential semantic space | |
| CN105843817A (en) | Method and apparatus for searching on terminal device, and device | |
| CN101639837B (en) | Method and system for automatically classifying objects | |
| KR20130033693A (en) | Method, apparatus and computer readable recording medium for search with exetension data-set of concept keywords | |
| CN111611499A (en) | Collaborative filtering method, collaborative filtering device and collaborative filtering system | |
| Sánchez et al. | On the effects of aggregation strategies for different groups of users in venue recommendation | |
| Zhang et al. | An interpretable and scalable recommendation method based on network embedding | |
| CN102298621B (en) | System for obtaining page user focus degree PageFocus by method for aggregating and displaying same source information search engine based on focus degree | |
| Rudinac et al. | Finding representative and diverse community contributed images to create visual summaries of geographic areas | |
| CN109299368B (en) | Method and system for intelligent and personalized recommendation of environmental information resources AI | |
| CN117056619A (en) | Method and device for determining user behavior characteristics | |
| Billig et al. | Residents' preferences for private amenities and trade-offs associated with various spatial densities and patterns | |
| De Maio et al. | Fine-grained context-aware ad targeting on social media platforms | |
| Casillo et al. | A Tailor made System for providing Personalized Services. | |
| Tan et al. | Mrainf: Multilayer relation attention based social influence prediction net with local stimulation | |
| Agagu et al. | Context-aware recommendation methods |
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 |