1. WSExpress: A QoS-Aware Search Engine for Web Services
ZHANG Yilei
15/03/2010

2. Outline Introduction System Architecture QoS-aware searching approach
Experiments
Conclusion

3. Problem
Web services provided by different organizations can be discovered and integrated to develop service oriented applications.
With the growing number of Web services in the Internet, many alternative Web services can provide similar functionalities to fulfill users’ requests.
How to find an appropriate Web service?

4. How to find an appropriate Web service?
UDDI
Syntactic or semantic matching approaches based on services’ tags in UDDI repository.
Limitations
UDDI can not provide liveness Web service information.
UDDI repositories use arbitrary tagging methods.

5. How to find an appropriate Web service?
WSDL
WSDL files are used for describing Web service.
WSDL files can be obtained in several ways.
Exist search engines

6. How to find an appropriate Web service?
Limitations of these Web service search engines
Functionality
keywords cannot represent Web services’ underlying semantics.
Keywords cannot precisely specify the information user needs and the interface acceptable to the user.
Non-functionality
cannot distinguish the non-functional differences between Web services.

7. How to find an appropriate Web service?
How about using QoS-driven selection approaches? No
A basic assumption: all the Web services in the candidate set share identical functionality.
selection approaches can only differentiate among Web services’ non-functional QoS characteristics, regardless of their functionalities.
Web services whose functionalities are not exactly equivalent to the user searching query are completely excluded from the result list.
Web services in the result list are ordered only according to their QoS metrics.

8. Outline Introduction System Architecture QoS-aware searching approach
Experiments
Conclusion

9. WSExpress
Input: user query
Output: Web service ranking list.
Pay respect to functional attributes as well as non-functional features of Web services.
Non-functional evaluation
Obtains QoS criteria values
QoS utility computation
Functional evaluation
WSDL preprocessing
Similarity computation

10. A motivating example

11. Outline Introduction System Architecture QoS-aware searching approach
Experiments
Conclusion

12. QoS Model qi,j represents the jth criterion value of Web service i.
Normalization: Q Q*
QoS utilities computation:
W is a weight vector used to represent user’s priorities on preferences given to different criteria.

13. Similarity search WSDL preprocessing Similarity computation
Word stemming
stopwords
Similarity computation
A user’s query:
A Web service operation:
Cosine similarity Si

14. Combination
calculates the appropriateness value of a candidate in position p for a query.
defines how much the functionality factor is more important than the non-functionality factor in the final recommendation

15. Three searching styles
Keyword specified
Operation specified
Similar operations

16. Outline Introduction System Architecture QoS-aware searching approach
Experiments
Conclusion

17. Evaluate QoS recommendation
To evaluate the qualities of top-k returned results in a result list, we employ DCG:
WSDL of 3,738 Web services (15,811 operations)
339 distributed computers in 30 countries from Planet-lab are employed to monitor these Web services.
20 test queries in 5 different domains

18. Functional matching evaluation
Benchmark:
OWLS-TC v2
1000 Web services, 32 test queries.

19. Outline Introduction System Architecture QoS-aware searching approach
Experiments
Conclusion

20. Conclusion A novel Web service search engine WSExpress.
Three searching styles in the WSExpress.
A large-scale real-world experiment in distributed environment and a experiment on benchmark OWLS-TC v2 are conducted to study the performance of our search engine prototype.
We publicly release our large-scale real-world Web service WSDL files and associated QoS datasets for future research.

21. Q & A