Deep knowledge-aware framework for web service recommendation

التفاصيل البيبلوغرافية
العنوان: Deep knowledge-aware framework for web service recommendation
المؤلفون: Xingjian Wang, Zixian Guo, Haochen Li, Rongen Yan, Depeng Dang, Chuangxia Chen
المصدر: The Journal of Supercomputing. 77:14280-14304
بيانات النشر: Springer Science and Business Media LLC, 2021.
سنة النشر: 2021
مصطلحات موضوعية: Information retrieval, Knowledge representation and reasoning, Artificial neural network, Computer science, business.industry, Cloud computing, Recommender system, computer.software_genre, Information overload, Theoretical Computer Science, Task (project management), Hardware and Architecture, Feature (machine learning), Web service, business, computer, Software, Information Systems
الوصف: Web services are products in the era of service-oriented computing and cloud computing. Considering the information overload problem arising from the task of selecting web services, a recommendation system is by far the most effective solution for performing such selections. However, users calling a limited number of services will cause severe data sparseness and a weak correlation with services. In addition, fully mining the semantic features and knowledge features of the text description is also a major problem that needs to be solved urgently. This paper proposes a deep knowledge-aware approach which introduces knowledge graph and knowledge representation into web service recommendation for the first time. We solve the data sparse problem and optimize the user’s feature representation. In this approach, an attention module is introduced to model the impact of tags for the candidate services on different words of user queries, and a deep neural network is used to model the high-level features of user-service invocation behaviors. The results of experiments demonstrate that the proposed approach can achieve better recommendation performance than other state-of-the-art methods.
تدمد: 1573-0484
0920-8542
الوصول الحر: https://explore.openaire.eu/search/publication?articleId=doi_________::3f7c981c79dfb1947b5e6f3cada9c0bbTest
https://doi.org/10.1007/s11227-021-03832-2Test
حقوق: CLOSED
رقم الانضمام: edsair.doi...........3f7c981c79dfb1947b5e6f3cada9c0bb
قاعدة البيانات: OpenAIRE