1. Web APIs recommendation with neural content embedding for mobile multimedia computing.
- Author
-
Xu, Yueshen, Ding, Yunpeng, Jiang, Zhiping, Yin, Yuyu, Hei, Lei, and Zhang, Shaoyuan
- Subjects
- *
MOBILE computing , *MULTIMEDIA systems , *MATRIX decomposition , *MOBILE apps , *APPLICATION program interfaces , *INTERNET users , *PREDICTION models - Abstract
In the development of applications for mobile multimedia computing, it has been a core problem to improve the efficiency of applications development. The use of Web APIs has played an important role in applications development and mashups creation, but it takes a lot of time to find a suitable Web API. In this paper, we propose to employ APIs recommendation to solve this problem. We use the matrix factorization (MF) model to predict the relationship between user and Web APIs. The relationship between user and Web APIs is represented by latent features. But different from MF model, we find that the relationship between Web APIs have an impact on prediction performance. This paper uses the embedding technique and mines the latent relationships between APIs and we develop a novel prediction model that leverages the latent relationships between APIs, which is named as LRM (LRM is short for latent relationships mining). The latent relations are input into the LRM model to produce the relation prediction between APIs and developers. The experimental results show that our method achieves high performance and behaves the best compared to existing baselines. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF