WSC2RCNN: A Deep Learning Actions-based Classifier for Improved Web Service Discovery

Abstract

Due to the increasing popularity of Web services and their tremendous number, discovery task is shown as the most important and difficult step. The difficulty lies in the limited data provided on them which is, in the most cases, a short textual description. Based on these descriptions, many interesting works have been proposed trying to classify them in an efficient way. In this work, we propose a new approach based on deep learning WSC2RCNN that uses Web services descriptions and actions within these descriptions to improve the classification task, which has given promising results and outperforms state-of-the-art approaches using the same data.