This workshop will focus on overfitting issues, such as architecture cost, design, algorithms and methodologies to solicit original research work with a particular emphasis on the challenges and future trends in cyber security, particularly for web-based services security, using machine learning applications. Machine learning approach has proven to be suitable for the WSS because it can help learn information and behaviour from the online and offline data sources in an automatic routine and reduce the workload of security threat analysis through human experts. In connection to this, emerging approaches, such as reinforcement learning, adaptive learning, deep learning, etc can be used for efficiently detecting any type of cyberthreat to WSS.