Machine Learning (ML) has been enjoying an unprecedented surge in applications that solve problems and enable automation in diverse domains. Primarily, this is due to the explosion in the availability of data, significant improvements in ML techniques, and advancement in computing capabilities. Undoubtedly, ML has been applied to various mundane and complex problems arising in network operation and management. There are various surveys on ML for specific areas in networking or for specific network technologies. This survey is original, since it jointly presents the application of diverse ML techniques in various key areas of networking across different network technologies. In this way, readers will benefit from a comprehensive discussion on the different learning paradigms and ML techniques applied to fundamental problems in networking, including traffic prediction, routing and classification, congestion control, resource and fault management, QoS and QoE management, and network security. Furthermore, this survey delineates the limitations, give insights, research challenges and future opportunities to advance ML in networking. Therefore, this is a timely contribution of the implications of ML for networking, that is pushing the barriers of autonomic network operation and management.

No abstract

Autonomic computing covers few self-abilities like self-configuration, self-healing, self-optimization, self-protection, self-adaptability, self-awareness, self-openness etc. in software systems. These self-abilities will lead towards lowering the overall maintenance cost of the software because of minimum level of human intervention. The term Autonomicity refers to the level of autonomic (self) features implemented in the system. The International software quality standard ISO 9126 is now replaced by new software product quality standard ISO/IEC 25010:2011 which defines the framework/model to specify and evaluate the quality of software as a product. However, this does not take into account the self-* features (autonomic aspects) and trust factor of modern day software systems. The present paper proposes here that autonomic characteristics of any system must be considered while assessing the quality of any software product. This autonomic-oriented quality model may be used to assess the software quality in a number of domains. Therefore, a new enhanced software quality model is proposed which considers autonomicity and trustworthiness as a factor of quality.

Bot detection using machine learning (ML), with network flow-level features, has been extensively studied in the literature. However, existing flow-based approaches typically incur a high computational overhead and do not completely capture the network communication patterns, which can expose additional aspects of malicious hosts. Recently, bot detection systems which leverage communication graph analysis using ML have gained traction to overcome these limitations. A graph-based approach is rather intuitive, as graphs are true representations of network communications. In this thesis, we propose BotChase, a twophased graph-based bot detection system that leverages both unsupervised and supervised ML. The first phase prunes presumable benign hosts, while the second phase achieves bot detection with high precision. Our prototype implementation of BotChase detects multiple types of bots and exhibits robustness to zero-day attacks. It also accommodates different network topologies and is suitable for large-scale data. Compared to the stateof-the-art, BotChase outperforms an end-to-end system that employs flow-based features and performs particularly well in an online setting.

No abstract