Intent-Aware Chatbot for the Refinement of Protection Recommender System

Currently, there are a number of on-demand network services (e.g., firewalls, load balancers, DHCP servers) and marketplaces available, which are not only offering network services, but also offer alternatives regarding the deployment and management aspects of such services [1]. However, it is not a trivial task for end-users to select one of them. Decision-making is even more critical when infrastructure is under attack and the decision to mitigate the attack should be provided on the basis of information about the infrastructure, such as its economic aspects, demands, and the characteristics of the attack. In this sense, recommender systems [2] provide, for example, a valuable security management tool to support decision during the detection and mitigation process. Thus, in order to help during the recommendation of network services, more high-level interfaces should be available in order to simplify the process of request for adequate services. In such a direction, Intent-based mechanisms [3] can be helpful to translate natural language in requests for recommender systems in order to find the best solution to supply the demands specified by end-users.

The goal of this thesis is to (i) research common languages to describe services and adapt/define standards to describe network services, (ii) definition of an intent-based language to describe the customer demands, and (iii) develop a chatbot interface to receive customers demands (i.e., intents) and transform in a data structure able to be used as input to filter network services from different providers that can support such demands.

References

[1] L. Bondan, M. F. Franco, L. Marcuzzo, G. Venancio, R. L. Santos, R. J. Pfitscher, E. J. Scheid, B. Stiller, F. De Turck, E. P. Duarte, A. E. Schaeffer-Filho, C. R. P. Santos, and L. Z. Granville: FENDE: Marketplace-Based Distribution, Execution, and Life Cycle Management of VNFs; IEEE Communications Magazine, vol. 57, January 2019, pp 1389-1406.

[2] M. Liphoto, C. Du, S. Ngwira: A survey on recommender systems; International Conference on Advances in Computing and Communication Engineering (ICACCE 2016), Durban, South Africa, November 2016, pp. 276-280.

[3] NewgenApps: Intent vs Flow Based Chatbot Communication; [Online]  https://www.newgenapps.com/blog/intent-vs-flow-based-chatbot-communication last visit July 2019.