Alana v2: Entertaining and Informative Open-domain Social Dialogue using Ontologies and Entity Linking Amanda Cercas Curry, Ioannis Papaioannou, Alessandro Suglia, Shubham Agarwal, Igor Shalyminov, Xinnuo Xu Ondˇ rej Dušek, Arash Eshghi, Ioannis Konstas, Verena Rieser and Oliver Lemon The Interaction Lab, Department of Computer Science Heriot-Watt University, Edinburgh, EH14 4AS, UK {ac293, i.papaioannou, as247, sa201, is33, xx6}@hw.ac.uk {o.dusek, a.eshghi, i.konstas, v.t.rieser, o.lemon}@hw.ac.uk Abstract We describe our 2018 Alexa prize system (called ‘Alana’) which consists of an ensemble of bots, combining rule-based and machine learning systems. This paper reports on the version of the system developed and evaluated in the semi- finals of the 2018 competition (i.e. up to 15 August 2018), but not on subsequent enhancements. The main advances over our 2017 Alana system are: (1) a deeper Natural Language Understanding (NLU) pipeline; (2) the use of topic ontologies and Named Entity Linking to enable the user to navigate and search through a web of related information; rendering Alana in part an interactive NL interface to linked information on the web; (3) system generated clarification questions to interactively disambiguate between Named Entities as part of NLU; (4) a new profanity & abuse detection model with rule-based mitigation strategies; and (5) response retrieval from Reddit. We also present several ablation studies that measure the performance contributions of specific features (e.g. use of Ontology-bot, Reddit-bot, rule-based systems, etc). We find that these features increase overall system performance. Our final score, namely averaged user ratings over the whole semi-finals period, was 3.4. We were also able to achieve long dialogues (average around 11 turns and 2.20 minutes) during the semi-finals period. 1 Introduction In this paper, we describe our entry to the 2018 Alexa Prize challenge semi-finals, a socialbot called Alana. Our system is based on an ensemble of different task/topic-specific bots, combining rule-based and machine learning systems. We focus mostly on the improvements we made over our last year’s entry (Papaioannou et al., 2017a,b). Our overall vision was to create an informative and engaging social chatbot that aims to keep users interested and enjoying a spoken interaction on topics of their choice for as long as possible. Our overarching inspiration for this vision is a mixture of topic-related chat, finding out about the user, and sharing amusing facts, jokes, stories, and items of news. Given that the system has better than human access to information on the web such as Wikipedia and news articles, our system also provides an engaging and interactive way to explore the web/news according to user preferences. As such, our social chatbot was designed to have the following behaviour: 1. It should be able to engage in open-domain topic-based conversations, to minimise re- sponses such as “I don’t know what you mean” or “I can’t answer that”. 1st Proceedings of Alexa Prize (Alexa Prize 2018).