Actionable knowledge discovery plays a critical role in deliberation processes as it is able to transform them into more effective practices, which enhance decision-making and facilitate consensus-building. This chapter presents an innovative software toolkit, which leverages AI techniques to meaningfully address the corresponding issues. The proposed solution is based on an integration of advanced deep learning and natural language processing techniques that articulate the structure and content of a deliberation to develop a semantically-rich pool of actionable knowledge. The toolkit comprises four main components, namely: (i) a pre-processsing component, which cleans the participants’ feedback and incorporates a polarity detection mechanism, for assigning each feedback into a sentiment group (ii) a clustering component, which groups the participants’ feedback according to its content and stance to the issue under discussion, (iii) a summarization component, which provides an informative summary of the content of each generated cluster and (iv) a keyphrase extraction component, which identifies the most influential phrases from the feedback of each cluster. Aiming to improve decision and policy making in public deliberation processes, our approach adopts a joint discourse-level and information-seeking perspective, while paying much attention to data aggregation mechanisms for the context under consideration.