Argument Theory offers the rAIson platform, a SaaS Artificial Intelligence (AI) platform specializing in the modeling and automation of complex and difficult decision-making problems in a changing environment and in the presence of incomplete information. Our approach is based on computational argumentation, a technology that allows an Artificial Intelligence system to have reasoning and behavioral capacities similar to those of a human.
The rAIson platform offers an argumentation system which is a generic inference engine that is endowed with both argumentative and abductive reasoning capabilities. This allows it to reason even in the presence of incomplete information by hypothetical reasoning. The system is capable of performing contextual reasoning and is able to explain its reasoning in order to justify its decisions. On the other hand, it is able to take into account the feedback of the user and to engage into a dialogue by arguing in favor of its decisions when these are challenged by the user, in order to arrive at a decision which is acceptable by the user.
Our argumentation technology is based on the Gorgias system and its extensive use in several types of real applications since 2004.
rAIson is a multi-purpose platform. It offers personalized solutions for companies facing problems such as automated compliance management, automated data access management, automatic risk assessment and mitigation, cybersecurity, etc. This is done using an authoring tool that allows the AT team to quickly and efficiently encode our clients' decision-making policies taking into account their preferences and other requirements (eg legal compliance). These policies constitute the knowledge of the AI system to be developed and which can be enriched, using machine learning methods, through interaction with the user. The systems thus developed can be installed at customers' premises and utilize the services of our argumentation system via the APIs provided by the platform.
The platform also allows IT specialist developers, who work on behalf of various companies in different fields of activity (finance, medical diagnosis, insurance, law, etc.), to develop in-house their own AI decisional systems based on our technology with the assistance of our AT team. This is made possible thanks to our authoring tool which allows these developers to model the knowledge of their AI system by working directly at the level of the language of their application without the need to have any technical expertise in computational argumentation. This knowledge is then coded by our platform in the language of our argumentation system in an automatic and transparent way for the developers. Developers can then query the system in the context of the application developed using the APIs provided by the platform.
Thanks to our argumentation system, our platform has the capacity to assess arguments for or against a decision and to provide explanations of the reasons and hypotheses in support of a specific decision. This allows the client to take final informed and explainable decisions.
Thanks to our argumentation system, our platform has the capacity to assess arguments for or against a decision and to provide explanations of the reasons and hypotheses in support of a specific decision. This allows the client to take final informed and explainable decisions.
The computational argumentation on which the rAIson platform is based, is an approach particularly suited to proposing solutions within the framework of Explainable, Trustable and Ethical AI.
