Towards a unifying framework for conceptual represention and reasoning in cognitive systems

Abstract

Abstract.In this paper we present the rationale adopted for the integration of the knowledge level of DUAL-PECCS, a cognitive system for conceptual representation and categorization, with two different cognitive architectures: SOAR and LIDA. In previous works we already showed how the representational and reasoning framework adopted in DUAL-PECCS was integrable with diverse cognitive architectures, i.e. ACT-R and CLARION, making different representational assumptions and adopting diverse knowledge processing mechanisms. The additional integrations presented here suggest that the underlying knowledge representation and reasoning structure adopted in DUAL-PECCS can be used as a unifying framework for the knowledge level of agents endowed with different cognitive architectures. The current version of the system has been experimentally assessed in a task of conceptual categorization where a target concept illustrated by a simple common-sense linguistic description had to be identified by resorting to a mix of categorization strategies. The output has then been compared to human and artificial responses. The novel integration allowed us to extend our previous evaluation.

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

“…In addition to these studies, there have been other applications in which sense and concept representations, in their broader meaning, have been effectively integrated: word sense or named entity disambiguation (Chen et al, 2014;Rothe & Schütze, 2015;Fang et al, 2016;Panchenko, Faralli, Ponzetto, & Biemann, 2017a;Peters et al, 2018), knowledge base completion (Bordes et al, 2013) or unification (Delli Bovi, Espinosa-Anke, & Navigli, 2015), common-sense reasoning (Lieto, Radicioni, Rho, & Mensa, 2017), lexical substitution (Cocos, Apidianaki, & Callison-Burch, 2016), hypernym discovery (Espinosa-Anke, Camacho-Collados, Delli Bovi, & Saggion, 2016), lexical entailment 24. It is worth mentioning that, while there is still a lack of sense-annotated multilingual corpora, recent efforts have directly addressed this issue by (semi-)automatically disambiguating large amounts of parallel corpora (Taghipour & Ng, 2015;Otegi, Aranberri, Branco, Hajic, Neale, Osenova, Pereira, Popel, Silva, Simov, & Agirre, 2016;Delli Bovi, Camacho-Collados, Raganato, & Navigli, 2017).…”

Beyond subgoaling: A dynamic knowledge generation framework for creative problem solving in cognitive architectures

“…Basically, when a production in [17], where a typicality-based KB properties is translated into a standard Description Logic knowledge base (corresponding to a chunk-based symbolic representation in SOAR). In particular, the overall approach is compliant with the idea of a goal-directed contextual activation of concepts obtained via a process of knowledge "proxyfication" [26] from the long-term memory to the short term memory of a cognitive agent (already employed in knowledge-based systems like DUAL-PECCS [33], integrated with different cognitive architectures, including SOAR [34,32]). A final element emerging from the described compliance consists in the fact that that the output of the new subgoaling procedure (i.e.…”

COVER: a linguistic resource combining common sense and lexicographic information

“…Conceptual categorization. COVER has been used as a knowledge base employed by a system designed to solve the task of conceptual categorization [40,42,43,44]. The task is defined as follows: given a a simple commonsense linguistic description, the corresponding target concept has to be identified.…”