Panesar, Kulvinder ORCID: https://orcid.org/0000-0002-4523-7218 (2020) Natural language processing (NLP) in Artificial Intelligence (AI): a functional linguistic perspective. In: Gouveia, Steven S., (ed.) The Age of Artificial Intelligence: An Exploration. Vernon Press Downloaded from: http://ray.yorksj.ac.uk/id/eprint/3406/ The version presented here may differ from the published version or version of record. If you intend to cite from the work you are advised to consult the publisher's version: https://vernonpress.com/book/935 Research at York St John (RaY) is an institutional repository. It supports the principles of open access by making the research outputs of the University available in digital form. Copyright of the items stored in RaY reside with the authors and/or other copyright owners. Users may access full text items free of charge, and may download a copy for private study or non-commercial research. For further reuse terms, see licence terms governing individual outputs. Institutional Repository Policy Statement RaY Research at the University of York St John For more information please contact RaY at ray@yorksj.ac.uk Natural language processing (NLP) in Artificial Intelligence (AI): a functional linguistic perspective Dr Kulvinder Panesar York St John University This chapter encapsulates the multi-disciplinary nature that facilitates NLP in AI. It identifies work of a linguistically orientated conversational software agent (CSA) (Panesar, 2017) framework sensitive to natural language processing (NLP) concepts, language and the agent environment. A long-standing issue within NLP CSA systems is refining the accuracy of interpretation to provide realistic dialogue to support the human-tocomputer communication. Motivated by this, we present novel computational approach of using Role and Reference Grammar (RRG) – a strong, mature, functional linguistic theory as the linguistic engine of a perceiving utterances (via speech act performatives), for further integration, cognitive manipulation and planning to create a grammatical correct RRG based response. Historically, the CSA’s links to the question ‘can machines think’ (Turing, 1950) making use of dialogue testing to assert intelligence. This question inspired much research and competitions with the yearly Loebner competition to demonstrate human-like conversation ("loebner.net", 2015) with a three times winner Chatbot Mitzuki (Worswick, 2017). We make a distinction between an applied Chatbot and CSA, in that the CSA has a deep strategic role to hold a conversation and enable the mechanisms to focus the conversation on achieving a goal, via NL dialogue (O'Shea et al., 2010). There is a need to plan, and to decide what do next, and manage the conversation - this is the work of the dialogue manager (DM) (Treumuth, 2011). The CSA’s role is that of a linguistic aware knowledge aware process simulating an empowered human to take part in the conversation and ask questions. So how is this goal achieved? This is ‘intentionality’, in that the agent displays beliefs, desires and intentions (BDI) concerning objects, events and states of affairs in the real world (Searle, 1983). Further, taking the viewpoint of language as action, we view utterances that change the state of the world, and hence speakers and hearer’s mental state change as a result of these utterances (Cohen and Levesque, 1988). To achieve this communication and interaction the plan based method of DM using the BDI model architecture is deployed, and is very flexible and supports a greater complexity of conversation (Kluwer, 2011). This CSA investigates the integration, intersection and interface of the language, knowledge, and speech act constructions (SAC) based on a grammatical object (Nolan, 2014a), and the sub-model of BDI (Rao and Georgeff, 1995) and DM for NLP. After deep requirements analysis and considering the works of (Nolan, 2014a, Nolan, 2014b) in regards to DM, a conceptual architecture of the CSA framework is devised, We present an investigation into the intersection and interface between our linguistic and knowledge (belief base) models for both dialogue management and planning. The architectural approach constitutes three phase models: (1) a linguistic model based on RRG; (2) Agent Cognitive Model (ACM) with two inner models: (a) knowledge representation model employing conceptual graphs (CGs) serialised to Resource Description Framework (RDF) (Chein et al., 2013); (b) a planning model underpinned by BDI concepts (Wooldridge, 2013) and intentionality (Searle, 1983) and rational interaction (Cohen and Levesque, 1990); and (3) a dialogue model employing common ground (Stalnaker, 2002). The CSA framework is mapped to an operational framework, and implemented as a Java prototype, developed in Eclipse IDE, predominantly as POJO (plain old java objects) with some API support, based on a food and cooking domain (ontology) due to its rich and representative (ranging lexical entries) nature. The validation and verification (VV) approach deployed is based on the phase models and thus is a multiapproach. It is driven by: (1) grammatical testing (English language utterances) and NLP pipeline tasks; (2) software engineering (UML modelling, architecture centric, data structures and algorithms); knowledge representation logics (first order logics and graph theory) and agent practice (message passing, and planning cognitive responses). The author deems that RRG is as a successful linguistic engine for the CSA, but identify the complexity of the semantic gap of internal representations with subsequent details of a conceptual bridging solution. References "LOEBNER.NET". 2015. Home Page of The Loebner Prize in Artificial Intelligence - "The First Turing Test" [Online]. Available: http://loebner.net/Prizef/loebner-prize.html [Accessed 12th March 2017]. CHEIN, M., MUGNIER, M.-L. & CROITORU, M. 2013. Visual reasoning with graph-based mechanisms: the good, the better and the best. The knowledge engineering review, 28, 249-271. COHEN, P. R. & LEVESQUE, H. J. 1988. Rational interaction as the basis for communication. DTIC Document. COHEN, P. R. & LEVESQUE, H. J. 1990. Intention is choice with commitment. Artificial Intelligence, 42, 213-261. KLUWER, T. 2011. From Chatbots to Dialog Systems. In: PEREZ-MARIN, D. & PASCUAL-NIETO, I. (eds.) Conversational Agents and Natural Language Interaction. USA: Information Science Reference. NOLAN, B. 2014a. Constructions as grammatical objects : A case study of prepositional ditransitive construction in Modern Irish. In: NOLAN, B. & DIEDRICHSEN, E. (eds.) Linking Constructions into Functional Linguistics: The role of constructions in grammar. Amsterdam/Philadelphia: John Benjamins Publishing Company. NOLAN, B. 2014b. Extending a lexicalist functional grammar through speech acts, constructions and conversational software agents. In: NOLAN, B. & PERIÑÁN., C. (eds.) Language Processing and Grammars: The role of functionally oriented computational models [Studies in language Companion Series 150]. Amsterdam and New York: John Benjamins Publishing Company. O'SHEA, K., BANDAR, Z. & CROCKETT, K. 2010. A conversational agent framework using semantic analysis. International Journal of Intelligent Computing Research (IJICR), 1. PANESAR, K. 2017. A linguistically centred text-based conversational software agent. Unpublished PhD thesis, Leeds Beckett University. RAO, A. S. & GEORGEFF, M. P. BDI Agents - From Theory to Practice. ICMAS, 1995. 312-319. SEARLE, J. R. 1983. Intentionality: An essay in the philosophy of mind, CUP. STALNAKER, R. 2002. Common ground. Linguistics and philosophy, 25, 701-721. TREUMUTH, M. 2011. A Framework for Asynchronous Dialogue Systems: Concepts, Issues and Design Aspects. PhD thesis, University of Tartu. TURING, A. M. 1950. Computing machinery and intelligence. Mind, 433-460. WOOLDRIDGE, M. 2013. Intelligent Agents. In: WEISS, G. (ed.) Multiagent Systems. 2nd edn. ed. USA: Massachusetts Institute of Technology. WORSWICK, S. 2017. Mitsuku - Loebner Prize 2017 [Online]. https://www.pandorabots.com/mitsuku/. Available: http://www.mitsuku.com/ [Accessed 12th January 2017]. Short biographical note Dr Kulvinder Panesar is a Senior Lecturer of Computer Science at the School of Art, Design and Computer Science at York St Johns University. Her research work focuses on NLP (Natural Language Processing) in AI (Artificial Intelligence), and meaning and knowledge representation in conversational software agents. Teaching commitments include both undergraduate and postgraduate courses, specifically software engineering design and development; management information systems and strategy; enterprise computing solutions – databases, networks and security; and data science. Additional roles include: course directorship for BSc (Hons) in Software Engineering; placement co-coordinator; ‘study abroad’ academic advisor and STEM (Science, Technology, Engineering and Mathematics) ambassadorship. List of similar titles     A linguistically centred conversational software agent Motivating a linguistically centred conversational software agent A functional linguistic approach applied to a conversational software agent Intersecting, interfacing and intersecting between language and knowledge models of a conversational software agent