How do we get students to think critically, think deeper, make connections and have meaningful…
A chatbot is a computer-controlled conversational agent that interacts with users using natural language in a manner that approximates regular conversation between humans. Chatbots have been employed across a wide range of domains for many different purposes including customer service and technical support, e-commerce, education and training, user assistance, and information access, organisation, and scheduling. Ongoing developments within this space suggest that interaction with technology using either natural language text or speech is becoming increasingly feasible as the technology matures and users become more accustomed to interacting with digital entities. The key metric in determining the value or effectiveness of a chatbot in this regard is the extent to which users can interact with it in a natural fashion; the Loebner prize for example evaluates chatbots according to how well they can fool people into thinking they are talking to a human (in an implementation of the Turing test).
At the forefront of this technology are dedicated digital assistants such as the Apple Siri, Amazon Alexa, Microsoft Cortana, Google Assistant, and Samsung Bixby platforms. This is complemented by a vast number of simpler and more domain-specific chatbots which are typically designed to fulfil a specific purpose such as booking a flight, making a reservation, etc. These types of chatbots typically employ a rules-based approach to respond to user input, which limits their generalisability but allows them to effectively respond to requests within a specific domain, particularly where these are structured in predictable ways in pursuit of a given outcome (e.g. I would like to book a flight to X). Conversely, the more complex and capable digital assistant type chatbots employ machine learning techniques and are trained using large quantities of conversational data, which allows them to respond to a wider variety of user input and requests.
Of particular interest are the educational applications of chatbot technologies which have demonstrated potential in terms of their ability to support and engage student learning. Pereira (2016) for example investigated the application of chatbots for training a group of computer science students using multiple-choice quizzes, with feedback indicating that the chatbot helped students to become more self-guided in terms of their engagement and preparation. Similarly, MOOCBuddy – a Massive Open Online Courses (MOOC) recommender chatbot developed by Holotescu (2016), assisted students to discover suitable courses for personal and professional development and teachers to integrate MOOCs into their courses. Gamified chatbots, such as the educational chatbot CiboPoli developed by Fadhil and Villafiorita (2017), can account for factors such as age and gender during the course of teaching children about healthy lifestyle within an interactive social environment. Social health and well-being applications can also be explored using chatbots; Crutzen and colleagues (2011) used a chatbot to answer adolescents’ questions on sex, drugs, and alcohol, with feedback indicating a more positive end-user experience compared to information lines and search engines.
In this way, suitably capable chatbot systems can offer students a range of benefits based on their ability to respond to their needs in an immediate and natural manner, particularly where interaction occurs within the broader context of an integrated learning environment. Furthermore, chatbots exhibit the capacity to create informal connections with users such that they can be leveraged to support ubiquity and self-directed learning as well as personal and professional development.
Crutzen, R., Peters, G. J. Y., Portugal, S. D., Fisser, E. M., & Grolleman, J. J. (2011). An artificially intelligent chat agent that answers adolescents’ questions related to sex, drugs, and alcohol: an exploratory study. Journal of Adolescent Health, 48(5), 514-519.
Dale, R. (2016). The return of the chatbots. Natural Language Engineering, 22(5), 811-817.
Fadhil, A., & Villafiorita, A. (2017). An adaptive learning with gamification & conversational UIs: The rise of CiboPoliBot. In Adjunct Publication of the 25th Conference on User Modeling, Adaptation and Personalization (pp. 408-412). ACM.
Holotescu, C. (2016). MOOCBuddy: a Chatbot for personalized learning with MOOCs. In A. Iftene J Vanderdonckt (Eds.) In RoCHI–International Conference on Human-Computer Interaction (Vol. 91) (pp. 91-94).
Pereira, J. (2016). Leveraging chatbots to improve self-guided learning through conversational quizzes. In Proceedings of the Fourth International Conference on Technological Ecosystems for Enhancing Multiculturality (pp. 911-918). ACM.
Satu, M. S., & Parvez, M. H. (2015, November). Review of integrated applications with AIML based chatbot. In Computer and Information Engineering (ICCIE), 2015 1st International Conference on (pp. 87-90). IEEE.
Shawar, B. A., & Atwell, E. (2003). Using dialogue corpora to train a chatbot. In Proceedings of the Corpus Linguistics 2003 conference (pp. 681-690).
Shawar, B. A., & Atwell, E. (2007). Different measurements metrics to evaluate a chatbot system. In Proceedings of the Workshop on Bridging the Gap: Academic and Industrial Research in Dialog Technologies (pp. 89-96). Association for Computational Linguistics.