Computational Linguistic Analysis of Engineered Chatbot Prompts
The use of chatbots has become increasingly popular in the field of education. Chatbots have proven to be an effective tool for providing personalized learning experiences to students. However, the quality of the prompts used by chatbots can greatly influence the effectiveness of the learning experience. Therefore, it is important to analyze and understand the linguistic features of effective chatbot prompts for education. In this paper, we present a computational linguistic analysis of chatbot prompts used for education. We extracted publicly available chatbot prompts, filtered and extracted only those used for education, and applied NLP indices to analyze the linguistic features. The analysis resulted in four factors: linguistic average collostructural strength, collostructural ratio diversity, specificity, and academic language use. These factors all moved in the same direction, indicating that effective prompts had high scores on all four factors. Using this This talk provides insights into the linguistic features of effective chatbot prompts for education and can aid in the development of more effective chatbot-based learning systems.
Michelle Banawan
Asst. Professor and Academic Program Director at Asian Institute of Management
Michelle Banawan, Ph.D. is an Asst. Professor at the Asian Institute of Management. She conducted her postdoctoral research at the Science of Learning and Educational Technology (SOLET) laboratory headed by Dr. Danielle McNamara of the Department of Psychology, Arizona State University from 2019 to 2022. Her specializations include Natural Language Processing (NLP) and Machine Learning. Her postdoc work is focused on the application of machine learning algorithms, computational linguistics and natural language processing to various components of the SOLET lab projects within the contexts of reading comprehension and writing proficiency studies. She also handles data curation, data wrangling and database management which entails writing queries to address the different data requirements of projects. She also participates in the design of intelligent tutoring systems and educational games. Her specialization and skills include data and database management, design, analysis, and visualization. The data science and machine learning tools that she is proficient on include MySQL, Python, and R. She has also more than ten years of experience in information systems analysis and design, business requirements modeling, human computer interaction, user-centered design, and systems documentation.
