This post is provided by guest blogger, Angelica Brodeur, graduate student University of St. Francis, MS Training and Development program

When planning e-learning projects, one of the most important aspects to consider is determining how to support learners in the e-learning environment. Learners will always need assistance navigating the e-learning system and depending on the ratio of learners to staff, there might not be enough assistance available to answer all questions in an efficient or effective time frame. Thus, it is recommended to incorporate a dynamic Q&A system to help address frequently asked questions (FAQ) which will then save staff the time and energy needed to address other learner concerns. This is where chatbots are recommended. Chatbots can be a great addition to an e-learning system but concerns of accuracy need to be addressed first to ensure that chatbots are viewed as a beneficial resource for learners instead of another challenge that causes learners to turn to staff for assistance, which would defeat the entire purpose of utilizing chatbots in the first place.

Author Sumikawa et al. (2020) provides examples of the advantages and disadvantages of utilizing chatbots to address FAQ, allowing readers to determine how to best utilize chatbots in their own e-learning system by creating a framework to address the discussed disadvantages. To best utilize chatbots, organizations need to support the creation of datasets to best improve the accuracy of the information provided by chatbots. Sumikawa et al. (2020) created a framework that provides two recommended algorithms to best increase accuracy: 1) creating new questions and 2) aggregating semantically similar answers. Thus, this could be a great solution to supporting learners in the e-learning environment for anyone interested.

Reference

Sumikawa Y., Fujiyoshi M., Hatakeyama H., Nagai M. (2020) Supporting creation of FAQ dataset for e-learning chatbot. Intelligent Decision Technologies 2019. Smart Innovation, Systems and Technologies, 142,3-13. https://doi.org/10.1007/978-981-13-8311-3_1