Overview

As part of their studies, students at HM Hochschule München University of Applied Sciences have to acquire so-called skills and competences that will be in demand on the labour market both now and in the future (so-called "future skills" or "digital skills").

Problem

Students in the final semesters of a Bachelor's or Master's degree programme in particular often do not know how well their future skills have been developed to date and how they could develop them further.

Solution

For this reason, a team of business administration and industrial engineering students designed a chatbot that analyses a student's level of competence and recommends suitable training courses for each competence level. The developed bot - called "Buddy" - acts in a friendly manner and is happy to help students with the diagnosis and further development of their competences.

Procedure

The challenge ran under the main title "Chatbot". A total of four student teams were commissioned to create chatbots, each for a specific topic area. Based on a user analysis on the topic of "Future Skills", it became clear that students often do not know what "Future Skills" are, what level they have already reached regarding them and how they could improve them. For this reason, the first team was given the task of designing a chatbot to help with this topic.

In three sprints, conversation flows were specified (including a battery of items derived from scientific diagnostic tools for future and digital skills) and a prototype chatbot was developed, tested and integrated into Slack on the basis of AWS Lex. The students were supported by Amazon Web Services (AWS) during the process.

Innovation in action

The chatbot Buddy first asks whether it is already clear what is meant by "future skills". If the user negates, he gives a definition. Next, Buddy asks the user if they would like to start a test to analyse their skill level. If this question is answered in the affirmative, 20 items on different facets of "future skills" follow, whereby the user has to assess himself or herself on a three-point scale for each item. The three possible answer options are displayed as card groups to click on, which simplifies and speeds up the interaction with the chatbot. After completing the questions, Buddy displays the calculated result and asks whether further information and training recommendations are desired. Depending on the result, and thus: competence level, links to trainings on the platforms of LinkedIn Learning and Smart VHB are then displayed. Both services can be used free of charge with an HM account.

Next steps

The students plan to refine the prototype technically in the future.