SwipeGuide is an innovative Startup company offers a Cloud platform for creating, sharing and using visual step-by-step instructions. The company recently won the Young Technology Award and has offices in Amsterdam & Enschede (team of 16 people). The SwipeGuide platform is based upon evidence-based instruction principles. It brings user & employee instruction to mobile devices like the smartphone and tablet. SwipeGuide offers a new user experience for instruction using the intelligent features of devices to make instructions more relevant and effective. Customers like Philips, Heineken, Eneco and Pepsico use the platform to instruct employees and/or customers.
SwipeGuide wants to enable everyone to create effective instructions. For this aim they want to utilize the data our platform collects on user behavior and user sentiment. The aim is to apply Machine Learning to learn from the effectiveness of the instructions they have and apply that new knowledge when guides are newly created. For designing this piece of intelligence for our platform they need to make sense of the data they capture. The main research questions:
§ What data should SwipeGuide capture in order for effectiveness of instructions to be measured?
§ What pattern recognition vectors do they need to apply on the content as a training set to turn effective instructions into recommendation when creating new content?
The date SwipeGuide currently collect are both anonymous behavioral data and sentiment data from their users. Behavioral data is for example time spent on instruction steps, user flows through instructions, and usage of pieces of information (e.g. warning, tips). Sentiment data contains qualitative user feedback, customer effort score, and net promotor score. SwipeGuide thinks of combining data elements to draw conclusions for intelligent editor advice. The research should result into feature designs that can be incorporated into the SwipeGuide platform.
SwipeGuide wants to work from customer use cases in the project to make the solution as concrete as possible and have realistic playgrounds for testing.
This could be a challenging assignment to fulfill your bachelor or master thesis.
For further information:
Science Shop Twente, Ir. Tim Jongman, The Gallery, tel. 053-489 3942, e-mail: firstname.lastname@example.org