At the end of the study unit, students will be able to… (between brackets the number of the corresponding intended learning qualification of the programme):
- explain the concepts of deep data in relation to big data (2.1)
- classify different machine learning methods (2.3, 3.2);
- apply predictive modelling using supervised machine learning methods (2.3)
- execute time series analysis and applications of predictive modelling such as autoregressive model (2.4)
- evaluate, interpret, and visualize the results of a predictive model (2.5, 2.6, 5.2).
|
|
As more and more of everyday mundane activities become digitized, big data grows with irrelevant, unusable, or broken information. For big data to become of high quality and actionable from a personal perspective, it requires care -as if it were a lively thing- that grows, changes, becomes broken, and needs repair. As such, big data is being turned into deep data; smaller numbers of dense, rich, and diverse streams of data that are relevant for within-person predictive modelling and personalized improvement strategies. During two weeks of the so-called research incubator of this Research study unit, students are going to use the mundane as a generative site of big data and retrieve information-rich streams of deep data. For this study unit students will collect data about themselves from at least two separate mundane sites that generate streams of big data. This can be either their own data that is already generated or by using an app to generate new data. The aim is to transfer broken big data into coherent deep data by combining separate data streams with the ultimate goal to predict future behaviours: If we combine streams of deep data, can we make sense of the data, use it to predict personal trends, and extract transferable insights? Students write an individual research report (8R1), using their own personal streams of deep data collected by 8 hours of weekly activities based on at least two sources.
This study unit is part of the Communication science module The Quantified Self. Because the four study units, which are part of the module, are highly related to each other it is not possible to follow this study unit separately.
|
 |
|