Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Step Prediction
This master thesis aims at writing a smartphone application that can predict how many steps a smartphone user will do on the next day.
Keywords: Big data, Artificial Intelligence, Time series prediction, HMM, ANN
The purpose of this study is to predict if someone will achieve a certain step goal on the next day. If this person is unlikely to achieve this given goal, then we will send the smartphone user a notification with a motivational message to promote physical activity.
We already have a dataset consisting of >230’000 data points of historical steps. This dataset can be used as a starting point for the prediction algorithm. However, in the real application the user specific dataset will consist of only 14 days initially, before it then will grow continuously every day.
The purpose of this study is to predict if someone will achieve a certain step goal on the next day. If this person is unlikely to achieve this given goal, then we will send the smartphone user a notification with a motivational message to promote physical activity. We already have a dataset consisting of >230’000 data points of historical steps. This dataset can be used as a starting point for the prediction algorithm. However, in the real application the user specific dataset will consist of only 14 days initially, before it then will grow continuously every day.
The goal of this master thesis is to write a machine learning application that can predict how many steps a smartphone user will do on the next day.
You should bring:
- Experience in programming
- Skills or the interest to learn the basics of machine learning
- Interest in human behavior prediction based on smartphone data
- ETH / UZH students are preferred
The goal of this master thesis is to write a machine learning application that can predict how many steps a smartphone user will do on the next day.
You should bring: - Experience in programming - Skills or the interest to learn the basics of machine learning - Interest in human behavior prediction based on smartphone data - ETH / UZH students are preferred
Please send your application with a short motivation letter (~250 words), transcript of records and CV to Florian Künzler (fkuenzler@ethz.ch).
If you have any further questions or comments, please don't hesitate to contact me.
Please send your application with a short motivation letter (~250 words), transcript of records and CV to Florian Künzler (fkuenzler@ethz.ch). If you have any further questions or comments, please don't hesitate to contact me.