This paper describes the application of machine learning algorithms on user mobility data to identify and understand potentially interesting events. The data for this research was collected from a sample of users consenting to be monitored through our in-house developed smart phone app. A pilot study that includes 227 users that were tracked over a period of 7 years yields fairly positive evaluation results in terms of predictive accuracy of identified events but succeeds in identifying exclusively “well-known” events related to users going to or coming from the office and/or lunch. This shows that machine learning methods can be a suitable choice for identifying events in mobility data but there is still room for improvement.
COBISS.SI-ID: 32861735
This paper presents an approach to predicting the future development of scientific research based on scientific publications from the past two centuries. We have applied machine learning methods on the Microsoft Academic Graph dataset of scientific publications. Our experimental results show that the best performance is obtained for a noticeable increase of the topic frequency in the last 5 years compared to the previous 10 years. In this case, our model achieves precision of 74.3, recall of 71.7 and F1 of 73.0. Some topics that our model identified as promising are: proton proton collisions, higgs boson, quark, hadron, mobile augmented reality, variable quantum, molecular dynamics simulations, hadronic final states, search for dark matter.
COBISS.SI-ID: 32857895