[WEEK 6 — Wi-Fi Based Indoor Positioning]
Team Members: Burak Emre Ozer, Huzeyfe Kocabas
This week we are going to make a classification by using the attributes which are buildingID and floor and also you can watch the video presentation of our project from the link below.
The accuracy of floor and building classification is important in the multi-building and multi-floor databases. Our goal is to evaluate the success of the classification algorithms for indoor positioning.
buildingID: ID to identify the building. Measures were taken in three different buildings. Categorical integer values from 0 to 2.
floor: Altitude in floors inside the building. Integer values from 0 to 4.
The first step of our task is building classification as mentioned at the introduction of the post. It is clear from this plot that RF is better than other classifiers.
In the second step, the classification was performed based on the floor attribute in each building.
The machine learning algorithms are compared in terms of accuracy. As a conclusion, we deduced that the algorithms had almost the success rate close to each other but SVM for building classification and RF for floor classification are superior to all other methods to classification.
