A classification machine learning algorithm which works by labelling each data point with a class. The class is assigned according to its nearest neighbors within the feature space, calculated via distance metric. This implementation calculates distance using Euclidean distance.
A quantum machine learning teqnique which structureally works as KNN, classifying the datapoints into classes according to its K - nearest neighbors. Differing from KNN, QKNN computes the distance between these points via quantum machine learning principles. This implementation uses a SWAP test, where each data point is translated into rotational angles and encoded onto a quantum feature map. The fidelity, which is the overlap between the two quantum states, is used as the distance metric.
This dataset documents the characteristics of given geographical locations to calculate their susceptibility to flooding. The susceptibility label divides the data into classes according to the locations' flood susceptibility: No Flood, Low, Moderate, High, and Very High. This is the target variable used in this project. The characteristics evaluated include drainage capacity, Topographic Wetness Index, and flow accumulation. These are the features used in this project.
"Pluvial_Flood_Susceptibility" https://www.kaggle.com/datasets/oladapokayodeabiodun/pluvial-flood-dataset
This dataset is available on Kaggle. It contains 1 target variable and 7 features.
SUSCEP is the target, a multiclass label with 5 categories.
The three features selected for this implementation: Drainage, TWI, and FA.