Data Mining
Model evaluation on a 75%-25% training-test split using f1 score, precision and recall, parameter tuning, SVM training and the use of neural networks.
Client University of Patras, Greece
01Overview
About the project
Training-test data set with 75%-25% ratio and model performance measurement using f1 score, precision and recall metrics. Improve results by experimenting with input parameters.
Training SVM with the best parameters and listing findings on how much the classification quality was affected.
Use of neural networks.
02Tech stack
Technologies
Related
More in AI
-
BI in a Box
End-to-end Business Intelligence platform that turns raw data from ERP systems and databases into insights via dashboards, reporting and time-series forecasting.
-
Multi-dimensional Data Structures
2D kNN query implementation using K-D Trees, Quad Trees and 2D Range Trees over a dataset of check-in events with x/y coordinates, accuracy, timestamps and place ids.
Keep exploring
More projects
Interested in something similar?
Tell me about your idea or project and let's see how we can build it together.