Analyzing and Inferring Distance Metrics on the Particle Competition and Cooperation Algorithm
| Autor: | Fabricio Breve, Lucas Guerreiro |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
business.industry
Computer science 02 engineering and technology Semi-supervised learning Machine learning computer.software_genre Graph Euclidean distance 020204 information systems 0202 electrical engineering electronic engineering information engineering Artificial intelligence business Algorithm computer |
| Zdroj: | Computational Science and Its Applications – ICCSA 2017 ISBN: 9783319624068 ICCSA (6) |
| Popis: | Machine Learning is an increasing area over the last few years and it is one of the highlights in Artificial Intelligence area. Nowadays, one of the most studied areas is Semi-supervised learning, mainly due to its characteristic of lower cost in labeling sample data. The most active category in this subarea is that of graph-based models. The Particle Competition and Cooperation in Networks algorithm is one of the techniques in this field, which has always used the Euclidean distance to measure the similarity between data and to build the graph. This project aims to implement the algorithm and apply other distance metrics in it, over different datasets. Thus, the results on these metrics are compared to analyze if there is such a metric that produces better results, or if different datasets require a different metric in order to obtain a better correct classification rate. We also expand this gained knowledge, proposing how to identify the best metric for the algorithm based on its initial graph structure, with no need to run the algorithm for each metric we want to evaluate. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|