Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Data analysis optimization using multi-criteria decision analysis and machine learning methods

Michail Marios-Alexandros

Full record

---

Summary

In the Big Data world, where information continuously and rapidly keeps increasing, research interest focuses on the discovery of new and more effective methods for data analysis. On this premise, studies are being conducted and combined from multiple research fields, such as Business Administration (Optimization, Multi-Criteria Aid, etc.), as well as new ways of implementation and collaboration with the development of new classifiers or ensemble classifiers. The purpose of this thesis is to develop a system that supports the application of optimization methods to Big Data Analysis, combining the Multi-Criteria Decision Aid methods TOPSIS, UTASTAR and UTADIS with machine learning methods, such as k-means. First off, various datasets were gathered from open data libraries on the web (UCI repository), followed by the required pre-processing procedures. The methods implemented on the system were applied on these datasets, both individually and in combination, with the purpose of supporting the greater knowledge extraction from data regarding ranking and classification/clustering problems. Lastly, the results of the aforementioned methods were evaluated for greater information gain and future research proposals.

Available Files

Services