Not logged in.
Quick Search - Contribution
Contribution Details
Type | Master's Thesis |
Scope | Discipline-based scholarship |
Title | Implementation of a Generalized Binary Machine Learning Tool for Classifying Scientic Papers Based on Natural Language |
Organization Unit | |
Authors |
|
Supervisors |
|
Language |
|
Institution | University of Zurich |
Faculty | Faculty of Business, Economics and Informatics |
Date | 2020 |
Abstract Text | Green Zora is a project with the intention of showing the world which scientific papers submitted at the University of Zurich can be classified as being about sustainability. It is composed of a dynamic website, a machine learning tool and a backend that puts everything together. This thesis has two goals: The first goal is to improve the existing web application in various aspects regarding both visual design and usability and to make the tool applicable for various classification tasks beyond sustainability. The second goal is to research different machine learning algorithms and implement the best fit to improve the classification results. In the first part of this thesis, I present the improvements to the existing project and the necessary steps to generalize the tool. The second part gives a short introduction to machine learning and natural language processing. Then I present different machine learning algorithms that I tested during this thesis and their evaluations. Finally, I share how the implementation of the best performing algorithm was done and what steps were taken to optimize the classification results. |
PDF File | Download |
Export | BibTeX |