A Novel Approach for SemiAutomatic Tech-Mining and Interactive Taxonomy Visualization

  • Ioannis Karakatsanis

Student thesis: Master's Thesis

Abstract

Detecting emerging technologies is critical to decision makers and stakeholders such as investors in order to follow the latest developments as well as the potential for development of the relevant fields. For this reason, a fully automatic taxonomy-based framework has already been proposed in previous studies for identifying emerging technologies and for visualizing their interrelationships. Although this framework yielded encouraging results, the fully automatic process proved a little too noisy. Thus, in this study, a novel approach for semi-automatic technology forecasting and interactive taxonomy visualization is presented. Main goal of the current work is to overcome the limitations of the fully automatic process, namely the inconsistencies occurring in the relationships between the keywords of the taxonomy by developing an easy to use platform that utilizes a semi-automatic approach and can be used by non-technical end-users interested in tech-mining. To achieve this, a unified web-based application was implemented which compiles all the analytical techniques used in previous studies and offers interactive visualization. The findings of this methodology are evaluated by domain experts who confirm that our method is capable of generating informative and accurate visualizations of technologies in their early growth phase. In addition, to demonstrate the applicability of the developed approach in a particular research domain, a case study is presented in the field of Renewable Energy (RE) and specifically in the Distributed Networks (DNs) subfield.
Date of AwardMay 2015
Original languageAmerican English
SupervisorWei Lee Woon (Supervisor)

Keywords

  • Emerging Technologies
  • Fully Automatic Taxonomy
  • Semi-Automatic Technology Forecasting
  • Interactive Taxonomy Visualization.

Cite this

'