Discovering Latent Knowledge about Individuals, Enterprises and Technologies

Gong, Xian

Discovering Latent Knowledge about Individuals, Enterprises and Technologies

Gong, Xian

Permalink

Publication Type:: Thesis
Issue Date:: 2026

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (19.47 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Gong, Xian
dc.date.accessioned	2026-08-04T01:33:04Z
dc.date.available	2026-08-04T01:33:04Z
dc.date.issued	2026
dc.identifier.uri	http://hdl.handle.net/10453/195906
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	Innovation, productivity, and long-term economic resilience depend on the effective interaction of individuals, enterprises, and technologies. Yet, despite their central role in shaping economic and societal progress, we lack systematic, scalable ways to detect and quantify how these domains are connected and how their interactions drive performance. This gap persists because such relationships are complex, multidimensional, and embedded within vast, unstructured, and heterogeneous datasets. To tackle these challenges, this research leverages machine learning algorithms, natural language processing techniques, and large language models to discover and map latent knowledge in these domains. The objectives of this thesis are to develop novel computational methods to (i) quantify the influence of individual-level characteristics, such as personality traits, on organisational outcomes, (ii) reveal enterprise-level online network structures and their economic implications, (iii) construct multidimensional representations of the global technology frontier, and (iv) link emerging technologies with leading enterprises. Chapter 1 outlines the motivation, background, and research questions that frame the work. Chapter 2 examines how founders’ personality traits, derived from large-scale linguistic analysis, impact startup success, illustrating how NLP and machine learning can extract latent individual-level attributes associated with entrepreneurial outcomes. This study highlights the role of individual psychological factors in shaping the trajectories of start-ups and their market performance. Chapter 3 introduces the first systematic application of harmonic centrality at the web-domain level, combining structural web-graph metrics with firm-level non-structural data to uncover latent relationships between enterprises’ online positions, economic diversity, and performance. This work demonstrates the potential of network science to reveal hidden drivers of enterprise competitiveness in the digital economy. Chapter 4 presents Cosmos 1.0, a multidimensional map of the emerging technology frontier constructed from Wikipedia entity embeddings and temporal dynamics to uncover structural patterns, temporal trends, and thematic clusters. This map provides a valuable resource for policymakers, investors, and researchers to monitor, anticipate, and respond to technological change. Chapter 5 develops an NLP-based methodology that uses Wikipedia-derived entity embeddings to link 100 emerging technologies with the world’s leading R&D spending companies. This contribution provides a dynamic and scalable framework for connecting corporate innovation strategies with the evolving technology landscape. Overall, this thesis advances computational social science by developing integrated methods to reveal hidden dynamics among individuals, enterprises, and technologies, informing innovation management, economic policy, and strategic decision-making.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/195906/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2026 Xian Gong
dc.rights	au.edu.uts.lib/cph
dc.title	Discovering Latent Knowledge about Individuals, Enterprises and Technologies	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

Innovation, productivity, and long-term economic resilience depend on the effective interaction of individuals, enterprises, and technologies. Yet, despite their central role in shaping economic and societal progress, we lack systematic, scalable ways to detect and quantify how these domains are connected and how their interactions drive performance. This gap persists because such relationships are complex, multidimensional, and embedded within vast, unstructured, and heterogeneous datasets. To tackle these challenges, this research leverages machine learning algorithms, natural language processing techniques, and large language models to discover and map latent knowledge in these domains. The objectives of this thesis are to develop novel computational methods to (i) quantify the influence of individual-level characteristics, such as personality traits, on organisational outcomes, (ii) reveal enterprise-level online network structures and their economic implications, (iii) construct multidimensional representations of the global technology frontier, and (iv) link emerging technologies with leading enterprises. Chapter 1 outlines the motivation, background, and research questions that frame the work. Chapter 2 examines how founders’ personality traits, derived from large-scale linguistic analysis, impact startup success, illustrating how NLP and machine learning can extract latent individual-level attributes associated with entrepreneurial outcomes. This study highlights the role of individual psychological factors in shaping the trajectories of start-ups and their market performance. Chapter 3 introduces the first systematic application of harmonic centrality at the web-domain level, combining structural web-graph metrics with firm-level non-structural data to uncover latent relationships between enterprises’ online positions, economic diversity, and performance. This work demonstrates the potential of network science to reveal hidden drivers of enterprise competitiveness in the digital economy. Chapter 4 presents Cosmos 1.0, a multidimensional map of the emerging technology frontier constructed from Wikipedia entity embeddings and temporal dynamics to uncover structural patterns, temporal trends, and thematic clusters. This map provides a valuable resource for policymakers, investors, and researchers to monitor, anticipate, and respond to technological change. Chapter 5 develops an NLP-based methodology that uses Wikipedia-derived entity embeddings to link 100 emerging technologies with the world’s leading R&D spending companies. This contribution provides a dynamic and scalable framework for connecting corporate innovation strategies with the evolving technology landscape. Overall, this thesis advances computational social science by developing integrated methods to reveal hidden dynamics among individuals, enterprises, and technologies, informing innovation management, economic policy, and strategic decision-making.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/195906