https://scholarworks.unist.ac.kr/handle/201301/114 Wed, 13 May 2026 12:45:45 GMT 2026-05-13T12:45:45Z https://scholarworks.unist.ac.kr/handle/201301/91098 Title: Development of text analytics methods for supporting intellectual property assessment Author(s): Jeon, Daeseong Abstract: In modern economies where knowledge and technology are central sources of growth, intellectual property (IP) serves as a core legal framework for defining and protecting intangible assets. Among various forms of IP, patents and trademarks are two major instruments that govern the treatment of technological inventions and commercial identifiers. As the number and complexity of patent and trademark filings have grown, the need for systematic approaches to assessing such IP assets has become a critical concern for firms, public institutions, and policymakers. This dissertation frames these issues under the notion of IP assessment, defined as the systematic examination and evaluation of IP assets for legal, strategic, and policy purposes, and develops text-based analytical methods to support three key stakeholder groups: evaluators, examiners, and applicants. The first study addresses the needs of patent evaluators who seek to measure the technological novelty of inventions in a more fine-grained manner than is possible with conventional indicators based on classifications or citations in prior studies. It adopts a patent text analysis in which patent claim texts are converted into numerical vectors in a continuous space using a doc2vec model and applies the local outlier factor (LOF) technique to compute novelty scores that reflect how atypical each patent is relative to its local neighbors. An empirical analysis of 1,877 medical imaging patents filed at the United States Patent and Trademark Office between 1984 and 2019 shows that the resulting novelty scores are significantly associated with established patent indicators and that patents identified as highly novel exhibit higher technological impact on average. The second study focuses on the needs of trademark examiners, who must assess the similarity between the goods and services designated in new applications and those in existing trademarks. Using refined product descriptions provided by the Korean Intellectual Property Office (KIPO), it develops TM-BERT, a domain-adaptive Sentence-BERT model trained on 41,587 goods and 15,344 services. Goods and services are embedded into a low-dimensional semantic space using the trained model, and similarity scores are computed based on cosine similarity of the resulting vectors. Empirical analyses using product descriptions from KIPO confirm that TM-BERT effectively differentiates goods and services within and across similar group codes, demonstrating strong alignment with the European Union Intellectual Property Office CF Similarity database. The proposed model enhances accuracy and consistency in similarity assessments, serving as a complementary tool for trademark examiners in evaluating the similarity of goods and services. The third study turns to the context of trademark applicants, who must decide whether to file trademark applications under refusal uncertainty and resource constraints. It proposes a prescreening approach that learns from historical decisions on refused and registered trademarks to evaluate the distinctiveness of candidate trademarks under the Korean Trademark Act. Trademark examination records from the Korean Intellectual Property Right Information Service are organized into a balanced question–answer dataset aligned with eight distinctiveness-related refusal categories derived from Articles 33 and 34, and the generative language model, Gemma-3, is adapted through parameter- efficient fine-tuning using Low-Rank Adaptation (LoRA). The fine-tuned model produces, for each category, a binary judgment label and legally grounded explanation based on the given trademark and its designated goods or services. Validation on a test set shows that the model achieves moderate accuracy with relatively high precision in identifying refusal cases, and an LLM-as-judge assessment indicates that its refusal explanations attain a nontrivial level of legal and linguistic quality, while remaining below examiner-written explanations. Taken together, the three studies seek to advance text-based methods for IP assessment by addressing stakeholder-specific problems for patent evaluators, trademark examiners, and trademark applicants. From a theoretical perspective, the dissertation extends research on patent novelty, goods and services similarity, and trademark distinctiveness by formulating these issues as text-based assessment tasks that are explicitly grounded in how patents and trademarks are examined and used in practice. From a methodological perspective, it demonstrates how doc2vec combined with LOF, a domain-adaptive Sentence-BERT model, and a fine-tuned generative language model with parameter- efficient adaptation can be applied to patent and trademark texts to construct more fine-grained approaches to support IP assessment. From a practical perspective, the proposed approaches are intended to serve as complementary tools that help patent evaluators scan large patent sets for novel technologies, support trademark examiners in evaluating the similarity of goods and services and enable trademark applicants to anticipate refusal risks and prepare more informed filing strategies. Major: School of Business Administration Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/91098 2026-01-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/91097 Title: Behavioral and Structural Dynamics in Cryptocurrency Markets Author(s): Jung, Jaemin Abstract: Cryptocurrency markets provide a distinctive environment characterized by extreme swings and strong behavioral biases. Considering these features, this dissertation examines how behavioral factors and structural mechanisms, within their respective domains, influence cryptocurrency market dynamics across three essays. The first essay investigates whether investor attention promotes or mitigates daily herding behavior and examines how this effect varies with market sentiment. The results show that higher investor attention tends to reduce the likelihood of herding, suggesting that attention-driven trading facilitates information dissemination and enhances price efficiency. However, this mitigating effect diminishes under heightened fear, implying that attention-driven trading becomes less effective in fearful market conditions. The second essay examines the determinants of herding persistence by analyzing how herding episodes evolve over time. Using a discrete-time hazard model, the study investigates whether the duration of an episode affects the likelihood that herding continues. The results show that duration alone is insignificant; however, longer episodes tend to sustain herding persistence, indicating that herding behaviors do not dissipate quickly once established. Investor attention and sentiment interact with duration to counteract this persistence, suggesting that information-related factors play a key role in moderating prolonged collective behavior. Additionally, after the COIVD-19 shock, duration becomes negatively associated with persistence, indicating a structural change in how herding episodes unfold following a major market disruption. The third essay focuses on the state dynamics of cryptocurrencies, examining whether regime transitions depend on the states of neighboring cryptocurrencies. It defines bubble exposure and distress exposure as the proportions of neighboring cryptocurrencies in bubble and distress regimes, respectively, and decomposes the transition process into two components: occurrence and direction. This decomposition reveals that the mechanisms governing transition occurrence differ from those determining direction. The results show that only bubble exposure significantly affects transition occurrence, and its effect varies depending on the initial regime. In contrast, neither bubble nor distress exposure significantly influences transition direction. These findings suggest that while neighboring states can trigger regime shifts, their direction is primarily determined by coin-specific and market-wide factors. Collectively, these essays provide complementary perspectives on cryptocurrency markets, spanning behavioral dynamics and state transition processes. Together, they reveal that investor behavior and adaptive learning represent the behavioral dimension of market adjustment, whereas Major: School of Business Administration Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/91097 2026-01-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/88131 Title: Three Essays on Innovation and Entrepreneurship Author(s): Min, Kyung-Baek Abstract: Startups play a central role in the development and diffusion of new technologies, often advancing innovation at the frontier while operating under conditions of uncertainty, limited resources, and institutional change. Understanding how these firms make strategic decisions related to technology development, knowledge exchange, and investment requires attention to both internal organizational characteristics and external environmental influences. The following three essays examine the mechanisms through which startups engage with technological innovation and make strategic choices across different contexts. Together, they contribute to a deeper understanding of the interplay between founder backgrounds, interorganizational relationships, and institutional structures in shaping innovation outcomes. The first essay focuses on the relationship between founders’ prior work experiences and the novelty of technologies developed by their startups. Existing research has emphasized the role of prior experience in enhancing innovative capacity, but less is known about how the nature of that experience influences the direction of technological search. This essay distinguishes between different sources of prior experience, such as large incumbents, universities, or cross-industry settings, and theorizes their effects on the development of novel technologies over the course of a startup’s lifecycle. The analysis demonstrates that experience in incumbent firms tends to constrain novelty during the early and growth phases, whereas university-based experience contributes positively to novelty at the initial stage. Cross- industry experience shows delayed benefits, enhancing novelty in later stages. These findings contribute to research on imprinting, learning, and innovation trajectories by identifying heterogeneous and time- dependent effects of founder experience. The second essay examines knowledge spillovers from startups to incumbent firms, with a focus on the conditions under which startup-originated technologies are adopted by established industry players. Drawing on perspectives from innovation diffusion and interorganizational learning, the essay argues that spillovers are shaped by both the characteristics of the knowledge being transferred and the relationships between the knowledge sender and receiver (Supply- and Demand-side factors). Technologies that are more novel are generally less likely to be cited by incumbents due to their higher complexity and evaluative uncertainty. However, when startups are financially linked to incumbents through investment ties—particularly corporate venture capital—these barriers are reduced. The presence of prior investment relationships facilitates greater exposure, alignment of interests, and mutual awareness, increasing the likelihood of knowledge adoption. Moreover, incumbent firms with higher status are generally more active in citing startup technologies, although they exhibit lower engagement with highly novel inventions. These findings offer a more conditional understanding of how novelty, organizational status, and investment ties jointly shape patterns of knowledge diffusion. The third essay considers how changes in merger review policy affect startup financing outcomes. Specifically, it examines a regulatory threshold change that altered the conditions under which mergers and acquisitions were subject to antitrust scrutiny. The analysis posits that such changes influence early- stage investment by affecting expectations around exit pathways. When more acquisitions are exempt from regulatory review, early-stage startups become more susceptible to acquisition before reaching technological maturity, weakening the case for long-term investor engagement. The essay shows that independent venture capital investment into early-stage startups declined following the policy change, while later-stage investment and corporate venture activity remained more stable. Additional analyses suggest that the decline was more pronounced for startups without strong reputational signals or prior funding history. These findings contribute to the literature on entrepreneurial finance and innovation policy by demonstrating that macro-level regulatory decisions can have selective and unequal effects across stages of startup development. Together, the three essays offer a theoretically grounded and empirically detailed account of how startups pursue innovation and adapt to institutional, relational, and organizational conditions. By focusing on the antecedents of novel technology development, the contingencies of knowledge spillovers, and the influence of policy on financing structures, the study advances understanding of strategic behavior in startup ecosystems. The findings contribute to research in strategic management, entrepreneurship, and innovation studies by clarifying the multilevel factors that shape the innovation trajectories of early-stage firms. Major: School of Business Administration Thu, 31 Jul 2025 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/88131 2025-07-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/82485 Title: UNDERSTANDING USERS’ INFORMATION PRIVACY BEHAVIORS Author(s): Park, Jonghwa Major: School of Business Administration Sun, 31 Jan 2021 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/82485 2021-01-31T15:00:00Z