580 California St., Suite 400
San Francisco, CA, 94104
Key research themes
1. How can symbolic and metaphorical approaches enhance our understanding of the role of information in knowledge acquisition?
This theme explores the conceptualization of information as an active and dynamic component in the knowledge acquisition process, using symbolic and metaphorical models to clarify its function and relation to data, knowledge, and wisdom. It addresses the theoretical challenges around defining and interpreting information's meaning and its foundational role in transforming data into knowledge within epistemological frameworks.
Key finding: This paper proposes a novel symbolic representation of the knowledge acquisition process as a 'tree of knowledge,' rooted in 'data earth,' rising with 'information sap,' and illuminated by a 'sun of the mind' symbolizing... Read more
2. What methods and frameworks are effective for capturing and representing expert knowledge for knowledge-based and intelligent systems?
This theme centers on methodologies, frameworks, and challenges involved in acquiring, encoding, and reusing expert knowledge in computational systems such as expert systems and knowledge bases. It emphasizes the process steps, constraints, and techniques necessary to transform tacit and explicit knowledge into usable representations that support inference and decision making in intelligent systems.
Key finding: The paper reports a comprehensive case study on the actual process of knowledge reuse in building a knowledge base to address crisis management problems. It outlines a multi-step process—comprehension, translation, slicing,... Read more
Key finding: This work proposes a structured framework for manual knowledge acquisition emphasizing both tacit and explicit knowledge from experts, tailored for expert and knowledge base systems. The framework identifies critical... Read more
Key finding: This paper presents a web-based knowledge acquisition system designed for capturing expertise from software designers, integrating process and decision laddering techniques with concept mapping to create hierarchical and... Read more
Key finding: The authors develop a knowledge acquisition model specifically for the maintenance department context, focusing on the formalization, systematization, and retrieval of explicit and tacit knowledge related to machinery... Read more
3. How can machine learning and data mining techniques facilitate knowledge extraction and prediction in occupational injuries and business contexts?
This theme addresses the application of machine learning algorithms and data mining methods to extract actionable knowledge from complex, heterogeneous data sources in occupational safety and software systems. It investigates predictive modeling for injury risk assessment and automated business knowledge extraction from legacy systems to improve organizational decision-making and reduce costs associated with maintenance or healthcare outcomes.
Key finding: This paper documents the successful application of various machine learning models—Support Vector Machines, Neural Networks, and XGBoost—in predicting occupational injuries across multiple industries, achieving high accuracy... Read more
Key finding: The chapter defines and distinguishes data, information, and knowledge, introducing data mining as a powerful technique to uncover hidden patterns from large data sets. It outlines the process flow of data mining, tasks... Read more
Key finding: This paper introduces the concept of Knowledge Discovery Guided by Domain Knowledge (KDDK), integrating ontologies into the knowledge discovery process to improve interpretability and reuse of discovered knowledge. It... Read more
Key finding: Through systematic literature review, this paper identifies, classifies, and critiques methods for extracting business knowledge from legacy software artifacts to aid software maintenance and evolution. It points out the... Read more
4. What models and technological frameworks support effective knowledge management, sharing, and use in organizational and community settings?
This theme investigates theoretical and practical frameworks that enable organizations and communities to capture, share, and utilize tacit and explicit knowledge. It explores knowledge management processes, enablers like collaborative technologies and communities of practice, as well as the challenges of managing distributed and dynamic knowledge assets, often within digital and semantic infrastructures.
Knowledge Management (KM) Framework for Representing Lessons Learned System for Communities of Pract
Key finding: The study adapts Nonaka and Takeuchi's knowledge spiral model to design a knowledge management framework that supports the creation, dissemination, and retention of lessons learned within communities of practice. It... Read more
Key finding: This empirical study investigates how high-skilled actors manage information and knowledge in knowledge-intensive work processes characterized by distributed collaboration. It reveals the complexities of articulating diverse... Read more
Key finding: Using pedagogical discussion cases co-developed by academia and industry to simulate real-life business challenges, this research develops a conceptual framework elucidating knowledge capture, sharing, and retention dynamics... Read more
Key finding: This work critically examines traditional knowledge representation formalisms (production rules, logic, frames, object-oriented, etc.) highlighting their suitability and limitations for diverse knowledge types and emphasizing... Read more
Key finding: The paper discusses integration approaches between Knowledge Organization Systems (KOS) and Metadata Registries (MDR), highlighting semantic alignment of conceptual vocabularies with data element descriptions per ISO/IEC... Read more
5. How can philosophical and epistemological perspectives on knowledge and knowing-how inform the modeling of knowledge action and intention?
This theme probes into the integration of knowledge-first epistemology with intellectualist theses that conceptualize knowing-how as a form of knowing-that. It explores theoretical reasoning about the roles of knowledge, belief, desire, intention, and action in practical reasoning and how these mental states are structured and interrelated to explain human agency and knowledge use.
Key finding: The paper offers a sophisticated integration of knowledge-first epistemology with intellectualist accounts of knowing-how, positing that intention logically stands to action as belief does to knowledge in practical reasoning.... Read more
6. What innovative conceptual models and digital tools facilitate the capture, representation, and visualization of knowledge within semantic web and organizational contexts?
This theme reviews contemporary approaches emphasizing the unified treatment of knowledge capture, representation, and visualization, especially through semantic web ontologies and graphical tools like concept maps. It underscores the importance of user-friendly, visually oriented knowledge representation to empower domain experts to codify and share complex knowledge structures, highlighting challenges in ontology instantiation and symbol grounding.
Key finding: This paper surveys recent advances in knowledge capture, representation, and visualization, emphasizing their interrelatedness and convergence in semantic web ontologies. It highlights how graphical knowledge representations,... Read more
7. How can large-scale data analysis and automated methodologies assist cultural heritage institutions in identifying and managing contentious or problematic language in historical collections?
This theme investigates computational techniques to detect outdated, offensive, or problematic terms in cultural heritage archives, focusing on the creation of annotated datasets and automated classification methods. It explores the challenges of subjectivity in determining contentiousness, differences between expert and crowd annotations, and the implications for large-scale semantic annotation and contextualization within digital cultural collections.
Key finding: The authors introduce 'ConConCor,' a richly annotated corpus of 2,715 text snippets with contentious terms in the context of historical Dutch newspaper archives, labeled by experts and crowd workers. They analyze inter-rater... Read more
8. How can expertization level ranking be accurately estimated using internet data to improve expert finding in e-learning systems?
This theme explores methodologies for assessing an individual's expertise level by leveraging internet-based data aggregated via search engines, aiming to improve upon traditional expertise detection methods limited to self-classification and local document relevance. It addresses data extraction, filtering, and weighting techniques to dynamically estimate expertization accurately in open and heterogeneous data environments.
Key finding: The paper proposes a novel Dynamic Expertization Estimating System (DEES) that leverages search engines as agents to extract and weight web-based data relevant to an expert's identity and claimed expertise, overcoming... Read more