Data science meets information science meets knowledge management
The discipline of knowledge organisation is largely informed by information science, and it is founded on a suite of methodologies, frameworks, standards and technologies for how humans can organise information (and other knowledge resources) for general access and use. There is also a natural affinity and complementarity between knowledge organisation systems (KOS) such as taxonomies, knowledge graphs and ontologies, and search tools.
In recent years, we have seen increasing convergence between information science and data science, expressed in the various sub-disciplines of data analytics, text analytics, machine learning and artificial intelligence, as advances in technology increase the opportunities for enhancing access to information, whether it is structured or unstructured.
On the other side, we have knowledge management, which sets the strategic purpose for knowledge management (and knowledge organisation) systems. Knowledge management is also very concerned with identifying and characterising the contexts of information use, so that KM systems and KO systems can serve real, practical needs.
These three strands do not always communicate well with each other. Knowledge organisation systems can sometimes appear overly conceptual, complex and impractical, leading to a flight to technology. KM systems are often implemented in ignorance of basic information science principles, and of how the underlying technology actually works. AI/ML applications are often implemented without rigorous conceptual underpinnings, and do not easily scale across multiple working contexts.
In this session, Dave Clarke of Synaptica and Patrick Lambe of Straits Knowledge will discuss the ways in which information science can combine insights from KM and from data science, to develop more rounded, more effective, and more sustainable knowledge organisation architectures within organisations.
In recent years, we have seen increasing convergence between information science and data science, expressed in the various sub-disciplines of data analytics, text analytics, machine learning and artificial intelligence, as advances in technology increase the opportunities for enhancing access to information, whether it is structured or unstructured.
On the other side, we have knowledge management, which sets the strategic purpose for knowledge management (and knowledge organisation) systems. Knowledge management is also very concerned with identifying and characterising the contexts of information use, so that KM systems and KO systems can serve real, practical needs.
These three strands do not always communicate well with each other. Knowledge organisation systems can sometimes appear overly conceptual, complex and impractical, leading to a flight to technology. KM systems are often implemented in ignorance of basic information science principles, and of how the underlying technology actually works. AI/ML applications are often implemented without rigorous conceptual underpinnings, and do not easily scale across multiple working contexts.
In this session, Dave Clarke of Synaptica and Patrick Lambe of Straits Knowledge will discuss the ways in which information science can combine insights from KM and from data science, to develop more rounded, more effective, and more sustainable knowledge organisation architectures within organisations.
About the Speakers
Dave Clarke is co-founder and CEO of the Synaptica® group of companies, providers of enterprise software solutions for knowledge organization and discovery. He served on the authoring committee of the 2005 version of the US national standard for controlled vocabularies, ANSI/NISO Z39.19. Dave leads research and development at Synaptica, including software solutions for taxonomy and ontology management, text analytics and auto-categorization, image annotation and indexing, and Linked Data management. He is involved in educational outreach programs including LD4PE, the Linked Data for Professional Education initiative of DCMI. Synaptica software solutions have attracted numerous international awards including: Knowledge Management World magazine’s 100 Companies that Matter in KM and Trend Setting Product of the Year (multiple awards between 2011 and 2017). In 2016 Clarke was awarded the Knowledge Management Leadership Award by the Global Knowledge Management Congress. Dave is a Fellow of the Royal Society of Arts, London, and a Leadership Fellow of St. George’s House, Windsor Castle. Patrick Lambe is a globally recognised knowledge management practitioner, and an expert in bringing KM principles to the discipline of knowledge organisation. Patrick was originally trained in Information and Library Science and arrived in KM via a second career in training and development. He is the author of Organising Knowledge: Taxonomies, Knowledge and Organisation Effectiveness (Oxford: Chandos 2007), and co-author with Nick Milton of the award-winning The Knowledge Manager's Handbook 2nd ed. (London: Kogan Page 2019). Patrick is Visiting Professor in the KIM PhD programme at Bangkok University, former President of ISKO Singapore, and his next book, on Principles of Knowledge Auditing: Foundations of Knowledge Management Implementation will be published by the MIT Press in early 2023. Patrick spends his time between Singapore and Dublin, Ireland. |
Date/Time
Venue Type of Event Who should attend Fee |
Friday, 15 July, 3.30-5.00pm SGT
This session was facilitated via Zoom. Attendance was free but prior registration was required. Case Discussion | Networking | Panel | Site Visit | Talk & Discussion | Workshop * Knowledge managers, managers responsible for data strategy, data policy, data governance, information managers, teams working in digital transformation, teams working on AI/ML applications, managers responsible for data systems, dashboards, analytics and big data applications. This event was free, but prior registration required. |
event materials
Here are the slides used by Dave and Patrick, updated with the results of the two polls taken during the session, and with links to two additional resources mentioned in the discussion. Download here.
Here are the edited notes of the discussion in the chat channel during the call. Download here.
In the discussion we mentioned the need for outputs of data science to be explainable if applications are to be capable of being directed and managed. We also spoke about the weakness of the simplistic DIKW pyramid. Here are two additional resources to follow up on those matters.
Critique of the DIKW pyramid for being too simplistic
Explainability of AI/ML/ data science outputs
The edited recording of the session is in two parts. In Part 1, Dave Clarke discusses the relationship between information science and data science, and we discuss the relevance of the DIKW pyramid. In Part 2, Patrick Lambe discusses the big picture view of how the three disciplines can relate to each other, and the orchestration role (and orchestration challenges) that KM can play.
Here are the edited notes of the discussion in the chat channel during the call. Download here.
In the discussion we mentioned the need for outputs of data science to be explainable if applications are to be capable of being directed and managed. We also spoke about the weakness of the simplistic DIKW pyramid. Here are two additional resources to follow up on those matters.
Critique of the DIKW pyramid for being too simplistic
- Williams, D. (2014). Models, metaphors and symbols for information and knowledge systems. Journal of Entrepreneurship, Management and Innovation, 10(1), 79–107.
- https://jemi.edu.pl/vol-10-issue-1-2014/models-metaphors-and-symbols-for-information-and-knowledge-systems
- Neves, Ana (2020) Data to Knowledge to Innovation. LinkedIn post, June 18.
- https://www.linkedin.com/pulse/data-knowledge-innovation-ana-neves/
Explainability of AI/ML/ data science outputs
- Klein, G. (2020). AIQ: Artificial Intelligence Quotient. Helping people get smarter about smart machines. Psychology Today, July 1st.
- https://www.psychologytoday.com/ie/blog/seeing-what-others-dont/202007/aiq-artificial-intelligence-quotient
The edited recording of the session is in two parts. In Part 1, Dave Clarke discusses the relationship between information science and data science, and we discuss the relevance of the DIKW pyramid. In Part 2, Patrick Lambe discusses the big picture view of how the three disciplines can relate to each other, and the orchestration role (and orchestration challenges) that KM can play.
ISKO Singapore is registered in Singapore. Society registration number T15SS0160B
By continuing to use the ISKO Singapore website you are agreeing that ISKO Singapore may collect, use and disclose your personal data obtained by ISKO Singapore as a result of your use of the ISKO Singapore website. Please consult our data protection policy, including how you may access and correct your personal data or withdraw consent to the collection, use or disclosure of your personal data.
By continuing to use the ISKO Singapore website you are agreeing that ISKO Singapore may collect, use and disclose your personal data obtained by ISKO Singapore as a result of your use of the ISKO Singapore website. Please consult our data protection policy, including how you may access and correct your personal data or withdraw consent to the collection, use or disclosure of your personal data.