Check out the Zubir Said Knowledge Graph launched in July 2021: https://ZubirSaid.sg
Also Singapore Pioneers – Knowledge Graph & Social Network launched in Dec. 2021: https://SingPioneers.sg
Polyglot Medicine Knowledge Graph launched in May 2023: https://kgraph.sg/polyglot

My knowledge organization projects are currently focused on ontology and knowledge graph design and development to support:

  • text summarization of research papers
  • automatic discourse analysis of research papers
  • research data management
  • digital heritage portal

Project descriptions below:

Polyglot Medicine Knowledge Graph

Check out the Polyglot Medicine Knowledge Graph launched in May 2023: https://kgraph.sg/polyglot

This is a project to develop a database of Asian traditional and herbal medicines. Current database coverage is mainly on Chinese traditional medicines, expanding to Malay and Southeast Asian traditional medicines. The knowledge graph currently links the medicine names in the original and English languages, to alternative names and scientific names, to plant/animal parts they are made from, to literary and historical sources they were mentioned in, to geographic areas they were associated with, and to external database records with information on them. There are plans to link the medicines to pharmacological information later.

A collaborative project with Dr Michael Stanley-Baker (School of Humanities, Nanyang Technological University.

Recent papers:

  • Khoo, C.S.G., Ng, S.G., Chan, C.F., Stanley-Baker, M., Tan, E.A.L., & Cheng, W.N. (2024). Knowledge graph visualization interface for digital heritage collections: Design issues and recommendations. Information Technology and Libraries, 43(1). https://doi.org/10.5860/ital.v43i1.16719
  • Khoo, C.S.G., Stanley-Baker, M., Zakaria, F.B., Chen, J., Ang, S.Q.R., & Huang, B. (2023). Development of the Polyglot Asian Medicine Knowledge Graph System. In D.H. Goh, S.J. Chen, & S. Tuarob (Eds.), 25th International Conference on Asia-Pacific Digital Libraries, ICADL2023 (Lecture Notes in Computer Science, vol 14458, pp. 3-11). Springer. https://doi.org/10.1007/978-981-99-8088-8_1

Knowledge Graph & Visualization Interface for a Digital Archive Collection—the Zubir Said Collection

Check out version 1 of the graph (network) visualization interface launched in July 2021: https://ZubirSaid.sg
Also a text-based browse interface implemented using React.JS framework: https://zubirsaid.sg/ZS.text.html

The project seeks to develop a knowledge graph and visualization interface for the Zubir Said digital archive collection – as an example of applying knowledge graph technologies to digital heritage/archive content.

Zubir Said was the composer of the Singapore national anthem and a pioneer songwriter. The knowledge graph represents materials and content in Zubir Said’s personal archive as a network of concepts to support visualization, associative browsing, and research. Its mindmap-like feature helps one make correlations between the various artefacts and events surrounding Zubir Said’s songs, manuscripts and printed scores, official and personal correspondence, as well as interviews on various print and online media – to help users develop coherent understandings/narratives on topics related to the national anthem and other compositions of Zubir Said, and the social and cultural milieu of the time.

Version 1 of the interface was launched in July 2021. The technologies employed include:

  • A WordPress (content management) installation to host articles and digital files (at https://zubirsaid.info)
  • Cytoscape.js javascript library for graph (network) visualization (hosted on https://zubirsaid.sg), with layout libraries (cola, fcose)
  • React.js javascript library for implementing the text-based browse interface (hosted on https://zubirsaid.sg)
  • Node.js javascript runtime environment + Koa Web framework (deployed on a Google App Engine) to host the middleware and API connecting the graph/text interface to the backend graph database.
  • Neo4j graph database (as a Neo4j Aura cloud service) to store the knowledge graph and support graph matching and other graph operations.

Future work includes converting the music scores to digital music notation, using VexFlow to display the music notation, and Web Audio API to synthesize the sound.

A collaborative project with Dr Eleanor Tan (Nanyang Academy of Fine Arts, Singapore), Ng Siam Gek and Chan Chwee Fong.

Social Network of Singapore Pioneers

Version 1 of SingPioneers.sg was launched in Dec 2021: https://SingPioneers.sg

SingPioneers.sg visualizes biographical information and social networks of famous persons in the history of Singapore – based on the Personalities pages of Singapore Infopedia, the National Library Board’s electronic encyclopedia on Singapore. Version 1 of the system features 20 SingPioneers with a handcrafted knowledge graph.

Work is ongoing to develop an automatic information extraction program to extract information from over 400 biographies to populate the knowledge graph and social network.

A collaborative project with the National Library Board and Ng Siam Gek (National Library digital fellow for 2021).

An Ontology for Conceptual Analysis of Signature Pedagogies

In collaboration with: Rebecca Kan (Nanyang Academy of Fine Arts, Singapore)
Signature pedagogy is a multifaceted concept that has been found useful by education researchers for analyzing and characterizing pedagogical methods for particular professions. A signature pedagogy must impart not just the formal disciplinary knowledge but also the cognitive, behavioral and attitudinal norms of the profession. The pedagogy is characterized by four dimensions of surface structure, deep structure, implicit structure and shadow structure. The abstract and multifaceted nature of the concept affords researchers latitude in construing and applying the concept in analyzing and describing pedagogies, making it difficult to compare pedagogies across professions and across academic papers. Based on a conceptual analysis of twenty-four journal papers describing signature pedagogies of various professions, the paper proposes and outlines a Research in Signature Pedagogy (ReSiPe) ontology.

Project: Ontology to support data curation of social science statistical data

Many social science datasets are now available for public access in research data repositories (e.g., ICPSR), government data repositories (e.g., data.gov.uk and data.gov.sg), and many university data repositories. Such data repositories are meant to promote data reuse, data integration and big data applications. However, it is difficult for human users (let alone computer applications) to find appropriate datasets, understand them, integrate them and prepare them for data mining. An ontology of concepts and data elements in social science datasets as well as a metadata application profile for describing the datasets are being developed. A graph database application (using Neo4j) will be developed to store the ontology and metadata records and to support graph mining. A visualization tool (Cytoscape) will be used to visualize the ontology and metadata information to support users in exploring, understanding and integration the datasets.

Project: Taxonomies and ontologies for digital heritage portals

Many memory institutions have set up online portals to provide public access to their heritage collections. However, little is known about how such portal interfaces should be designed and the content organized to support user browsing and learning. Many online cultural heritage portals adopt an organization scheme that is either content-oriented or institution-oriented rather than one that is user-oriented.

Few heritage portals provide a browse structure for users to explore heritage content. Users are expected to think of specific topics and keywords to search. A limited kind of browse structure is often provided in the form of an alphabetic list of “collections.” There is no bird’s eye view of what the heritage portal contains.

The portals also do not support users in learning and synthesizing an understanding of a heritage topic. They do not indicate what aspects or attributes of a heritage topic is covered in the portal content/resources, and relationships to other topics.

The study makes a few assumptions about learning:

  • Learning about a topic involves synthesizing information into a coherent understanding and linking together related concepts (i.e. aspects and attributes of the topic, and relations to other topics). This coherent understanding can be represented as an ontology—a set of concepts linked with conceptual relations.
  • In the course of reading the memory postings, users will identify salient, important or interesting aspects/attributes about a topic, as a first step in synthesizing a coherent understanding.
  • Users already have some prior knowledge of a topic as well as opinion about the important aspects and attributes of a topic. This prior knowledge guides searching and browsing, and provides an initial template structure for synthesizing an understanding. Some of the associated aspects/attributes may be tacit—something the user can recognize but might not remember in free recall. So the system can support the user’s learning by providing a taxonomy or mind map to guide the user in browsing and making sense of the topic and memories.

Recent papers:

  • Khoo, C.S.G., Ta, M.T., Win, K.P., & Thi, C.S.S. (2016). Visualization of heritage content in the Singapore Memory Portal to support user learning. In Proceedings of the 7th Asia-Pacific Conference on Library & Information Education & Practice (A-LIEP 2016) (pp. 230-236). Nanjing, China: School of Information Management, Nanjing University. [PDF]
  • Khoo, C.S.G., Teng, T.B.R., Ng, H.C., & Wong, K.P. (2014). Developing a taxonomy to support user browsing and learning in a digital heritage portal with crowd-sourced content. In W. Babik (Ed.), Proceedings of the 13th International ISKO Conference, 19-22 May 2014, Krakow, Poland (pp. 266-273). Wurzburg: Ergon Verlag. [PDF]
  • Srieedar, J., & Khoo, C.S.G. (2013). A user study of the Singapore Memory Portal to derive a taxonomy for content organization. In Proceedings of the 2013 Digital Heritage International Congress, 28 Oct-1 Nov 2013, Marseille, France (IEEE catalog no. CFP1308W-USB, pp. 297-305). [PDF]

Studies of ontology design

I have used ontologies to represent knowledge for particular domains and applications:

  • Ontology to represent the knowledge base of a clinical decision support system [PDF]
  • Ontology to represent disease-treatment information found in abstracts of medical articles [PDF]
  • Ontology to represent the research objectives of social science research [PDF1, PDF2]