Ontologies as Nested Facet Systems for Human-Data Interaction

Submitted by GQ Zhang on 05/06/2019 - 07:03

Tracking #: 2201-3414

Authors: 
GQ Zhang
Shiqiang Tao
Ningzhou Zeng
Licong Cui1

Responsible editor: 
Guest Editor 10-years SWJ

Submission type: 
Other
Abstract: 
Irrespective of data size and complexity, query and exploration tools for accessing data resources remain a central linkage for human-data interaction. A fundamental barrier in making query interfaces easier to use, ultimately as easy as online shopping, is the lack of faceted, interactive capabilities. We propose to repurpose existing ontologies by transforming them into nested facet systems (NFS) to support human-data interaction. Two basic issues need to be addressed for this to happen: one is that the structure and quality of ontologies need to be examined and elevated for the purpose of NFS; the second is that mappings from data-source specific metadata to a corresponding NFS need to be developed to support this new generation of NFS-enabled web-interfaces. The purpose of this paper is to introduce the concept of NFS and outline opportunities involved in using ontologies as NFS for querying and exploring data, especially in the biomedical domain.
Full PDF Version: 
PDF icon swj2201.pdf
Revised Version:
Ontologies as Nested Facet Systems for Human-Data Interaction
Tags: 
Reviewed

Decision/Status: 
Minor Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
By Stefano Borgo submitted on 17/Jun/2019
Suggestion:
Minor Revision
Review Comment:

Nice paper and nice topic. I have some comments to improve the presentation.
- the Amazon example is very useful but may be confusing: the Amazon user can be a layman, while I guess to use a service based on NFS one needs some domain knowledge. Pls, clarify better the expected user of the NFS.
- I would like to know how the NSF approach works in practice and why isn't enough to present or highlight the corresponding fragment of the ontology.
- what happens when two distinct ontologies cover the same concept? Do we end up with two distinct NSF? isn't this problematic?
- pls, integrate better the material on resources and ontologies of Sect. 4 with the rest of the paper.
- the authors report that the approach has already been developed to some extent, any information of its use so far?

Review #2
Anonymous submitted on 01/Jul/2019
Suggestion:
Major Revision
Review Comment:

In this paper, the authors propose the notion of nested facet systems (NSFs) for human-data interaction. They then propose a method to transform ontologies to NFSs. The idea of introducing NSFs is quite interesting and seems to be useful in practice, especially for information retrieval. However, I think the presented work is too preliminary to ensure publication in a journal. Significant extension is necessary. Some suggestions are given as follows.

First, completeness of NFS should be defined in section 2 and some properties of the soundness and completeness should be given.

Second, examples of NFS should be given, in which areas can NFS be applied? My understanding is that NFS can be used in many applications where IR is applied, but the authors should specify these applications.

Third, Section 4 is not well written. The authors claim that they will discuss about completeness and soundness of NFS in details and give examples, but the "examples" given in this section are high level and not intuitive, and no formal proof of the completeness and soundness of NFS for SNOMED CT and NCIt is given. This section should be significantly modified.

Fourth, related work should be given for NFS. Hierarchical labels should have been used in machine learning, authors should do a survey on this and discuss related work in details. It seems that the authors totally ignore related work and only give a formal definition of the hierarchical labels.

Fifth, the authors should provide a method for transforming an ontology in OWL to NFS and prove its soundness and completeness if possible.