Epistemics and Fragments

Developing the epistemic model of documents with marked up text from articles based on the 'Fragmenter'

June 2016

  1. Dry Run II Papers from MITRE

July 2015

  1. Two-Hop Ras BioPax Model

May 2015


April 2015

  1. Open Access Pathway Logic Papers and Figures
  2. Extended Coprecipitation Frames v2

January 2015

  1. Coprecipitation Frames v2
  2. List of Experimental Motif Types + Definitions
  3. KEfED Database Construction

December 2014

  1. KEfED Modeling of Coprecipitation Ras Papers
  2. Initial Extraction Study of Results-Based Epistemics

October 2014

  1. Pathway Logic Experiment Types
  2. Building a Database of Observations from Result Text
  3. Deploying the BioScholar System
  4. Reading Against a Model of Experimental Evidence

August 2014

  1. Developing NL Annotations for KEfED Elements
  2. Epistemics and Fragments
  3. Generating the Big Mechanisms Evaluation Corpus
  4. A Generative Story for Scientific Text from Experimental Data
  1. Epistemic types and fragments

Based on Anita de Waard’s work, we use a simple representation of ‘Epistemic Segment Types’. These consist of the following codes:

  1. Fact, fact: a claim that has been accepted to be true, a known fact.
  2. Hypothesis, hypothesis: a proposed idea, not supported by evidence
  3. Problem, problem: unresolved, contradictory, or unclear issue
  4. Goal, goal: research goal
  5. Method, method: experimental method
  6. Result, result: a restatement of the outcome of an experiment
  7. Implication, implication: an interpretation of the results, in light of earlier hypotheses and facts
  8. Other-Goal, other-goal: Goal described in a cited paper
  9. Other-Implication, other-implication: Interpretation made in a cited paper
  10. Other-Problem, other-problem: Problem described in a cited paper
  11. Other-Fact, other-fact: Fact with an explicit reference
  12. Other-Hypothesis, other-hypothesis: Hypothesis presented in a cited paper
  13. Other-Method, other-method: Method description in a cited paper
  14. Other-Result, other-result: Experimental Result in a cited paper

  15. BioScholar, the Digital Library subsystem and the Fragmenter

The BioScholar system is the web application I am building to house the framework for knowledge management and engineering based on the KEfEd / CoSI model. The Digitial Libary is concerned with all aspects of managing papers and the Fragmenter is simply the subcomponent within that that handles delineating fragments of text from within the overall body of the document. We have advanced this system significantly.

Figure 1 shows the latest version of the annotation system. This image shows the basic formulation of the Digital Library as a way of managing the full text documents of papers (PDFs, XMLs, HTMLs). It provides fragmenting capability over PDF files where each numbered fragment may be made up of several blocks, each fragment may also be assigned a fragment type (based on the epistemic categories shown above).

Figure 1: Screenshot of the fragmenter superimposed over Sasaki et al. 2011

Figure 2 shows the list of fragments.

Figure 2: Screenshot of the list of fragments themselves (there is a bug where the fragments are ordered by numbers in their string order, not numeric).

Note that Figure 1 has a button labeled brat. This permits the user to automatically dump the current fragments to an installation of the brat annotation system for subsequent markup and processing (see Figure 3).

Figure 3: A list of 14 fragments pertaining to experimental work from Sasaki et al. 2011 where each fragments delineates a separate experiment.

The Bioscholar system now has implemented an annotation framework that permits annotators to process texts according to these epistemics codes is shown at the following link: