In Silico Hypothesis Discovery - Translational Informatics: Realizing the Promise of Knowledge-Driven Healthcare - page 138

used during reasoning processes [ 31 - 36 ]. Using frequency-based measures (e.g.,
measuring the frequency with which a given entity is related to other entities within
the knowledge collection) in addition to simple statistics can allow for the assess-
ment of the degree of interrelatedness of a set of multiple hypotheses [ 37 ].

8.4.3 Information Theoretic Methods

Information theoretic methods are most commonly applied to measure multi-source
agreement in an aggregate collection of multiple hypotheses. The use of informa-
tion theory to evaluate the agreement between multiple sources is based on the
argument that if such agreement exists, it will be manifested as repetitive patterns
within the resulting information constructs. To utilize this verifi cation and validation
approach, the relationships between units of knowledge that make up each constitu-
ent hypothesis must be represented as a numerical matrix, where each cell contains
a numerical indication of the strength of the relationship between the two units of
knowledge identifi ed by the corresponding row and column indices. Given such a
matrix, repeating patterns can be quantifi ed based on their effect on information
content or complexity. Matrix complexity is determined by calculating the number
of repeating patterns within the matrix less the contribution of the overall envi-
ronment within which the matrix is constructed. The probability of each repeating
pattern detected in the actual matrix occurring randomly or as a result of the envi-
ronmental contribution can be computed by generating multiple random matrices.
As matrix complexity decreases, the degree of multi-source agreement increases
[ 35 ]. This type of evaluation method is summarized in Fig. 8.11 , and further detail
can be found in the work reported on by Kudikyala et al. [ 35 ].

8.4.4 Graph Theoretic Methods

Graph theoretic methods are based on the ability to represent knowledge-based
formulations, such as the output of intelligent agents, as graph constructs, where
individual units of information or knowledge are represented as nodes, and the rela-
tionships between these units as arcs. Such graph representation of knowledge col-
lections has been described in a number of areas, including ontologies [ 32 , 38 ],
taxonomies [ 39 , 40 ], controlled terminologies [ 41 ] and semantic networks [ 40 , 42 ].
Given a particular graph representation of a hypothesis or set of hypotheses, the
degree of interrelatedness of those knowledge-based products can be assessed using
a group of graph-theoretic techniques known as class cohesion measures. Such
metrics are used to assess the degree of cohesion, a property representative of con-
nectivity within a graph. Specifi c class cohesion measurement algorithms include
the Lack of Cohesion of Methods (LCOM), Confi gurational-Bias Monte Carlo
(CBMC), Improved Confi gurational-Bias Monte Carlo (ICBMC) and Geometrical