Page 151 - AIH-2-3

Artificial Intelligence in Health Explainable solutions from AI for HSSs

5.1. Implementing KESs in manufacturing “covers” several classes of tasks (for example, diagnosis of
environments diseases regardless of their etiology ). The knowledge base
28
An example of an environment for producing clinical is an ever-changing component and should be formed based
KESs is thematic medical ontological portals, and an on the created ontology. KES is a special case of applied
example of a toolkit for modern KES production and software services on the IACPaaS platform (IACPaaS
maintenance environments is the Intelligent Adaptive services). They need the information resources (of the
Clinical Platform as a Service (IACPaaS) cloud platform. portal), such as knowledge bases. The set of IACPaaS-portal
tools for the formation of KES’s information components
The creation of an ontological portal starts with the are the IACPaaS-editor of knowledge base, generated
creation of an ontology as a semantically structured basis in terms of ontology with a self-adaptive user interface
for the creation and processing of ontological information (Figure 2), and the IACPaaS data editor (with self-adaptive
resources (Figure 1).
user interface) (Figure 3).
An ontology (as a template or “meta-information”)
The regular generator of information editors makes
defines a semantic model, structure, rules for the formation knowledge base editors available and constantly operating
of information resources, limitations of its interpretation, in the production environment (Figure 2).
or processing rules. Usually, cognitive scientists (users of
27
the IACPaaS platform) form such semantic structures and In addition, the IACPaaS toolkit to form software
rules for the community of experts and specialists (users solvers (in addition to сoding tools for software units)
of the IACPaaS portals) (Figure 1). The tool for ontology contains (Figure 4) (i) an IACPaaS meta-information editor
formation is the IACPaaS meta-information-editor to explain the resulting structure, (ii) a “master” of the
“ontology editor.” formation of declarative parts of software units and their
blocks-reasoners, which conduct reasoning on ontological
Creating an ontology is a creative process requiring information, (iii) a generator of code blanks (according to
extensive analytical work and a systematic domain analysis declarative parts) for new IACPaaS software units, (iv) a
to identify common patterns in forming knowledge, solver constructor from GUI (or “root”-unit or IACPaaS-
structure, and integrity constraints. The ontology is agent), a software unit (being represented by its declarative
generally not changeable throughout the life cycle of the parts), including connector modules, and (v) tools for
KES. The separation of an ontology from a knowledge testing IACPaaS agents and preparing them for reuse.
base (and a set of facts) leads to the ability to interpret
them with a specialized ontology-based algorithm. The 5.2. Creating clinical KESs using Intelligent Adaptive
algorithm (ontological reasoner) searches for or refutes Clinical Platform as a Service environment tools
hypotheses by traversing the (declarative) knowledge The KES design technology in the proposed environment
base. It “sorts” the knowledge base statements of each type provides for a sequence of activities.
related to the hypothesis, comparing these sentences of (i) Find pre-existing knowledge ontologies whose
input information (patient’s document). concepts and relationships are sufficient for the tasks
The medicine ontology was carried out by cognitive and data ontology. If the IACPaaS platform does
scientists (knowledge engineers) together with experts. It not already have an ontology for the problem under

Figure 1. Basic components of the environment for developing basic components of a knowledge-enabled system

Volume 2 Issue 3 (2025) 145 doi: 10.36922/aih.5736