Bayesian Network-Based Model for the Diagnosis of Deterioration of Semantic Content Compatible with Alzheimer’s Disease - Foundations on Natural and Artificial Computation - page 425

CD ⇒ CognitiveDeterioration ⇒∀ x 1 ∈{ absent, slight, moderate, serious }

CDLC ⇒ CognitiveDeteriorationLivingCreatures ⇒∀ x 2 ∈

{ absent, slight, moderate, serious }

}
N=> Is a function that counts the number of records in the database of
instances that fulfil the query criteria.

AD ⇒ Alzheimer sDisease ⇒∀ x 3 ∈{ absent, present

Fig. 4. Clusters for cognitive deterioration generated with WEKA

Once the clusters for Cognitive Deterioration and Cognitive Deterioration in
Living Creatures and Non-Living Creatures have been defined, each instance of
the database has to be categorised according to the number of attributes that the
patient has produced for each object. We have to make the symptom variables
discrete, and for this we use the K-means algorithm, with as many clusters
and states that we want the variables to have. Once the symptom variables are
discrete, the database is analysed to obtain the conditional probabilities thus:

N ( CATx 1 ,DCox 2 )
N ( DCOx 2 )

P ( CATx 1 | DCOx 2 )=

where

⎧
⎨

⎫
⎬ Taxonomical

Types
Parts
Functional
Evaluative
Place/Habitat
Behaviour
Causes
Procedural
LyfeCycle
Others

CAT ⇒ SemanticCategories ⇒∀ x 1 ∈

⎩ ⎭

{ absent, slight, moderate, serious }