Table 6.

Performance metrics across different concept features while using only GPT-4 on the manually annotated test-set.a

Concept featureConcept accuracyConcept F1
Fluctuating cognitive performance0.850.83
Difficulty answering questions0.790.72
Reduced speech output0.940.92
Forgetfulness0.50.47
Bladder dysfunction0.850.80
White-matter changes0.900.88
Delirium0.890.87
A loss of libido1.01.0
Substance dependence1.01.0
Low energy0.890.88
Concept featureConcept accuracyConcept F1
Fluctuating cognitive performance0.850.83
Difficulty answering questions0.790.72
Reduced speech output0.940.92
Forgetfulness0.50.47
Bladder dysfunction0.850.80
White-matter changes0.900.88
Delirium0.890.87
A loss of libido1.01.0
Substance dependence1.01.0
Low energy0.890.88
a

Each concept feature is of the form “The patient exhibits ….”.

Table 6.

Performance metrics across different concept features while using only GPT-4 on the manually annotated test-set.a

Concept featureConcept accuracyConcept F1
Fluctuating cognitive performance0.850.83
Difficulty answering questions0.790.72
Reduced speech output0.940.92
Forgetfulness0.50.47
Bladder dysfunction0.850.80
White-matter changes0.900.88
Delirium0.890.87
A loss of libido1.01.0
Substance dependence1.01.0
Low energy0.890.88
Concept featureConcept accuracyConcept F1
Fluctuating cognitive performance0.850.83
Difficulty answering questions0.790.72
Reduced speech output0.940.92
Forgetfulness0.50.47
Bladder dysfunction0.850.80
White-matter changes0.900.88
Delirium0.890.87
A loss of libido1.01.0
Substance dependence1.01.0
Low energy0.890.88
a

Each concept feature is of the form “The patient exhibits ….”.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close