COR = class of recommendation.
In the primary prevention of SCD, ICD should be implanted within 40 days after MI in patients with the EF 30-40%, in NYHA class II-III, receiving optimal pharmacological treatment and with predicted survival in good functional condition for at least one year (class I of recommendation, level A of evidence) [ 68].
Most related to our work is the class of hashtag recommendation systems.
According to them, an ICD should be implanted as the secondary prevention in patients who survived cardiac arrest in the course of malignant ventricular arrhythmias unless arrhythmia occurred within the first 24-48 h after fresh myocardial infarction (class I of recommendation, level A of evidence).
Thus, for this class of antiemetics, treatment recommendations can be based on strong evidence.
To help find the set of dishonest recommendation classes from the set of recommendations in R domain, Arning et al.[27] defined a measure called smoothing factor (SF).
The algorithm uses a smoothing factor which detects malicious recommendations by evaluating the impact on the dissimilarity metric by removing a subset of recommendation classes from the set of recommendations.
We compared the prevalence of ILRs actually implanted (indicated by the physician investigator) and that estimated using restricted criteria based on Class I recommendations of the recently published EHRA and ESC guidelines.
Let R c k be the kth recommendation class of R domain and SRdomain be the set of suspicious recommendation classes from R domain, i.e., SRdomain ⊆ R domain.
In order to find out the set of dishonest recommendation R domaindishonest from R domain, the mechanism defined by the proposed approach is as follows: Let R c k be the kth recommendation class of R domain and SRdomain be the set of suspicious recommendation classes from R domain, i.e., SRdomain ⊆ R domain.
