Naive Bayers Classifier Algorithm¶
Naive Bayers Classifiers is the most prominent algorithm used in Sugaroid. Most responses are classified with Naive Bayers Classifiers. Naive Bayes Algorithm is a linear and scalable algorithm which can guess the most appropriate answer based on statistical probability
Sugaroid uses Naive Bayers Algorithm on a list of responses saved as a
portable SQL database; the most probable reponses are then filtered out
using summation algorithm. Naive Bayers Classifiers is a light weight
algorithm which uses less CPU load, but has high memory usage. Using
mysql instead of sqlite3 can however, considerably decrease
memory usage sacrificing portability
The lesser complex version of the Naive Bayers Classifier is used explicitly in Sugaroid for deriving the responses. The equation used, is given below
where p is the Probability, D|C is the conditional probability.