Drug Recommendation System based on Sentiment Analysis of Drug Reviews using Machine Learning

Abstract

Ever since the coronavirus emerged, there
has been an extreme scarcity of genuine
clinical resources, such as doctors, nurses,
and other clinical professionals, as well as
essential clinical gear and meds. An
extraordinary number of individuals pass on
the grounds that the entire clinical local area
is in a condition of frenzy. Since the
medication was not promptly accessible,
individuals started utilizing it all alone
without appropriate meeting, which
exacerbated their medical issues. Recent
years have seen a rise in inventive work for
automation, and machine learning has proven
useful in many domains. A medication
recommender system that may significantly
lessen the burden on experts isthe goal of this
article. In this review, we foster a framework
for medication suggestion in light of patient
surveys. We use vectorization methods like
Bow, TF-IDF, Word2Vec, and Manual
Element Investigation to foresee the feeling
of audits. The framework then utilizes these
expectations to decide the best medication for
a particular sickness utilizing different
characterization calculations. Quality
measurements like exactness, accuracy,
review, and region under the bend (AUC)
were utilized to evaluate the projected
mentalities. With an exactness pace of 93%,
the discoveries show that classifier Linear
SVC using TF-IDF vectorization is the best
model.