Now a Days cancer is a common diseases among people. And is generally having an obstruction in getting the cancer cured. If the cancer is not predicted in an early stage it can lead to a treacherous death of a person. Our paper describes a classification method known as Naïve Bayes classifier for the prediction of cervical cancer which is a kind of cancer occurring in females. Cervix is a lower portion of uterus in the human female reproductive system. Our algorithm is based on the concept of conditional probability. Naïve Bayes classification algorithm is based on the assumption of independent amount predictors. Bayes classification assumes that the presence of an attribute in a class is unrelated to any other attribute of that class and also if both are related they are independent of the existence of each other. Since all the attributes are contributing independently to the probability. Another feature of this algorithm is that it can be applied to both binary and multi. Further we require less training datasets unlike other algorithms of machine learning. In this algorithm we consider a hypothesis. And we get the probability of hypothesis before we get its evidence and the same hypothesis after getting its evidence. We are using gaussian naïve Bayes algorithm for our prediction algorithm. For this probability calculation we need to consider a large number of attributes such as age of women, Number of intercourse with multiple men, first mating, Number of times of conceiving, Smoking habits, frequency and duration of Smoking(no. of years), Dx, Hinselmann, Schiller, Citology, Biopsy etc. And using these attributes of dataset we calculate the probabilities of occurrence and then finally we use those probabilities for our final predictions, Here we are taking common ratio of training and testing data sets which is 70% and 30%.
Girdonia, M., Garg, R., Jeyabashkharan, P., & Minu, M. S. (2019). Cervical cancer prediction using naïve bayes classification. International Journal of Engineering and Advanced Technology, 8(4), 784–787.