Data mining techniques have taken a significant role in the diagnosis and prognosis of many health diseases. Still, very little work has been initialized in neurological medical informatics or neurodegenerative disease. Parkinson's Disease (PD) is the second significant neurodegenerative disease (after Alzheimer's), which causes severe complications for patients. PD is a nervous disorder that affects millions of people worldwide. Most of the cases go undetected due to a lack of standard detection methods. This paper attempts to review literature related to PD diagnosis, its stages, and its management using data mining techniques (DMT). The review has been done by exploring the Scopus indexed literature using the query containing the keywords data-mining and Parkinson's disease. This study's focus is to observe how DMT, its applications have developed in PD during the past 16 years. This paper reviews data mining techniques, their applications, and development, through a review of the literature and articles' classification, from 2004 to 2020. We have used keyword indices and article abstracts to identify 273 articles concerning DMT applications from 159 academic journals from Scopus online database. Another objective of this paper is to provide directions to researchers in data mining applications in Parkinson's disease.