Data cleaning has long been recognized as a necessary step in data analysis, but its implementation is not uniform and differs amongst researchers. The purpose of this article is to investigate the impact of various data cleansing aspects and make recommendations. Normality, outliers, and missing values are the stages that are examined. Advances in information technology (social networks, mobile applications, the Internet of Things, and so on) have recently generated a flood of digital data; nevertheless, converting this data into valuable information for business choices is becoming increasingly difficult. Identifying legitimate, new, potentially relevant, and intelligible patterns from a large volume of data is part of the Knowledge Discovery (KD) process. Preparing the data, on the other hand, is a non-trivial refining effort that necessitates technical skill in data cleansing methods and algorithms. As a result, inexperienced users have a difficult time selecting an appropriate data analysis approach.

turkey tours