While conducting data analytics, raw data are not suitable to be directly used in the
process. Data analytic requires data that has been analyzed and sampled to suit the expected
results. However, with raw data, it is difficult to manipulate the data and typically needs crucial
processing before they can be applied in a standard data analysis using appropriate software
(Beer, 2018). Further raw data may contain exaggerations and errors that require to be reviewed
and clarified. Considerably, if data scientists use the raw data without clearly sorting out during
the analysis, the column, measurements, characteristics, and analytic quality may not reflect the
correct results.
Data preprocessing in data analytic follow crucial step in ensuring the quality of the data
to reflect the extraction of substantial insights from original data set. The process requires several
steps which includes acquiring the data set from multiple and disparate sources, importing the
significant libraries using software such as python, importation of the data set to the working
directory, identification of the missing data values and encoding the data that has been
categorized within the data set acquired. Additionally, the data set are spilt and considered as the
output while using analytics such as machine learning data analysis and subset the test data
(Yang, 2018). Finally, feature scaling is conducted to end the data preprocessing, resulting in the
desired results.
Following the required and effectively applying the preprocessing steps in data analytics
is advantageous to the data scientists in achieving their desired objectives. Indeed the process
would ensure all the outliers continued in the raw data are removed for accuracy of the objected
results. Further, the preprocessing steps provide a platform for data scientists to standardize their
DATA ANALYTICS 3
data in forms that can create models and provide reasonable findings. Data algorithms and
structures are also well developed by adhering to the steps in the preprocessing of data.
DATA ANALYTICS 4
References
Beer, D. (2018). Envisioning the power of data analytics. Information, Communication &
Society.
Yang, H. (2018). Data preprocessing.