Exploratory data analysis (EDA) helps to understand the data better.
The significance of EDA includes :
* Understanding the Data : EDA allows analysts to understand the data they are working with. By creating visualizations, calculating statistics, and exploring the data structure, analysts can begin to understand the patterns, trends, outliers, and anomalies in the data.
* Guiding Future Analysis : EDA can help analysts decide which types of models or algorithms would be most appropriate to apply, and what data cleaning or transformation might be necessary.
* Data Cleaning : During EDA, analysts can identify errors or inconsistencies in the data that need to be corrected, and missing values that need to be addressed, before modeling can take place.
* Assumption Checking : Many statistical models and machine learning algorithms make assumptions about the data (for example, that it is normally distributed, or that its variables are scaled similarly). EDA allows these assumptions to be checked.
* Feature Selection : By exploring correlations and relationships between variables, EDA can help in selecting features for machine learning models.
Explain Outlier :
An outlier is a data point that significantly differs from other similar points. It’s an observation that lies an abnormal distance from other values in a random sample from a population. In other words, an outlier is very much different from the “usual” data.
Depending on the context, outliers can have a significant impact on your data analysis. In statistical analysis, outliers can distort the interpretation of the data by skewing averages and inflating the standard deviation.
