Data Science – EDA/Descriptive statistics(Part – 1)
What you will learn
Students will get an elaborate understanding of exploratory data analysis, also known as descriptive statistics.
We dig deep into the first-moment business decision, aka measures of central tendency.
We gain an understanding of second-moment business decisions, aka measures of dispersion.
We further understand the importance of third and fourth-moment business decisions, aka skewness.
Finally, we also look at the multitude of graphical representations like univariate, bivariate, and multivariate plots.
Description
This program will help aspirants getting into the field of data science understand the concepts of project management methodology. This will be a structured approach in handling data science projects. Importance of understanding business problem alongside understanding the objectives, constraints and defining success criteria will be learnt. Success criteria will include Business, ML as well as Economic aspects. Learn about the first document which gets created on any project which is Project Charter. The various data types and the four measures of data will be explained alongside data collection mechanisms so that appropriate data is obtained for further analysis. Primary data collection techniques including surveys as well as experiments will be explained in detail. Exploratory Data Analysis or Descriptive Analytics will be explained with focus on all the ‘4’ moments of business moments as well as graphical representations, which also includes univariate, bivariate and multivariate plots. Box plots, Histograms, Scatter plots and Q-Q plots will be explained. Prime focus will be in understanding the data preprocessing techniques using Python. This will ensure that appropriate data is given as input for model building. Data preprocessing techniques including outlier analysis, imputation techniques, scaling techniques, etc., will be discussed using practical oriented datasets.
Content
