Understanding the model does not end with finding accuracy. Each model performance measure is important and tells us many important aspects of the data and model itself. In this blog, I will try to explain every small inference which we can draw out of the measure.
When it comes to data manipulation, I prefer pandas over SQL. I feel pandas is easy to understand and faster. I will discuss some of the widely asked questions on SQL which we can do in pandas.
Following are two tables 1. Employee 2. EmployeePosition, Let’s keep the 1st table in data frame df_1 and the 2nd table in data frame df_2.
The thing is, I don’t want to be sold to when I walk into a store. I want to be welcomed. – Angela Ahrendts, Senior Vice President of Retail at Apple Inc.
Data science helps retail businesses to capture behaviors and utilize the customer data and turn it into actionable insights that will boost revenue. The following are a few examples of how data is used in retail:
A decision tree is a supervised learning algorithm used for classification that uses a flowchart where each internal node denotes a condition ( a test) and each terminal node holds a class label (1 or 0, Yes or No, Default or Not Default).
Let us understand a decision tree with an example:
Clustering is an unsupervised learning algorithm that groups the data and these groups show similar properties. Clustering is unsupervised because data does not contain any dependent variable.
The principal component analysis is one of the dimensionality reduction techniques widely used in Machine Learning. In a huge dataset, reduce the dimensions with minimal loss of information. Reducing dimensions will reduce the time and storage space required.
So, the main idea is to look for accurate data representation in a lower-dimensional space.
When anyone interviews you for any data science or Machine Learning role, the 1st question would be: Tell me the techniques you know in Machine Learning. Then you will start answering the question by stating many algorithms you have studied and worked upon. The interviewer will pick some keywords when you are explaining the concept and ask you things related to that.
I will also explain things in that way so that we can connect the dots.
Data Science || Machine Learning || Business Analytics