Imagine a world without data, where there is no concept of Social Media, No Netflix, No Online Shopping. Would you live in that?
It’s even hard to imagine that, right?
That is the importance of data in our daily life, Because data is everywhere and it’s mandatory for all kinds of businesses which gain insights from data and pivot their strategy depending on those insights.
Data Scientists are the unsung heroes of our daily life dopamine, who leverage their skills in the collection, cleaning and analysis of data to extract valuable insights and tell the algorithms to show us what our desires are, Good or Bad, it is another debate.
So, these heroes need to understand the first word of their heroism which is DATA.
There are two types of data on the scale of measurement in data science: Qualitative and Quantitative
Qualitative data is non-numerical data that describes or characterizes something. It can be words, images or videos.
Some examples of Qualitative Data:
Customer satisfaction ratings
Social Media Posts
Nominal data is the most basic type of data. It is qualitative data that can be categorized but cannot be ordered.
For Example: Eye Colour, Gender and country of origin, Customer ID
The type of Qualitative data that can be ordered.
For Example: A survey ranking of customer satisfaction from 1 to 5, education level
Quantitative data is numerical data that can be measured and counted.
Some Examples of Quantitative Data
The type of Quantitative Data that can be counted.
For Example: The number of children in the family or the number of goals scored in a soccer match
The type of Quantitative data that can be measured.
For Example: A person’s height or weight of a parcel, the customer’s age, and product price.
A better understanding of data: Classifying data types helps us understand what data can be used for. For Example: if we know that a piece of data is continuously numerical, operations like mean and standard deviation can be done on it.
Choosing the right tools: Different tools and algorithms are used for different types of data. If we data is already classified, we can pick the right tools to perform operations on the data.
Improvement in Efficiency: Understanding data types can help improve the efficiency of our model because we are less likely to stray away from the path as we already know what we should be doing with the data.
Effective Communication: Proficiency in getting insights from data from relative data types brings an effective description of our findings as we can clearly communicate the process involved in our findings.
A data scientist working on a fraud detection model will be using nominal data such as transaction type to identify fraudulent activities.
A marketing analyst working on a customer segmentation campaign will be using ordinal data such as customer purchase history to identify possible customer segmentation.
In a product recommendation system, a data scientist will possibly be using discrete numerical data such as the number of times a customer has viewed some type of product to recommend him.
A data scientist working in a financial company might be working on a risk assessment model for certain investments, and he will be using numerical data like customer age or income to assess the risk of defaulting on the loan.
Understanding the different data types is mandatory for data scientists. By ensuring your knowledge of data types, you can use the right tools and algorithms that will improve the efficiency and accuracy of the machine learning model.
More core concepts about Data Science and Machine Learning Models are on their way. So, Follow for more.
Thanks for reading, Happy Learning!