Introduction
This article will focus on most frequently used terminologies related to Machine Learning and Artificial Intelligence in a much simpler way.
To elaborate on different terminologies, we need to emphasize on the following chart below.
A bubble chart is created with Business Value on the X-axis and Complexity on the Y-axis. The data point on the bubble chart shows as the progression of business value which you can give to the company, so does the complexity for deliverance in your role also increases.
Reporting
The first term you might come across is Reporting. This is at the lowest end of the Data Science spectrum. These are those kinds of reports which you can automate with minimum effort. Normally, it involves lesser amount of data cleaning function. Additionally, it is related to delivering daily, weekly, monthly, and yearly reporting with normally the same data sources and data structures.
MIS or Management Information Systems is the most basic term being used here. Generally, MIS is used to track the progress of different departments within the organization towards it’s goals by gauging its performances against a pre-defined KPI (Key Performance Indictor). A KPI is a sort of a measurement or a benchmarking tool which exists to analyze the performances of quantitative data against its targets.
Instead of going through heaps of sales data, this cross tab shows how much business was source and what was the rejection rate in terms of quantitative integer and percentage wise data. Additionally, conditionally formatting in terms of coloring the background color of the cells makes the manager’s life easier by helping them to conclude that Delhi’s North Region didn’t perform as expected from them by scoring more than 13 percent rejection rate from the business sourced from their sales representatives.
Dashboard
The next item you might have heard is Dashboard or Business Intelligence or BI. It is the utopia of the world. Imagine a world where every action about your business is reflected in front of screens of your business executives. That is business intelligence for you. Dashboards are used to answer, “What is happening now in my business?”.
Here’s an example of a Dashboard developed by me few months earlier.
The following Dashboard developed from Google Analytics shows the democratic of the website by revealing the number of users visiting the website.
Predictive Modeling
If we look up to find out what is Predictive Modeling, the easier to understand definition that comes up is:
Well, when you gather all the data to predict what is likely to happen at a granular level, this is called predictive modelling.
For example, predictive modeling helps us to predict which customer is likely to default in the next 30 days.
To do this, we need to gather up all the historical data about the customers of the bank by analyzing trends and insights to predict which customers are likely to default in near future. Please note that it happens at a very granular level.
Forecasting
The next common terminology is Forecasting:
Forecasting is a process of estimating the future based on past as well as present data.
It is normally done at the aggregate level. For example, how many customers will fly on a particular flight, or many sales of a particular product we can expect in the next week?
This is how the outcome of a forecast would look like typically-
The line graph above is the perfect example of as to how easy it comes for forecasting using tools like Microsoft Excel, Tableau or Power BI.
In this graph, the dates or equal interval is plotted on the X-Axis and count of passengers in on the Y-Axis. Legends are a clever way of describing that blue line shows the historical data and the orange line shows the forecast of the number of passengers, calculated from the historical data of the passengers who flew from one destination to the another with this airline. The confidence level is nothing but a graphical representation of floor and ceiling values of the forecasted quantitative data.
Machine Learning
Now, the hottest terminology used in the Data Science world today is Machine Learning.
Basically, what is Machine Learning?
The machine learning is a system of teaching machines to learn things and improve predictions or behavior, based on data on their own.
The above definition means that Machine Learning is nothing but enabling an algorithm to learn to predict the result of the next dataset, after it has analyzed enough labeled or unlabeled historical datasets.
There are two types of Machine Learning algorithms
- Supervised Algorithms:
These algorithms are the most used ones by the Data Analysts and Scientists. In fact, more than 90% of the analytical work is accomplished using them as they are easier to implement, and the results are easier to interpret. It is imperative to feed them with labeled datasets as they are unable to work without it. Some of the supervised algorithms are linear regression, non-linear regression, classification, etc.
- Unsupervised Algorithms:
These algorithms are not much in numbers as they are hard to implement. Plus, they require unlabeled data. So, it means that you do not have that much control or supervision over the data as the algorithm analyses it on its own. Examples are K-means clustering and Apiori algorithm.
Artificial Intelligence
And the final word in a spectrum is Artificial Intelligence. The applications of AI include chatbots and robots. There is a plethora of other applications in use today which fall into the bracket.
