The Cleansing, Transforming and Modeling of data to extract insights form data for making decisions is called Data analysis or Data Analytics. For example Uber is going to launch new service in Pakistan. For this they will analyze their data, from data they will extract useful information. Based on these insights / Information they will make decision against or in favor of service. Now let’s go one step ahead, they are trying to predict future trends. For example “WHAT-IF UBER START SERVICE XYZ”. This scenerio can be solved by using advance statistical models and Machine Learning techniques (List of ML Techniques). They might use Big Data systems; Analysis on such scale is called as Advanced Data Analysis or Advanced Data Analytics.
Types of Advanced Data Analytics
Data Analysis; as discussed is used to assisting human experts for decision making. Now the decision can be one of the following:-
- What is happening?
- Why an event is happening?
- What is possibility of an event to happen?
- What do I need to do this Action?
There are 4 types of Advanced Data Analytics
- What is happening? ——– Descriptive Analytics
- Why an event is happening? ——– Diagnostic Analytics
- What is possibility of an event to happen? ——- Predictive Analytics
- What do I need to do this Action? ——– Prescriptive Analytics
Amazon decision makers wants to know what happened during the months of COVID-19. They collect raw data from different sources. Then create dashboards explaining different features and their correlations. They understand the demographics information of their customers such as age, country and sex etc. These visualizations will depicts insights from the data. Based on this data decision makers in Amazon will set goals for next month. This type of analysis is known as Descriptive Analytics. Insights that managers of Amazon extracts from data using visualization can be following:-
- Analyzed monthly revenue and income per product group
- Which Customer Sex buy more product
- Country wise sale of products
- Total quantity of each item delivered per week
- Timings on which product was demanded by customer
- What types of product were observed out of order
- What was the relation between different products
Descriptive analytics’ findings simply signal that something is wrong or right, without explaining why. For this reason, our data analyst do not recommend highly data-driven companies to settle for descriptive analytics only, they’d rather combine it with other types of data analytics. In business it provides the analyst a view of key metrics and measures within the business.
Alibaba’s manager at US region set their sales targets for 2020. Now in the mid year 2020, they didn’t make sales as per expectations. They wants to know why we miss those targets. They will use historic data and different measures from data to find answer of “Why we missed targets?”. Diagnostic Analysis is solution to this problem. After effective Descriptive data, diagnostic analytical tools will empower an analyst to drill down and in so doing isolate the root-cause of a problem.
Google Adsense place ads on web page, and they want to know what will be the best position for placing ads and appropriate ads. Best place is where maximum users click on ads and appropriate ads means ads of user interest. Now this is something predicting future trend. Finding the possibility or likelihood of events (click or not) is called Predictive analytics. Findings of Descriptive Analytics and Diagnostic Analytics will help us to detect clusters (segments) leads to effective Predictive Analytics. Predictive analytics belongs to advanced data analytics types and brings many advantages like sophisticated analysis based on Machine Learning or Deep learning and Proactive approach that predictions enable. The results of forecasting or Predictive Analytics depends on the quality of data.
HBL Bank wants to known what actions (products and offers) will bring more customers and how HBL retain their old customers. Now this actually predicting the future and eliminate the future problems. Prescriptive Analytics is the answer to these questions. Prescriptive analytics is most complex as compare to other types and it uses advanced tools and technologies, like machine learning (Deep Learning), Business Rules and Algorithms. Besides, this state-of-the-art type of data analytics requires not only historical internal data but also external information.