Introduction
This is the third post in a series of posts discussing the types of analytics. They are as the following: Exploratory, Descriptive, Predictive, and Inferential. This post will be focusing on Predictive analytics which as defined by the article “Predictive Analytics: Techniques and Applications” by Neelam Tyagi is an advanced form of data analytics, understands historical data behaviour and anticipates data-driven insights as future outcomes(the article can be found here). In other words, predictive analytics is using historical data to make predictions about the future. It is used a lot in many industries from healthcare to finance and has helped companies prepare for their future endeavors in a very systematic, data-driven way.
Summary of Article
First, the article gets into the definition of predictive analytics and then shows the techniques that predictive analysis uses(data mining, statistical modeling, and machine learning tools). All of these are used, in essence, to determine the relationships within the historical data and then use those to forecast the outcomes in the future. Then the article gets into why predictive analytics is important and it cites that predictive analytics can be used for a plethora of business practices. Practices such as fraud protection, optimization of campaigns, minimization in risk, and improvements in business operations. All of these are important to businesses and if they are more successful it can really help boost a business to new levels. Finally, the article discusses the applications of predictive analysis in the world today, and here are a couple of them:
- Marketing – predictive analytics can help with leads, segmenting the market, and can help identify which consumers are expected to purchase which product
- Retail – predictive analytics can help manage inventory and logistics.
- Manufacturing – predictive analytics helps operate and optimize the manufacturing process at all stages.
- Healthcare – predictive analytics can help save money and improve healthcare efficiency.
- Finance – predictive analytics can help minimize risk and improve customer satisfaction
My Take
All things considered, this article was a good one. It gave me an overview of predictive analytics and gave me some examples of predictive analytics. One thing I would have liked, however, is the addition of more technical content. What are some types of statistical models used, what are some types of machine learning models used, what are some software used to perform all of this, etc. I also would have liked a case study in which the author took us through an iteration of predictive analysis with a dataset to show us how it is done. In my opinion, this would have really helped the reader understand the content and show the reader how to use this type of analysis in their own workflow. I did really enjoy the detail and the content about the real-world application of this and it really showed me where it is used in the real world and how companies have adopted it in a lot of the major sectors of the economy. I also really liked the links to further reading and the use of a video which gave another means of presenting content to the reader. Another thing that I felt the article did well was the way it was written with the formatting playing a key role in helping the reader understand the content. The use of good formatting really supplemented the lack of visuals presented to the reader.
Conclusion
All in all, it was a good article. Its combination of different media forms, as well as the detail on the application of predictive analysis made it a good read. Although I would have liked a larger focus on the technical aspect it was nonetheless a worthwhile article. I recommend you read it when you get the chance(the article can be found here).