Data Science

Data science can be understood even to those who are new to it, since it does not require programming, math or even past experience to understand how it is done.  The taxonomy of data starts with getting more data, which involves things that can be measured such as numbers, temperatures, names and types. They are things that if they are changed in any way, they can lead to different results. Data are of different types such as names that can be turned into numbers and that appear in a defined order. The process of organizing and transforming data into a definite and useable form is referred to as data engineering (Rohrer, 2016).

After obtaining new data, it is important to ask a sharp question which must be answered using a number or name. To get answers that are in quantity form such as expected sales volume, the market share and products can only be answered through data analysis. The data that is used to explore the relationships between various factors such as how market share relates to sales is referred to as target data.  It is possible to have a question that one does not have target data for and in this case it is necessary to get more data that addresses that specific question.  After getting the data and asking a sharp question, the data has to be put in a table. The data in a table has to be aggregated, distributed, computed and then measured. Estimation can then be done to get a reasonable guess of a specific expected outcome such as stock price, market share and total users of a product (Rohrer, 2016). This process allows for easy analysis of the entire target data collected.

Reference

Rohrer, B., (2016).Data Science for Absolutely Everybody. Retrieved from: https://channel9.msdn.com/Events/Machine-Learning-and-Data-Sciences-Conference/Data-Science-Summit-2016/MSDSS02