Business Inteligence Techniques

To start let's explain what business inteligence is. It can be explained as a mixture of data skills, business knowledge, business intuition, and explanation of past performance.

There are 6 steps to this process. First comes the observation, which basically means the gathering of the data. For example the sales per month or new customers per month. The next step is to quantify the data which means representing the observations with a numerical value. Then we have to accumulate the observations to show information, this step is called the measure. As our 4th step comes building the Metric to do so we have to give the measure a business meaning to gain business performance or progress. Like in the 4th step we have to use the metric plus a business objective to create the KPI (or Key Performance Indicator)and these act as methods to achieve your business goals. The last step is to use the KPI to visualize our results with BI reports or BI dashboards which consist of graphs or diagrams to explain the data.

Predictive Analytics Traditional Methods Techniques

The goal of the traditional methods is to assess potential future scenarios by using advanced statistical methods. One of these methods is called regression which is a model used to quantify casual relationships among the different variables incñluded in your data set. There are different types of regression. For example, Logistical Regression which operates with logical values like 0s and 1s in a graph. Another type is cluster analysis which looks for groups of similar data in our data pool.

Machine Learning Types

The goal of creating an algorithm, which a computer then uses to find a model that fits the data as best as possible. And makes very accurate predictions based on that. The most important thing to remember is Machine Learning is that you give the model no instructions just an algorithm that gives it directions. This algorithms are usually a trial-and-error process of continuous trails that are at least as good as the previous one.

There are 3 types of Machine learning: First, Supervised learning which reasembles a teacher teaching his students. Second, Unsupervised learning in wich data us used and the model learns by itseld by looking at the patterns. Third, Reinforces Learning Wich rewards the best patterns and dosent give any reward if its worse.