Predictive analysis is a group of analysis which uses machine learning, data mining, statistical algorithms to analyze data to make predictions. The goal of predictive analysis is to cross the capability of descriptive statistics and reporting and provide the best assessment on what will happen in the future. The end result helps in streamline decision making and produce new insights. Predictive models discovers patterns, identify relationships between different factors, associated risks or probabilities from data.
Types of Predictive Analysis
While discussing about predictive analysis, we usually point to the predictive models and forecasts. But the usage of the terminology to related to widespread analytical disciplines. We can broadly divide in to three classes :
Predictive models : relationship between the specific performance of a subject in a sample and one or more attributes or characteristics of the same subject. It focuses on predicting the behavior of a target.
---
Descriptive models : quantify the relationships between the data so that it is often used to classify into groups. It identifies different relationships between two or multiple things.
Decision models : describe the relationship between all the elements of a decision, development of a logical decision to reach the desired result for a circumstance.
Application of Predictive Analysis
Below are some of the use cases of predictive analysis.
CRM or customer relationship software : CRM can use predictive analysis to create marketing, sales or professional services campaigns.
Healthcare : may determine which patients are at risk of developing certain disorders such as diabetes, asthma or cardiovascular diseases among the chronic diseases.
Customer Loyalty : companies need to focus their efforts on maintaining the continuity of consumer satisfaction rewarding loyalty. Small increases in customer retention results a much higher benefit. 5% increase in the retention percentage has a 25% to 95% increase in benefits.
Direct Marketing : predictive analysis can help by identifying the most effective combinations of products, marketing material, communication channels and time margins that should be considered for a targeted consumer.
Fraud Detection : predictive model can help reduce or eliminate the risk of companies of fraud identifying high risk matters including the high-risk related personal accounts, potentially fraudulent travel agents and suspicious suppliers.
Tagged With characteristics of predictive analysis , is predictive analysis part of data analysis