Question

You are the data analyst on the project team building a data warehouse for an insurance company. List the possible data sources from which you will bring the data into your data warehouse. State your assumptions. Support with research.

Answer 1

Before thinking of harvesting the data and looking for the source we should think of what kind of data we are looking at and how it is going to benefit us

assumption: i am assuming the company  to be a general insurance company , but while defining the data i will give examples of the specific types of the companies relying on those type of data , the stats i will be providing are all from researches

first of all lets look at how data can help us . Data can help us in taking various decision .

• assessing the risk : for eg we can analyse whether a person buying a car insurance is prone to accident risk base on his driving decision , or whether a person buying health insurance is prone to Arrhythmia or not
• detecting fraud : so we can analyse previous harvested records of people who have taken insurance to detect a fraud pattern
• customers behaviour and their Insight : these help in shaping customer policy and decision making , we know their behaviour and the basis of their decision .
• Marketing : once we know the customer behaviour and their segmentation we can use that to target our product and services , based on their behaviour , products can also be personalised  for example targeted ads on facebook(Trattner and Kappe)

now we know what kind of data we need now we will see their sources

lets say we are a health insurance based company

1. assessing the risk : first of all for accessing the risk we will need clients medical record , this can be taken either from the client or from Medical association . we can also use fitbit devices to take physical activity reading from customers . Use of data in risk analysis is going to grow to 77% by 2021
2. detecting fraud: for detecting fraud we will need fraudulent data from various other insurance companies and use those data to determine whether a customer is fraudulent or not , we can also use the social network data of those fraud people to determine risk using statistical measure . these researches are based on fraud detection .Identify the signs of fraudulent accounts and the patterns of fraudulent transactions(Quah and Sriganesh), Identification of fraudulent financial statement (Kirkos)
3. customers behaviour and their Insight : for this we can have an app on customer device through which we can take permission and accumulate user data to provide him with personalised experience , this data can include user gps log or his social network activity and etc but this should not affect users privacy . Assess disease outbreaks from tweetsAssess disease outbreaks from tweets(Bodnar and Salathé) , Detect public health events(Fisichella)
4. this is also done through taking customerpersonal data over their choices and preferences , this data can be taken through application or web interface to provide them personalised  experiences (Viral marketing in social networks mar at el)
5. other than this we can also take data from private sectors and academic researchers . we can also take data that are harvested by government.

all his data can be imported in data hive or a database through various channel some of which can be IOT based like fitbit some of can be web and app interface by using firebase db and some can be direct interactive device based .we can even use the old data stored by the company by importing it into our data base and digitalizing it