Help needed on making graph database of a fraud detection system

Hi all Seniors

I m Samveda currently started learning graph database. Gone through Packt book on graph database & studying O`reilly book on same topic.

I m working/learning on a fraud detection system as a student. I have four column heading in excel/CSV file

  1. From
  2. To
  3. Amount
  4. Data

Now in the start , the fraud started from 3 origin accounts which i marked as Origin_1,origin_2, origin_3.

Now the amount from these 3 origin accounts went to further 13 accounts. From these 13 , it went to further 47 accounts. Now these 47 accounts have the original money as well as their personal transactional money. The money at each stage is split and sent in pieces and send on diff days.

From these 47 accounts, it is further sent, rotated on previous accounts making a complex web of transactions. In the end after 1-2 month gap , the money finally lands in 7 destination accounts.

Now the filter i want to apply is +/- 3 days before and after of transaction.
All my data is in excel/CSV sheets. My main concern is filter by "date of flow of amount" or filter by amount

Can anyone guide how to build relationships in cypher and create neo4j database of it so that i can visualize in Neo4j browser

If you can share some sample data,that would be helpful in giving some guidance.