I am working for the first time on building a unsupervised learning model to detect suspicious transactions using DBSCAN, Do I train the model on all data columns(Columns like account number, transaction date, transaction details ) or only on the ones that I want to detect suspicion (money withdrawal and deposit columns)?
This is the data set that I am working on https://www.kaggle.com/apoorvwatsky/bank-transaction-data
Please help me
