ISSS608 2017-18 T3 Assign LUO Haoran Data Preparation

VAST Challenge 2018 MC3:
Catch the Unsub – Who hurts Eurasian Pipit?

Overview

Data Preparation

Methodology

Insights

Raw Data Overview

1. Event Data Tables Overview

• In total there are 9 event data tables, recording calls, emails, purchases and meetings across the company.
• Insider has already marked out suspicious calls, emails, purchases and meetings and made them isolated data tables.
• No column name for all 9 data tables. However, the format of them are the same.
  • 1. Source: contains the company ID# for the person who called, sent an email, purchased something, or invited people to a meeting
  • 2. Etype: contains a number designating what kind of connection is made.
    • a. 0 is for calls
    • b. 1 is for emails
    • c. 2 is for purchases
    • d. 3 is for meetings
  • 3. Destination: contains company ID# for the person who is receiving a call, receiving an email, selling something to a buyer, or being invited to a meeting.
  • 4. Time stamp: is in seconds starting on May 11, 2015 at 14:00.

2. Company Index Overview

• The company index shows the name of everyone in the company and their associated ID#. There are 642,631 individuals in the index.

3. Letter from Insider

• There is a letter from the insider in which he/she shares a suspicious employee list.

Data Aggregation

1. Add Column Name

• Applied the same command to 9 event data tables to add column names.
• Corresponding Column Name: Source, Etype, Destination, TimeStamp

2. Convert TimeInterval to TimeStamp

• Changed TimeInterval in original data tables into TimeStamp in format of dd/mm/yyyy hh:mm:ss.
• Initial time is 11/05/2015 14:00:00.

3. Save New Data Tables

• Saved new data tables with TimeStamp to the path.

4. Data Modification in JMP

• Relabeled values in Etype column.
• Added names for sources and targets according to SourceID (company index) and TargetID (company index).
• created new columns indicating different time dimensions.

5. Create Nodes and Edges Data Using Suspicious Employee List

5.1 Edges_Suspicious Event Data

• Select out all events initiated or accepted by the suspicious employees.
• Concatenate data tables together and get a merged data table as Edges_SusEvent.

5.2 Nodes_Suspicious Employee Data

• By using the Edges table created, all the employees involved in the suspicious behavior can be filtered out.
• Except for 20 employees the insider points out, there might be more people doing bad things.