Cyberlium
Cyberlium
Cyberlium on Android·Google Play Store

Learn cybersecurity with hands-on labs and AI mentor support on your phone.

Digital › Module 7 › Lesson 2

BeginnerModule 7Lesson 2/5

Correlation

Correlation literacy — join PCAP, logs, disk, and memory rows on UTC and entity keys from $DFIR_LAB practice cases.

15 min+39 XP3 quiz
Module progress2 of 5

Visual · t26_correlation

Correlation = join keys across artifacts. $DFIR_LAB. Original Cyberlium.

Opening

Correlation is where scattered artifacts become a story — join with keys and skepticism, not with unauthorized data.

Correlation literacy joins artifacts via UTC window, hostname, username, IP:port, file hash, process name, and session ID — document join key per row on $DFIR_LAB practice evidence. Example literacy: DNS query in PCAP within ±60s of 4688 process start in log and file write MAC on disk image — three-source row with confidence note (strong/medium/weak). Cyberlium teaches join methodology — NOT assuming causation from coincidence, NOT correlating employer prod logs with lab fiction, NOT 'hacking back' to validate IPs. Purple analyst marks conflicting rows instead of deleting them. Output: correlation matrix or three cross-source rows with join keys explicit.

1. Join keys

UTC ± window, IP, user, host, hash, PID — pick key per pair; document in column.

Weak joins labeled weak — do not upgrade to 'confirmed attacker' without evidence class.

Command guide

Try these commands — Join keys

═══ LINUX / macOS (Malware Static & Dynamic Triage) ═══

Inspect PE / ELF binary headers and sections

Command — copy this

readelf -h sample.bin 2>/dev/null || xxd -g 1 -l 32 sample.bin

Check binary entropy (high entropy often signals packed or encrypted payloads)

Command — copy this

python3 -c "
import math, sys
data = open('sample.bin', 'rb').read() if len(sys.argv) > 1 else b'MZX\x00\x90'
ent = -sum((data.count(bytes([b]))/len(data))*math.log2(data.count(bytes([b]))/len(data)) for b in set(data))
print(f'Calculated File Entropy: {ent:.4f} (Entropy > 7.0 suggests packed/encrypted)')
" sample.bin 2>/dev/null

Primary tools to practice this lesson: grep, python3. Reference sites: Event correlation (https://www.sans.org/cyber-security-skills/digital-forensics/); CISA IR (https://www.cisa.gov/topics/cybersecurity-best-practices/incident-response); NIST CFReDS (https://www.nist.gov/itl/ssd/software-quality-group/computer-forensics-reference-data-sets-cfreds). Run every command in the box — install first, then the usage lines — only on YOUR lab / program scope.

2. Multi-source rows

Minimum literacy: two sources correlated; stretch: three (net + log + disk).

Module 4 disk and Module 5 memory rows may enter joins when brief provides images.

3. Refused

No correlation using unauthorized evidence, no deletion of contradicting rows.

Correlation supports hypotheses — does not replace legal reporting standards Module 8.

4. What you ship: correlation row template

Join key glossary + three cross-source example rows + confidence column + scope line.

5. What you record before the next lesson

Correlation row template path.

6. Wrong vs right: tampering evidence vs chain of custody

Worked failure — same MSF word, opposite target. Right never needs a café Wi-Fi or classmate laptop.

  • Wrong

    Delete log row that contradicts your theory before submitting timeline.

  • Right

    Correlation template with explicit join keys on $DFIR_LAB. Next: Hypotheses.

Mission: correlation row template

1) List five join keys. 2) Define confidence weak/medium/strong. 3) Write two cross-source rows from practice case. 4) Write NEVER delete contradicting rows.

Stuck? Ask Cyberlium AI Mentor

Ask Mentor: “±60s window — when too wide?”

Knowledge Check

1

APPLY: Correlation uses:

Multiple choice

Knowledge Check

2

APPLY: True or False: Delete contradicting rows to fit theory.

True or False

Knowledge Check

3

APPLY: Weak correlation should:

Multiple choice

← Previous

Answer all 3 knowledge checks to continue. (0/3 answered)