vault backup: 2024-10-17 23:28:49

.obsidian/workspace.json

@@ -21,7 +21,7 @@
             "state": {
               "type": "markdown",
               "state": {
-                "file": "Data Structures/Week 5/Lecture 9 - Stacks.md",
+                "file": "AI & Data Mining/Week 5/Lecture 9 - PRISM.md",
                 "mode": "source",
                 "source": false
               }
@@ -79,7 +79,8 @@
       }
     ],
     "direction": "horizontal",
-    "width": 395.5
+    "width": 395.5,
+    "collapsed": true
   },
   "right": {
     "id": "ee8f6df41634f71e",
@@ -95,7 +96,7 @@
             "state": {
               "type": "backlink",
               "state": {
-                "file": "Data Structures/Week 5/Lecture 9 - Stacks.md",
+                "file": "AI & Data Mining/Week 5/Lecture 9 - PRISM.md",
                 "collapseAll": false,
                 "extraContext": false,
                 "sortOrder": "alphabetical",
@@ -112,7 +113,7 @@
             "state": {
               "type": "outgoing-link",
               "state": {
-                "file": "Data Structures/Week 5/Lecture 9 - Stacks.md",
+                "file": "AI & Data Mining/Week 5/Lecture 9 - PRISM.md",
                 "linksCollapsed": false,
                 "unlinkedCollapsed": true
               }
@@ -135,7 +136,7 @@
             "state": {
               "type": "outline",
               "state": {
-                "file": "Data Structures/Week 5/Lecture 9 - Stacks.md"
+                "file": "AI & Data Mining/Week 5/Lecture 9 - PRISM.md"
               }
             }
           },
@@ -187,40 +188,43 @@
   },
   "active": "3ed058b7ba32ddc0",
   "lastOpenFiles": [
-    "Client Server Systems/PHPStorm Repo/workshop4/Views/index.phtml~",
+    "Pasted image 20241017132059.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/calculator.php~",
+    "Pasted image 20241017132019.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/calculator.php",
+    "Pasted image 20241017131912.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/index.php~",
+    "Pasted image 20241017131740.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/Views/template/header.phtml",
+    "Pasted image 20241017131525.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/Views/template/footer.phtml",
+    "Pasted image 20241017131053.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/js/bootstrap.min.js",
+    "Pasted image 20241017130933.png",
-    "Client Server Systems/PHPStorm Repo/workshop5/js/bootstrap.js",
+    "Networking and Security/Week 4/Workshop 4 - Routing.md",
+    "AI & Data Mining/Week 5/Lecture 9 - PRISM.md",
+    "AI & Data Mining/Week 5",
+    "Penetration Testing/Week 4/Workshop 4 - Google Dorking.md",
+    "Penetration Testing/Week 4/Week 4 - Pre-Engagement.md",
+    "Penetration Testing/Week 3/Lecture 3 - Blue Team.md",
+    "Penetration Testing/Week 5/Lecture 5 - Reconnaisance.md",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/css",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/page1.php",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/page1.php.partial",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/phpinfo.php",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/phpinfo.php.partial",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/Models",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/js",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/index.php",
+    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/index.php.partial",
+    "Data Structures/Week 5/Lecture 9 - Stacks.md",
     "Client Server Systems/PHPStorm Repo/workshop5/images/new_uos_logo.jpg",
-    "Client Server Systems/PHPStorm Repo/workshop5/fonts/glyphicons-halflings-regular.woff",
-    "Client Server Systems/PHPStorm Repo/workshop5/fonts/glyphicons-halflings-regular.ttf",
     "Client Server Systems/PHPStorm Repo/workshop5/fonts/glyphicons-halflings-regular.svg",
-    "Client Server Systems/PHPStorm Repo/workshop5/fonts/glyphicons-halflings-regular.eot",
     "Client Server Systems/PHPStorm Repo/workshop4/images/new_uos_logo.jpg",
     "Client Server Systems/PHPStorm Repo/workshop4/fonts/glyphicons-halflings-regular.svg",
-    "Client Server Systems/PHPStorm Repo/workshop 5/images/new_uos_logo.jpg",
-    "Client Server Systems/PHPStorm Repo/workshop 5/fonts/glyphicons-halflings-regular.svg",
-    "Client Server Systems/PHPStorm Repo/MVCtemplate-24-with-top-navbar/images/new_uos_logo.jpg",
-    "Client Server Systems/PHPStorm Repo/MVCtemplate-24-with-top-navbar/fonts/glyphicons-halflings-regular.svg",
-    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/images/new_uos_logo.jpg",
-    "Client Server Systems/PHPStorm Repo/MVCtemplate-23-24/fonts/glyphicons-halflings-regular.svg",
     "Client Server Systems/PHPStorm Repo/README.md",
     "PHPStorm/README.md",
     "Software Project with Agile Techinques/Week 3/Study Skills - Class Diagrams.md",
-    "Data Structures/Week 5/Lecture 9 - Stacks.md",
     "README.md",
     "AI & Data Mining/Week 4/Tutorial 4 - Nearest Neighbor.md",
     "AI & Data Mining/Week 4/Lecture 7 - Nearest Neighbor.md",
     "AI & Data Mining/Week 4/Workshop 4 - Nearest Neighbor.md",
     "Penetration Testing/Week 3/Workshop 3 - netcat.md",
-    "Penetration Testing/Week 4/Week 4 - Pre-Engagement.md",
-    "Penetration Testing/Week 4/Workshop 4 - Google Dorking.md",
     "Software Project with Agile Techinques/Week 4/Workshop 4 - SPAT.md",
-    "Networking and Security/Week 4/Workshop 4 - Routing.md",
     "Networking and Security/Week 3/Workshop 3 - Subnetting.md",
     "Networking and Security/Week 2/Lecture 3 - IP4 and Subnetting.md",
     "Client Server Systems/Week 3/Workshop 3.md",
@@ -230,9 +234,6 @@
     "AI & Data Mining/Week 3/Tutorial 3.md",
     "AI & Data Mining/Week 1/Lecture 2 - Input and Output.md",
     "AI & Data Mining/Week 3/Lecture 5 - Naive Bayes.md",
-    "Data Structures/Week 1/Lecture 2 - Data Structures & ADTs.md",
-    "Data Structures/Week 1/Lecture 1 - Module Intro.md",
-    "Penetration Testing/Week 3/Lecture 3 - Blue Team.md",
     "Database Systems/Untitled.canvas",
     "Untitled 1.canvas",
     "Untitled.canvas"

AI & Data Mining/Week 5/Lecture 9 - PRISM.md

@@ -0,0 +1,95 @@
+# Covering Algorithms
+
+- Each class in turn; find set of rules covering all examples
+- At each stage, rule is identified that covers some examples
+- Example is covered if satisfying conditions in the antecedent (LHS) of the rule
+- Consider dataset with 2 predicting numeric attributes, and two class values.
+- ![](Pasted%20image%2020241017130933.png)
+
+# PRISM: Simple Covering Algorithm
+
+- Generates rule by adding tests that maximise probability of desired class
+- Similar to situation in decision trees; problem of selecting attribute to split on
+- Each new test reduces rules coverage
+- Rule becomes more specific as tests are added
+- Search strategy is general-to-specific
+- ![](Pasted%20image%2020241017131053.png)
+
+## Selecting a Test
+
+Goal: Maximise probability of desired class
+- $t$ = total number of examples covered by rule
+- $p$ = number of positive examples of the class covered by rule
+- $t - p$ = number of errors made by rule
+- => Select test that maximises ratio $p/t$
+Stop Condition: $t-p=0$
+- $p = t$, $p/t=1$
+- Or, set of examples cannot be split further.
+
+## Example: Contact Lenses Dataset | Class = Hard
+
+### Selecting 1st Test of 1st Rule
+
+- Rule to Seek: If ? { then recommendation = Hard }
+![](Pasted%20image%2020241017131525.png)
+
+#### Modified Rule and Coverage
+
+- Rule with best test added: If astigmatism = Yes { then recommendation = Hard }
+![](Pasted%20image%2020241017131740.png)
+
+### Selecting 2nd Test of 1st Rule
+
+- If astigmatism = Yes and ? { then recommendation = Hard }
+
+![](Pasted%20image%2020241017131912.png)
+
+#### Modified Rule and its Coverage
+
+- Rule with best test added: If astigmatism = Yes and tear rate = Normal { then recommendation = Hard }
+![](Pasted%20image%2020241017132019.png)
+
+### Selecting 3rd Test of 1st Rule
+
+- If astigmatism = Yes and tear rate = Normal and ? { then recommendation = Hard }
+![](Pasted%20image%2020241017132059.png)
+- PRISM will use test with highest sample size, therefore using Myope.
+
+### 1st Rule for Class = Hard
+
+- Final Rule:
+If astigmatism = Yes
+	and tear rate = Normal
+	and spectacle prescription = Myope
+	then recommendation = Hard
+
+$p/t = 3/3 = 1$
+
+# Pseudo-code for PRISM
+
+For each class C
+	Init E to set of training examples
+	While E contains examples in class C
+		Create rule R with empty LHS predicting class C
+		Until p/t=1, do
+			For each attribute A not mentioned in R, and each value v
+				Consider adding condition A=v to LHS of R
+				Select A and v to maximise p/t
+					Break Ties by choosing largest sample
+			Add A=v to R
+		Remove examples covered by R from E
+
+# Separate and Conquer
+
+- PRISM with outer loop removed generates list of rules for one class
+- PRISM with outer loop removed is separate and conquer algorithm
+	- Identify useful rule
+	- Separate examples covered
+	- Conquer remaining examples
+
+# Rule Execution
+
+- Default Rule
+	- If no rules cover example, prediction is the majority class (most frequent in training data)
+- Conflict Resolution Strategy
+	- If more than one rule covers an example, select predicted class with highest recurrance in training data

Penetration Testing/Week 5/Lecture 5 - Reconnaisance.md

@@ -0,0 +1,107 @@
+# Intelligence Gathering
+- More information gathered, more vectors of attack may be able to use
+- Better knowledge of target, more likely to succeed
+- Better target company knows what is common knowledge, better it can prepare.
+
+## Open-source Intelligence (OSINT)
+
+- Gathers information from publicly available sources and analyses it, producing intelligence
+	- May not be up to date, accurate or complete.
+	- Could be deliberately manipulated to provide false intelligence.
+- Many companies may fail to take into account public information, and how it could be gathered, organised and made searchable
+	- Physical (locations / relationships)
+	- Logical (business partners, job openings, meeting minutes, professional licenses)
+	- Org chart (important people)
+	- Electronic (document metadata, marketing information)
+	- Infrastructure (email addresses, technologies used)
+- Many employees fail to realise information published on the public domain about themselves.
+	- Social Media
+	- GDPR gives right to ask to remove.
+
+# Limits
+
+- Gathering information to identify entry points
+	- physical, electronic, human...
+- and try to map out internal structure
+	- physical, network, organisational
+- and external dependencies
+	- outsourcing, financial
+- It does not involve trying to test or use entry points
+	- "potential vulnerability" more interesting
+	- cyclic lifecycle, we can do more recon later
+
+# Levels
+
+- Level 1
+	- Automated tools to gather information
+	- Generally a simple list of what exists
+- Level 2
+	- Combination of tools and manual searching / analysis
+	- Good understanding of physical locations, business relationships, organisation charts, naming policies, etc.
+- Level 3
+	- Heavy use of manual techniques
+	- Deep understanding of business and how it operates
+	- Highly strategic and planned, time consuming
+
+# Considerations in Commercial Pentest
+
+- Keep to RoE
+	- Avoid legal issues and avoid scope creep
+	- Avoid being sidetracked by interesting sideroads
+- Have a Goal
+	- What is relevant to the target you have been engaged to attack
+- Have a deadline
+	- Make sure time allocated to use intelligence
+
+# Passive vs Active Reconnaissance
+
+## Passive
+
+- Collecting data using publicly available information without direct contact with target
+	- Open web resources, public company information
+	- How they operate, how large they are, contact info, etc.
+
+## Active
+
+- Direct interaction with target by any means to gather information
+	- Port scanning, vulnerability scanning, etc
+	- Illegal without permission.
+
+## Semi-Passive
+
+- Collecting data with methods that appear like normal internet traffic and behaviour.
+	- Looking at metadata in published documents and files. Not actively seeking hidden content.
+
+# Semester 1 Assignment
+
+- Choose company
+	- Should be small, but not too small
+	- Likely IT business
+- Passive recon using OSINT sources
+- Include some semi-passive recon
+- Write report, outlining what has been found and why company should be aware.
+
+- Look for:
+	- Corporate
+	- Personal
+	- Technical information
+- http://www.pentest-standard.org/index.php/Intelligence_Gathering
+
+## How to Obtain Information
+
+- Google Dorking, search for information to see who else has it, and what else they have.
+- Information Gathering tools built into Kali
+- Google for OSINT sources.
+- Google Hacking Database (GHDB)
+- Maltego
+- DMitry
+- Dnmap
+- Ike-scan (Discover IPsec VPNs)
+- P0f (Passive traffic fingerprinting)
+
+### Note on Packet Sniffing
+
+- Some tools rely on network inspection between you and target
+- "Active Packet Sniffing" means specific things cause traffic to flow to you
+- "Passive Packet Sniffing" means you inspect the traffic that happens to come past sniffer.
+-