addition of unit 1 3 4 5
This commit is contained in:
19
unit 5/00_Index.md
Normal file
19
unit 5/00_Index.md
Normal file
@@ -0,0 +1,19 @@
|
||||
# Unit 5: Advanced Data Mining Techniques
|
||||
|
||||
Welcome to your simplified notes for Unit 5.
|
||||
|
||||
## Table of Contents
|
||||
|
||||
1. [[01_Ubiquitous_Data_Mining|Ubiquitous & Invisible Data Mining]]
|
||||
- Mining everywhere (IoT, Mobile)
|
||||
- Invisible Mining (Background processes)
|
||||
2. [[02_Web_Mining|Web Mining]]
|
||||
- Content, Structure, and Usage Mining
|
||||
3. [[03_Spatial_and_Temporal_Mining|Spatial & Temporal Mining]]
|
||||
- Mining location data (Maps/GIS)
|
||||
- Mining time-based data (Trends)
|
||||
4. [[04_Other_Mining_Types|Other Mining Types]]
|
||||
- Text, Visual, Audio, and Process Mining
|
||||
5. [[05_Applications_and_Impact|Applications & Social Impact]]
|
||||
- Real-world uses (Healthcare, Retail)
|
||||
- Privacy and Ethical concerns
|
||||
30
unit 5/01_Ubiquitous_Data_Mining.md
Normal file
30
unit 5/01_Ubiquitous_Data_Mining.md
Normal file
@@ -0,0 +1,30 @@
|
||||
# Ubiquitous and Invisible Data Mining
|
||||
|
||||
## Ubiquitous Data Mining (UDM)
|
||||
**"Ubiquitous"** means existing everywhere.
|
||||
- **Definition**: Mining data from everyday objects and devices (Smartphones, IoT, Wearables) in real-time.
|
||||
- **Goal**: To provide insights anytime, anywhere, without you asking for it.
|
||||
- **Characteristics**:
|
||||
- **Mobile**: Uses GPS and sensors.
|
||||
- **Context-Aware**: Knows where you are and what time it is.
|
||||
- **Real-Time**: Processes data instantly.
|
||||
|
||||
### Examples
|
||||
- **Smartphones**: Google Maps predicting traffic.
|
||||
- **Wearables**: Smartwatches tracking your heart rate.
|
||||
- **Smart Homes**: Alexa learning your voice commands.
|
||||
|
||||
## Invisible Data Mining
|
||||
- **Definition**: Mining that happens **silently** in the background. You don't see it happening.
|
||||
- **Why "Invisible"?**: It is embedded in apps and systems. You only see the result (like a recommendation).
|
||||
- **Examples**:
|
||||
- **Amazon**: "People who bought this also bought..."
|
||||
- **Google Search**: Auto-completing your sentence.
|
||||
- **Banks**: Detecting fraud without you knowing.
|
||||
|
||||
### Difference
|
||||
| Feature | Ubiquitous Mining | Invisible Mining |
|
||||
|---|---|---|
|
||||
| **Focus** | Mining **everywhere** (IoT, Mobile) | Mining **hidden** from user |
|
||||
| **Awareness** | You might know it's happening (e.g., wearing a watch) | You usually don't know |
|
||||
| **Key Tech** | Sensors, Mobile Devices | Software Algorithms, Background Processes |
|
||||
20
unit 5/02_Web_Mining.md
Normal file
20
unit 5/02_Web_Mining.md
Normal file
@@ -0,0 +1,20 @@
|
||||
# Web Mining
|
||||
|
||||
**Web Mining** is using data mining techniques to discover useful information from the World Wide Web.
|
||||
|
||||
## Types of Web Mining
|
||||
|
||||
### 1. Web Content Mining
|
||||
- **What**: Mining the **actual content** of web pages.
|
||||
- **Data**: Text, images, audio, video.
|
||||
- **Example**: Analyzing reviews on Amazon to see if people like a product (Sentiment Analysis).
|
||||
|
||||
### 2. Web Structure Mining
|
||||
- **What**: Mining the **links** (hyperlinks) between pages.
|
||||
- **Goal**: To find important pages (Authorities) and pages that link to many others (Hubs).
|
||||
- **Example**: Google's **PageRank** algorithm uses this to rank search results.
|
||||
|
||||
### 3. Web Usage Mining
|
||||
- **What**: Mining **user activity** logs.
|
||||
- **Data**: Server logs, browser history, clicks.
|
||||
- **Example**: Analyzing which pages users visit most often and where they leave the site.
|
||||
17
unit 5/03_Spatial_and_Temporal_Mining.md
Normal file
17
unit 5/03_Spatial_and_Temporal_Mining.md
Normal file
@@ -0,0 +1,17 @@
|
||||
# Spatial and Temporal Data Mining
|
||||
|
||||
## Spatial Data Mining
|
||||
- **Spatial Data**: Data related to **location** or geography (Maps, GPS).
|
||||
- **Goal**: Finding patterns in space.
|
||||
- **Tools**: GIS (Geographic Information Systems).
|
||||
- **Examples**:
|
||||
- Finding the best location for a new store.
|
||||
- Tracking the spread of a disease on a map.
|
||||
|
||||
## Temporal Data Mining
|
||||
- **Temporal Data**: Data related to **time**.
|
||||
- **Goal**: Finding patterns that change over time (Trends).
|
||||
- **Tasks**:
|
||||
- **Trend Analysis**: Is the stock market going up or down?
|
||||
- **Sequence Analysis**: "If event A happens, does event B follow?"
|
||||
- **Example**: Analyzing weather patterns over 10 years to predict climate change.
|
||||
17
unit 5/04_Other_Mining_Types.md
Normal file
17
unit 5/04_Other_Mining_Types.md
Normal file
@@ -0,0 +1,17 @@
|
||||
# Other Types of Data Mining
|
||||
|
||||
## 1. Text Mining
|
||||
- **Data**: Unstructured text (Emails, Tweets, Documents).
|
||||
- **Technique**: Natural Language Processing (NLP).
|
||||
- **Goal**: To understand meaning, sentiment, and topics.
|
||||
- **Example**: Classifying customer feedback as "Angry" or "Happy".
|
||||
|
||||
## 2. Visual and Audio Mining
|
||||
- **Data**: Images, Videos, Sound.
|
||||
- **Goal**: To find patterns in visual or audio data.
|
||||
- **Example**: Face recognition in photos, or detecting keywords in a voice recording.
|
||||
|
||||
## 3. Process Mining
|
||||
- **Data**: Event logs from business systems (ERP, CRM).
|
||||
- **Goal**: To see how a business process *actually* works vs how it *should* work.
|
||||
- **Example**: Finding out why it takes 5 days to approve a loan instead of 2.
|
||||
22
unit 5/05_Applications_and_Impact.md
Normal file
22
unit 5/05_Applications_and_Impact.md
Normal file
@@ -0,0 +1,22 @@
|
||||
# Applications and Social Impact
|
||||
|
||||
## Applications of Data Mining
|
||||
Data mining is used everywhere!
|
||||
|
||||
1. **Healthcare**: Predicting diseases, finding side effects of drugs.
|
||||
2. **Retail (Market Basket Analysis)**: Placing Bread near Butter to increase sales.
|
||||
3. **Finance**: Detecting credit card fraud, approving loans.
|
||||
4. **Education**: Tracking student performance to help them improve.
|
||||
5. **Crime**: Identifying crime hotspots and predicting criminal behavior.
|
||||
|
||||
## Social Impact and Issues
|
||||
|
||||
### Positive Impact
|
||||
- **Convenience**: Personalized recommendations save time.
|
||||
- **Safety**: Fraud detection and medical diagnosis save money and lives.
|
||||
|
||||
### Negative Impact (Ethical Issues)
|
||||
1. **Privacy Invasion**: Companies know too much about you.
|
||||
2. **Discrimination**: Profiling can lead to unfair treatment (e.g., denying loans based on where you live).
|
||||
3. **Security**: Large databases can be hacked (Data Breaches).
|
||||
4. **Manipulation**: Targeted ads can influence your behavior or political views.
|
||||
Reference in New Issue
Block a user