Please enable JavaScript.
Coggle requires JavaScript to display documents.
[AADS] Chapter 1: Introduction to Data Analysis - Coggle Diagram
[AADS]
Chapter 1: Introduction to Data Analysis
1. Characteristic of Data
Variety
Variety of new data sources, formats and structures
eg: photo, video
Velocity
High velocity data with rapid data ingestion and near real time analysis
eg: streaming data
Volume
Billions of rows and millions of column
eg: sensor data
2. Type of Data
Semi Structured
form of structured data that does not obey the tabular structure of data models
contains tags or other markers to separate semantic elements
eg: HTML code
Structured
Data conforms to a data model and has easily identifiable structure
Data is stored in the form of rows and columns
eg: Database
Quasi Structured
consists of textual content with inconsistent data formats
eg: data about webpages a user visited and in what order // clickstream data
Unstructured
data that not defined
It can be textual or non-textual.
eg : email, social media data
3. Type of Analysis in Data
Cognitive Analytic
Question
: "What is the best action?"
Purpose
: to blend traditional analytics techniques with AI and ML features for advanced analytics outcomes.
Example
: Customer Support - Siri (technology with cognitive computing abilities to understand natural language)
Techniques in Cognitive Analytic
: blend of artificial intelligence (AI), machine learning (ML), deep learning (DL), and semantics
Diagnostic Analytic
Question
: “Why did it happen?”
Purpose
: search for insights, and uncover the reasoning behind certain results.
Example :
make root cause analysis
Techniques
: drill-down, data discovery, data mining and correlations.
Predictive Analytic
Question
: "What could happen?"
Purpose
: seeks to predict what is likely to happen in the future based on past patterns and historical data.
How
: use the relationship between a set of variables to make predictions
Example
: sales prediction (predict the revenue sales for the following tenure / quarters)
Techniques in Predictive Analytic
: classification - uses logistic regression algorithm
Extra:
Machine learning is a branch of predictive analytics - designed to recognize patterns and evolve to make accurate predictions
Prescriptive Analytic
Question
: “How can we make it happen?”
Purpose
: to suggests various courses of action and outlines what the potential implications would be for each.
Example :
action plan of what that have we predicted (Google Maps - consider all the possible modes of transport)
Techniques
: Advanced algorithms to test potential outcomes of each decision (involving algorithms, machine learning, statistical methods, and computational modeling procedures)
It based on current data analytics, predefined future plans, goals and objectives.
Descriptive Analytic
Question
: "What has happened?"
Purpose
: to simply describe what has happened
Example
: Google analytics (provides a simple overview of what’s been going on with your website)
Techniques in Descriptive Analysis
: Data aggregation and data mining