The document discusses spatial data mining and geographic knowledge discovery. It describes motivations like the availability of large geospatial datasets and lack of analysis tools. As a case study, it analyzes a one day taxi trip log with 53 variables through univariate and bivariate exploration, data cleaning, and deriving new variables. It explores various visualizations including interactive brushing, animated maps, trellis maps, and density maps to detect spatial and temporal patterns in the taxi trip data.

1. A Journey through the
Spatial Data Mining and
Geographic Knowledge
Discovery Jungle
Dr. Kam Tin Seong PhD
Associate Professor of Information Systems (Practice)
School of Information Systems
Singapore Management University
E-mail: tskam@smu.edu.sg
Copyright © 2011, SAS Institute Inc. All rights reserved.

2. Content
 Motivations
 Interactive exploratory analysis
 Distribution analysis
 Geographic data visualisation
 Visualising and detecting spatio-temporal patterns
2
Copyright © 2011, SAS Institute Inc. All rights reserved.

3. Motivations
 Availability of massive, high dimensional, and complex
geospatially-referenced data
 General lack of spatial data visualisation and analysis
functions in data analysis software
 General lack of data analytics techniques in
conventional GIS
 There is an urgent need for effective and efficient
methods to visualise and detect unknown and
unexpected information from these massive datasets
3
Copyright © 2011, SAS Institute Inc. All rights reserved.

4. Taxi Travel Log Case Study
 One day taxi travel log – 278676 trips
 Number of variables: 53
4
Copyright © 2011, SAS Institute Inc. All rights reserved.

5. Initial Data exploration: Univariate
 Overview of the data
 Detect outliers
 Missing data
 Identify new variables
5
Copyright © 2011, SAS Institute Inc. All rights reserved.

6. Initial Data exploration: Bivariate
6
Copyright © 2011, SAS Institute Inc. All rights reserved.

7. Data Cleaning and Transformation
 Data cleaning
 Derive new variables: Time interval, travel time etc
7
Copyright © 2011, SAS Institute Inc. All rights reserved.

8. Geographic Data Visualisation
8
Copyright © 2011, SAS Institute Inc. All rights reserved.

9. Visualising and Detecting Spatio-temporal
Patterns with Interactive Brushing
9
Copyright © 2011, SAS Institute Inc. All rights reserved.

10. Visualising and Detecting Spatio-temporal
Patterns with Animated Map
10
Copyright © 2011, SAS Institute Inc. All rights reserved.

11. Visualising and Detecting Spatio-Temporal
Patterns with Trellis Maps
11
Copyright © 2011, SAS Institute Inc. All rights reserved.

12. Visualising and Detecting Spatio-Temporal
Point Patterns
12
Copyright © 2011, SAS Institute Inc. All rights reserved.

13. Visualising and Detecting Spatio-Temporal
Point Patterns with Density Map
13
Copyright © 2011, SAS Institute Inc. All rights reserved.

14. Q&A
Copyright © 2011, SAS Institute Inc. All rights reserved.