Please enable JavaScript.
Coggle requires JavaScript to display documents.
Data Mining (visualization methods (Geometric projection (Projection…
Data Mining
visualization methods
Icon-based
Chernoff Faces
Stick Figures
Shape coding
Color icons
Tile bars
Hierarchical
Tree-Map
Cone Trees
Worlds-within-Worlds
Dimensional Stacking
InfoCube
Geometric projection
Projection pursuit
Prosection views
Landscapes
Hyperslice
Scatterplot
Parallel coordinates
Direct visualization
complex data and relations
Non Numeric data
Tag cloud
Pixel-oriented
Dirty Data Problems
Noise (SMOOTHING)
Errors
Binning
Equal depth "Frequency"
Equal Width "Distance"
Outliers
Clustering
Regression
Inconsistent
Data Type
Conflict in Duplicated Records
Incomplete (Missing values)
Ignore
Replacing
Estimating
Binning
Eliminating
Duplicated Data
Data Preprocessing
Data Transformation
Generalization
Concept hierarchy
Normalization (NUMERIC)
Decimal scaling
Less than 1
v' = v/ 10^ number of digits
Z - score
Grater then 1
V' = (v-M) / Q
Min - Max
0 to 1
-1 to 1
v' = (v- min)/(max-min) [ Nmax- Nmin] + Nmin
Aggregation
Stigmatization
Sum
Avg
Max
Data cube construction
Data whorehouse
Attribute/feature construction
Discretization
Binning (NUMERIC)
Equal depth
Equal width
Concept Hierarchy (low level to high level)
Numeric
Interval label
Nominal
Smoothing
Noise
Error
Binning
Outliers
Clustering
Regression
Data Cleaning
Noise
Errors
Binning
Smooth by bin median
Smooth by bin boundaries
Smooth by bin means
Outliers
Clustering
Regression
Duplicated Data
Incomplete (Missing values)
Ignore
Replacing
Estimating
Binning
Eliminating
Inconsistent
Data Type
Conflict in Duplicated Data
Data Reduction
Numerosity reduction (NUMERIC)
Non-parametric
Histogram
Clustering
Sampling
Simple Random Sampling with Replacement (SRSWR)
Simple Random Sampling without Replacement (SRSWOR)
Clustering Sample
Stratified Sample
Parametric
Function take parameters
Ex: Regression
y=x+1
we send x
Data Compression
Lossless
Same data after re-compressed
Ex .. String data
Lossy
If some info. are lost
Ex: Audio - Video
Dimensionality reduction (Attributes)
step-wise backward elimination
Start with full set.
In each step, remove one worst attribute from the set.
combining
decision-tree induction
information gain
step-wise forward selection
Start with empty set.
In each step, select one best attribute to add in set.
Data Integration
Data Mining Functions
Classification
Cluster Analysis
Association Analysis
Outlier Analysis
Types of Data Sets
Graph
World Wide Web
Molecular Structures
Record
Data Matrix
Document Data
Transaction Data
Ordered
Temporal Data
Sequential Data
Spatial Data
Genetic Sequence Data
Data Types
Nominal
Binary
Symmetric
Asymmetric
Ordinal
Numeric
Interval
Ratio
Descriptions of Data
Measures of central tendency
Midrange
Mode
Median
Mean
Dispersion of the data
quartiles
five-number summary
range
boxplots
variance
standard deviation
ساجن أسد