Computational Biology Classification (some parts taken from Introduction to Data Mining by Tan, Steinbach, Kumar) Lecture Slides Week 10

MBG404 Overview Processing Pipelining Generation Data Storage Mining

Data Mining Data mining Machine learning noun Digital Technology. The process of collecting, searching through,  and analyzing a large amount of data in a database,  as to discover patterns or relationships:  e.g.: the use of data mining to detect fraud. Machine learning a branch of artificial intelligence, concerning the construction and study of systems that can learn from data. For example, a machine learning system could be trained on email messages to learn to distinguish between spam and non-spam messages. After learning, it can then be used to classify new email messages into spam and non-spam folders.

Application in Biology Data exploration Microarrays Next generation sequencing Prediction microRNAs Protein secondary structure

Parametrization

Types of Attributes There are different types of attributes Nominal Examples: ID numbers, eye color, zip codes Ordinal Examples: rankings (e.g., taste of potato chips on a scale from 1-10), grades, height in {tall, medium, short} Interval Examples: calendar dates, temperatures in Celsius or Fahrenheit. Ratio Examples: temperature in Kelvin, length, time, counts

Data Type Distinctness (nominal) Order (ordinal) Addition (interval) Equal or unequal Order (ordinal) >,<,>=,<= Addition (interval) +,- Multiplication (ratio) *,/

Data Quality Missing data Noise False measurements Outliers Duplicate data Precision Bias Accuracy

Data Preprocessing Aggregation Sampling Dimensionality Reduction Feature subset selection Feature creation Discretization and Binarization Attribute Transformation

Aggregation Variation of Precipitation in Australia Standard Deviation of Average Monthly Precipitation Standard Deviation of Average Yearly Precipitation

Dimensionality Reduction: PCA

Similarity/Dissimilarity for Simple Attributes p and q are the attribute values for two data objects.

Similarity Eucledian distance Simple matching coefficient Correlation Jaccard coefficient Correlation Cosine similarity ...

Sampling Curse of dimensionality Feature selection Dimensionality reduction Principal component analysis Aggregation Mapping of data to different space

Sampling Dividing samples into Using not all samples from both sets Training set Test set Using not all samples from both sets

Classification Examples with known classes (labels) Learn rules of how the attributes define the classes Classify unknown samples into the appropriate class

Classification Workflow

End Theory I 5 min Mindmapping 10 min Break

Practice I

Exploring Data (Irises) Download the file Iris.txt Follow along

Exploring Data Frequencies Percentiles Mean, Median Visualizations

Data Selection Selecting columns Filtering rows

Data Transformation Discretize Continuize Feature construction

Visualizations

End Practice I Break 15 min

Theory II

Classification Workflow

Illustrating Classification Task

Example of a Decision Tree categorical continuous class Splitting Attributes Refund Yes No NO MarSt Single, Divorced Married TaxInc NO < 80K > 80K NO YES Training Data Model: Decision Tree

General Structure of Hunt’s Algorithm Let Dt be the set of training records that reach a node t General Procedure: If Dt contains records that belong the same class yt, then t is a leaf node labeled as yt If Dt is an empty set, then t is a leaf node labeled by the default class, yd If Dt contains records that belong to more than one class, use an attribute test to split the data into smaller subsets. Recursively apply the procedure to each subset. Dt ?

Hunt’s Algorithm Refund Refund Refund Marital Marital Status Status Don’t Cheat Yes No Don’t Cheat Refund Don’t Cheat Yes No Marital Status Single, Divorced Married Refund Don’t Cheat Yes No Marital Status Single, Divorced Married Taxable Income < 80K >= 80K

How to Find the Best Split Before Splitting: M0 A? B? Yes No Yes No Node N1 Node N2 Node N3 Node N4 M1 M2 M3 M4 M12 M34 Gain = M0 – M12 vs M0 – M34

Underfitting and Overfitting Underfitting: when model is too simple, both training and test errors are large

Overfitting due to Noise Decision boundary is distorted by noise point

Overfitting due to Insufficient Examples Lack of data points in the lower half of the diagram makes it difficult to predict correctly the class labels of that region - Insufficient number of training records in the region causes the decision tree to predict the test examples using other training records that are irrelevant to the classification task

Cost Matrix C(i|j): Cost of misclassifying class j example as class i PREDICTED CLASS ACTUAL CLASS C(i|j) Class=Yes Class=No C(Yes|Yes) C(No|Yes) C(Yes|No) C(No|No) C(i|j): Cost of misclassifying class j example as class i

Cost-Sensitive Measures Precision is biased towards C(Yes|Yes) & C(Yes|No) Recall is biased towards C(Yes|Yes) & C(No|Yes) F-measure is biased towards all except C(No|No)

Receiver Operating Characteristic (ROC) Curve (TP,FP): (0,0): declare everything to be negative class (1,1): declare everything to be positive class (1,0): ideal Diagonal line: Random guessing Below diagonal line: prediction is opposite of the true class

Using ROC for Model Comparison No model consistently outperform the other M1 is better for small FPR M2 is better for large FPR Area Under the ROC curve Ideal: Area = 1 Random guess: Area = 0.5

End Theory II 5 min Mindmapping 10 min Break

Practice II

Learning Supervised (Classification) Classification Decision tree SVM

Classification Use the iris.txt file for classification Follow along as we classify

Classification Use the orangeexample file for classification We are interested if we can distinguish between miRNAs and random sequences with the selected features Try yourself