Scribd
Upload
67 views

Syllabus: 1. Introduction To Data Science

This syllabus outlines a data science course covering key topics including introduction to data science concepts and tools, exploring and preparing data, predictive modeling techniques, model evaluation, data engineering, probabilistic models, exploratory data mining, and case studies. The course consists of 9 sections, beginning with introductions and working towards hands-on challenges. Students will learn foundational skills like visualizing, cleaning and transforming data, as well as algorithms including decision trees, clustering and Naive Bayes. Practical exercises using the WEKA tool are integrated throughout the sections.

Uploaded by

Uday Kumar
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
67 views

Syllabus: 1. Introduction To Data Science

This syllabus outlines a data science course covering key topics including introduction to data science concepts and tools, exploring and preparing data, predictive modeling techniques, model evaluation, data engineering, probabilistic models, exploratory data mining, and case studies. The course consists of 9 sections, beginning with introductions and working towards hands-on challenges. Students will learn foundational skills like visualizing, cleaning and transforming data, as well as algorithms including decision trees, clustering and Naive Bayes. Practical exercises using the WEKA tool are integrated throughout the sections.

Uploaded by

Uday Kumar
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

SYLLABUS

1. INTRODUCTION TO DATA SCIENCE

What is data science, relation to data mining, machine learning, big data and
statistics
Motivating examples

Why is it interesting?

Several data science settings

Introduction to the WEKA tool

Practical information

2. GETTING TO KNOW YOUR DATA

From data to features


o Interactive group discussion
o

Representing problems with matrices

Representing problem with relations

Example: Text with TFIDF

Computing simple statistics


o

Means, variances, standard deviations, weighted averaging, modes, quartiles

Example: Political predictions

Simple visualizations
o

Histograms

Boxplots

Scatterplots

Time series

Spatial data

Case studies
o

X & Y examples

Medical data

3. OVERVIEW OF TASKS & TECHNIQUES: PREDICTION

The prediction task


o Definition
o

Examples

Format of input / output data

Prediction algorithms
o

Decision trees

Rule learners

Linear/logistic regression

Nearest neighbour learning

Support vector machines

Properties of prediction algorithms and practical exercises

Combining classifiers

4. EVALUATION AND METHODOLOGY OF DATA SCIENCE

Experimental setup
o Training, tuning, test data
o

Holdout method, cross-validation, bootstrap method

Measuring performance of a model


o

Accuracy, ROC curves, precision-recall curves

Loss functions for regression

Interpretation of results
o

Confidence interval for accuracy

Hypothesis tests for comparing models, algorithms

5. DATA ENGINEERING

Attribute selection
o Filter methods
o

Wrapper methods

Data discretization

Unsupervised discretization

Supervised discretization

Data transformations
o

PCA and variants

Exercises

6. OVERVIEW OF TASKS & TECHNIQUES: PROBABILISTIC


MODELS

Introduction
o Probabilities
o

Naive Bayes
o

Rule of Bayes and Conditional Independence

Application to spam filtering

Bayesian Networks
o

Graphical representation

Independence and correlation

Temporal models
o

Markov Chains

Hidden Markov Models

7. OVERVIEW OF TASKS & TECHNIQUES: EXPLORATORY


DATA MINING

Introduction to Exploratory Data Mining


Association discovery
o

What is association discovery?

What are the challenges?

In detail: Apriori

Clustering
o

What is clustering?

What are the challenges?

In detail: agglomerative clustering

Hands-on: clustering in WEKA

8. CASE STUDIES IN DATA SCIENCE

Eve, the Pharmaceutical Robot Scientist: Data Science for Drug Discovery
Data science for sports analytics

Data science for sensor data (Introduction to challenge)

9. CHALLENGE

Introduction
Hands-on by participants

Discussion of results

You might also like