KU-AICC202-Intro-to-Data-Science

Kathmandu University Department of Computer Science

Subject: Introduction to Data Science

Course Code: AICC 202

Level: BTech in AI 2nd year 1st semester

Credit Hours: 3

Type: Core [Theory + Practical]

Course Description

Data Science called as “the sexiest job of the 21st century” by Harvard Business Review., and in this period “the world’s most valuable resource is no longer oil, but data”. This course helps students to understand the basic concepts of Data Science like information extraction from the vast amounts of data using different scientific methods. It explains how the data is manipulated and processed to unravel useful underlying information from the raw and unstructured data. The course follows an example-based approach in terms of providing a better understanding on how data science techniques are applied in the real-world problems. Furthermore, the course provides insights to the ethical use of data, which has become very crucial with the rampant and abundant generation of data.

Programming Language used: Python, Julia, R

Course Objective

• To introduce the basic concepts and techniques of Data Science.
• To give an insight to data visualization
• To understand the current scenario and application of Data Science.
• To understand how Data Science techniques are used in real-world problems.

Prerequisites

It is expected that students have prior knowledge of mathematical preliminaries such as Probabilities and Statistics, Calculus and Linear Algebra. Besides, students should have the knowledge of high-level programming languages like C and C++ or Python to understand the concepts of Data Science and implement projects.

Course Evaluation

Internal Examination: 50% Final Examination: 50%

Chapters

Chapter 1: Introduction to DataScience [6 Hrs.]

• Evolution of Science: Empirical, Theoretical, Computer Science and Data Science
• Current Trends in Data Science
• DIKW Pyramid
• Asking Good Questions
• Data Types and Properties
• Data Analytics Value Escalator
• Data Science Hierarchy
• Data Science Lifecycle

Practical:

• Python
• Julia
• R

Chapter 2: Data Wrangling [9 Hrs.]

• Languages for Data Science
• Data Discovery
• Data Structuring
• Data Cleaning
• Data Transformation
• Data Sampling Techniques
• Data Enriching
• Feature Selection
• Data Validating
• Data Publishing

Chapter 3: Data to Models: Score, Rank, Class, Clusters [6 Hrs.]

• Why Model ?
  • Philosophies of Modeling
• Data to Score
  • The Body Mass Index (BMI)
  • Developing Scoring Systems
  • Linear Regression
  • Metrics
• Data to Rank
  • Advanced Ranking Techniques: PageRank, Elo Ranking, Machine-learned ranking (MLR)
  • Arrow’s Impossibility Theorem
  • Metrics
• Data to Class
  • Types of Classification
  • Metrics
• Data to Clusters
  • Types of Clustering
  • Cluster Validation

Chapter 4: Visualizing Data [6 Hrs.]

• Exploratory (EDA) and Explantory Data Analysis
• Grammar of Graphics
• Developing a Visualization Aesthetic
• Chart Types
• Data Visualization Perspectives
• Principles of Visual Design
• Interactive Visualization
• Common tools for Data Visualization
• Storytelling with data
• Misinformation and Disinformation

Chapter 5: Data Analysis [12 Hrs.]

• Descriptive Analysis
• Diagnostic Analysis
• Predictive Analysis
• Prescriptive Analysis
• Case Studies

Optional

• Process Analysis
• Text Analysis
• Qualitative Research & Analysis

Chapter 6: Big Data [6 Hrs.]

• Introduction to Big Data
• Big Data Ecosystem
• Parallelism
• Hadoop and MapReduce
• Apache Spark

Chapter 7: Future Trends in Data Science [3 Hrs.]

• Current Operational trends
• Next-Generation Data Scientists
• Cultivating Soft Skills
• Being an Ethical Data Scientist

Text Books

1. Steven S. Skiena. The Data Science Design Manual. Springer Cham, 2017
2. Cathy O’Neil, Rachel Schutt. Doing Data Science. O’REILLY Media, Inc., 2013

Reference Books:

1. Joel Grus. Data Science from Scratch 2e: First Principles with Python. O’REILLY, 2019
2. Ivo D. Dinov. Data Science and Predictive Analytics. Springer, 2018

This site is open source. Improve this page.