Introduction to Data Science
1 like214 views
This document provides an introduction to data science. It discusses the rapid growth of data and defines data science as extracting insights from vast amounts of data using scientific methods. The document outlines the typical steps in the data science process: acquire, prepare, analyze, report and act on data. It also discusses career opportunities in data science and common tools used, including programming languages, mathematics/statistics foundations and visualization/modeling tools.
Introduction to Data Science
- 1. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Introduction to Data Science Dr Ahmed Rebai, Phd in nuclear physics Dr Lotfi Ncib, PhD in applied maths
- 2. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Plan ▪ The Data Explosion ▪ The Data Hystory ▪ Why Data Science? ▪ What is Data Science ▪ Steps in The Data Science Process ▪ Career in Data Science ▪ Data Science Tools 1
- 3. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn The Data Explosion How Much Data Is Collected Every Minute of The Day in 2019 ? 2
- 4. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn The Data History Since the dawn of time… up until 2005 Humans had created 130 EXABYTES of Data 2005-130 EXABYTES 2010-1200 EXABYTES 2015-7900 EXABYTES 2020-40900 EXABYTES Byte Kilobyte(KB) 1.000=103 Megabyte(MB) 1.000.000=106 Gigabyte(GB) 1.000.000.000=109 Terabyte(TB) 1.000.000.000.000=1012 Petabyte(PB) 1.000.000.000.000.000=1015 Exabyte(XB) 1.000.000.000.000.000.000=1018 3
- 5. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn The Data History A 1 BYTE of space if we zoom out 1000 times we will get a page of letter (1 kB) about 500 characters Now let zoom another 1000 times and we will get a book - about 500 pages to take 1MB Now lets zoom another times and we will get 1GB(1 GB is sufficient to fit all human genomes once coded (Usually it takes 725MB) If we zoom another 1000 times we will get into TB(enough to fit some one’s life recorded for 8 years(everything they do- every minute or second If we zoom another 1000 times we will get into PB(Amazon rain forest is 1.4 Billion acres about 500 tree per acre / 700 billion trees). If you shup all these trees down and turn them in to papers and fill the papers with letters both side- close to 1PB. If we zoom another 1000 times we will get into XB(1000 TB) 4
- 6. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn The Data History 1 ZettaByte=1000 ExaByte 5
- 7. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Why Data Science? Salary trends have followed the impact of data science. With a national average salary of $118.000(which increase to $126.000 in Silicon Valley), data science has become a lucrative career path where you can solve hard problems and drive social impact. Data scientist is the sexiest career of the 21st century Statistical Analysis and Data Mining wher the hottest skills that got recruiter’s attention in 2014/2015/2016/2017/2018 The US alone faces a shortage of more than 150.000 data analyst and an additional 1.5 million data savy managers 6
- 9. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn What is Data Science? “The ability to take data — to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it — that’s going to be a hugely important skill in the next decades.” - Hal Varian, chief economist at Google and UC Berkeley professor of information sciences, business, and economics DATA SCIENCE is the area of study which involves extracting insights from vast amounts of data by the use of various scientific methods, algorithms, and processes. Data Science is the science wich uses computer science, statistics and machine learning, visualization and human-computer interactions to collect, clean integrate, analyze, visualize, interact with data to create data products, 8
- 10. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Steps in The Data Science Process ACQUIRE PREPARE ACTREPORTANALYZE Data Engineering Computational Data Science 9
- 11. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Steps in The Data Science Process ACQUIRE Step 1: Acquire Data ▪ Identify data sets ▪ Retrieve data ▪ Query data 10
- 12. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Step 2: Prepare Data ▪ Explore Data ➢ Understand nature of data ➢ Preliminary analysis ▪ Pre-process Data ➢Clean ➢Transform PREPARE Steps in The Data Science Process 11
- 13. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Step 3: Analyze Data ▪ Select analytical techniques ▪ Build modelsANALYZE SPAM Dimensionality Reduction Clustering Regression Classification Steps in The Data Science Process 12
- 14. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn REPORT Step 4: Communicate Results ▪ What to present ▪ How to present Steps in The Data Science Process 13
- 15. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Steps in The Data Science Process ACT Step 5: Turning Insights into Action ▪ Results ▪ Purpose 14
- 16. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Career in Data Science Domain Expertise Programing languages Math/ statistic/ Probability Lingo/Fondations Projects ❖ Health care ❖ Retail ❖ Finance ❖ Eduction ❖ … ❖ Python ❖ R ❖ C++ ❖ Java ❖ Julia ❖ Scala ❖ … ❖ Sentiment analysis ❖ Card Fraud detection ❖ Customer segmentation ❖ Image classification ❖ Loan default detection ❖ … ❖ Machine Learning ❖ Deep Learning ❖ Classification ❖ Rgression ❖ Clustering ❖ Decision trees ❖ KNN ❖ SVM ❖ Kmeans ❖ PAC ❖ … ❖ Linear algebera ❖ Bayes theorem ❖ Mean, Median and Mode ❖ Covariance and correlation ❖ Central Limite Theorem ❖ Normal Distrubtion ❖ … 15
- 17. ahmed.rebai@esprit.tnLotfi.ncib@esprit.tn Data Science Tools Visualization ToolsModeling Tools 16