Data Science

IS EATING THE WORLD

by mart van de ven of droste

Data Science

is

creative

problem solving

with

data

as the

raw

material

y = f(x) + \epsilon
y=f(x)+ϵy = f(x) + \epsilon
\text{Data Science vs Data Analysis}
Data Science vs Data Analysis\text{Data Science vs Data Analysis}
\text{Data Science vs Data Engineering}
Data Science vs Data Engineering\text{Data Science vs Data Engineering}

Data engineers essentially

data for their evaluations and

experiments.

easily retrieve the needed

analyst or data scientist to

lay the groundwork for a data

\text{Data Science vs Machine Learning}
Data Science vs Machine Learning\text{Data Science vs Machine Learning}

Machine Learning automates modeling. Using

algorithms that iteratively learn from data,

ML allows computers to surface relationships

without explicitly telling them what to look.

DATA DRIVEN EXPERIENCES

\text{Example: Waze}
Example: Waze\text{Example: Waze}

Text

\text{Level 1: Data Collection}
Level 1: Data Collection\text{Level 1: Data Collection}

Text

\text{Level 2: Data Application}
Level 2: Data Application\text{Level 2: Data Application}
\text{Level 3: Data Driven Experience}
Level 3: Data Driven Experience\text{Level 3: Data Driven Experience}

In a world where software builds itself,

computers will only be limited

 by the data they can or cannot access,

not by their algorithms.

software builds itself,

Our Stack is Open

Data Storage
Data Pipelines
Machine Learning
Analysis
Visualisation

Our Stack is Open

Our Stack is Open

SQL,HDFS,Cassandra
Airflow,Luigi,OpenRefine
SciKit,MLLib,TensorFlow Pandas,R,
GGplot,Bokeh,Plotly
\text{supervised} \space \text{vs} \space \text{unsupervised}
supervised vs unsupervised\text{supervised} \space \text{vs} \space \text{unsupervised}
\text{classification} \space \text{vs} \space \text{estimation}
classification vs estimation\text{classification} \space \text{vs} \space \text{estimation}
\text{Example : Automated Risk Credit Management}
Example : Automated Risk Credit Management\text{Example : Automated Risk Credit Management}
\text{Example : Payments Analytics}
Example : Payments Analytics\text{Example : Payments Analytics}
\text{Example : Customer Segmentation}
Example : Customer Segmentation\text{Example : Customer Segmentation}
\text{Cultivating a Data Science Team}
Cultivating a Data Science Team\text{Cultivating a Data Science Team}

A Data Scientist is a

better programmer than

most statisticians, and

a better statistician

than most programmers.

\text{Code Demo : Predicting P2P Loan Interest Rates}
Code Demo : Predicting P2P Loan Interest Rates\text{Code Demo : Predicting P2P Loan Interest Rates}

Level 2 : GA HK

 Level 1 : DS Prologue

Level 3 : OSDS Masters

LEARNING DATA SCIENCE

Mart van de Ven

DIRECTOR

DROSTE

M @ DROSTE.HK

DATA SCIENTIST

Thank you.

Data Science is Eating the World @ UBS

By Droste

Data Science is Eating the World @ UBS

Introductory talk on Data Science for UBS

  • 1,708