Back

What is Data Science

Systems, algorithms, and processes managing and deriving insights from heterogeneous and/or large data science.

Typical Workflow for a Data Scientist

1. Data Collection
2. Data Processing
3. Exploration / Visualization
4. Analysis / Machine Learning

Data scientists spend the majority of time performing the first two stages. Data fusion is long and tedious. Most data will be unstructured. Data management tools are very useful, and will make data management as well as processing much more efficient (AWS, Azure for example).

Structured Data

A Relation R is a subset of $S_1 x S_2 x ... x S_N$ where $S_1$ is Domain of attribute i in [1, n], and n is a number of attributes of R A Tuple t is an element of $S_1xS_2x...xS_2$

• A relation is a set of tuples Relational Database is a collection of relations

Relation Schema

• Relation name
• Name of each attribute
• Type of each attribute Database schema
• A set of all relation schema's

Back