Fundamentals of Data Science Quiz 1

Study Notes

Data Science: Concepts and Practice (Vijay Kotu and Bala Deshpande, 2019)
DATA SCIENCE: FOUNDATION & FUNDAMENTALS (B. S. V. Vatika, L. C. Dabra, Gwalior, 2023)

Data science is a compilation of techniques to extract value from data.
Data science techniques use patterns, connections, and relationships in the data.
Data science roots are in applied statistics, machine learning, visualization, and computer science.
Data science is also called knowledge discovery, machine learning, predictive analytics, and data mining.
Underlying methods of data science are decades (if not centuries) old.
The term "science" in data science implies an evidence-based approach, built on empirical knowledge, especially historical observations.
Data science increases company efficiencies, manages costs, and boosts market advantage.

Artificial intelligence, machine learning, and data science are related.
These terms are used interchangeably in media and business communication.
Artificial Intelligence aims to give machines the ability to mimic human cognitive behavior (facial recognition, automated driving, mail sorting).
Machine learning is a subfield or a tool of AI, enabling machines to learn from experience.
Machine learning can fail if an AI system isn't properly programmed, or if given inaccurate or incomplete data (e.g., self-driving car misidentifying a detour).
Data used to teach machines is called "training data".

Capture: Gathering structured and unstructured data from various sources (manual entry, web scraping, systems, devices).
Prepare and Maintain: Converting raw data to a consistent format for analytics/machine learning. Data cleaning, deduplication, reformatting, and ETL (extract, transform, load) are involved.
Preprocess/Process: Analyzing data (biases, patterns, ranges, distributions) to confirm suitability for predictive analytics/machine learning. Data scientists use methods like statistical analysis, predictive analytics, regression, and machine learning algorithms.
Analyze: Performing statistical analysis, predictive analytics, machine learning, and more to extract insights from the processed data.
Communicate: Presenting the insights through reports, charts, and visualizations to stakeholders for easier understanding of the impact.

Data scientists analyze business data to extract insights and solve problems.
Data scientists need business acumen, technology expertise, and mathematics expertise.
- Business Acumen: understanding domain, business strategy, problem-solving, communication, presentation, inquisitiveness
- Technology Expertise: Good database knowledge (RDBMS, NoSQL databases like MongoDB, Cassandra, HBase), programming languages (Java, Python), open-source tools (Hadoop, R), data warehousing, data mining, and visualization tools (Tableau, Flare, Google visualization APIs).
- Mathematics Expertise: mathematics, statistics, artificial intelligence (AI), machine learning, pattern recognition, natural language processing.

Before data collection, a data scientist defines the problem, through questions and understanding.
Data scientists determine which variables and data sets are correct.
Data scientists gather data from various sources, converting it into a suitable format for analysis.
The data is cleaned and validated for uniformity, completeness, and accuracy.
Data inputs are fed into analytic systems (ML algorithms or statistical models)
Data scientists analyze/identify patterns and trends.
Data scientists interpret data to identify opportunities and solutions.
Data scientists present insights through reporting, charts, and visualizations.