Data Science Fundamentals Vocabulary

A collection of key vocabulary flashcards grounded in the introductory data science lecture material, covering core concepts, computational thinking, data formats, and pre-processing techniques.

Data Science

A multidisciplinary, independent field of study and practice that combines statistical analysis, computer science, and domain expertise to collect, store, and process data to derive important insights and knowledge.

Data Collection

The process of gathering data from various sources such as databases, web scraping, sensors, and surveys.

Data Cleaning

Preparing data for analysis by removing errors, handling missing values, and standardizing formats.

Data Analysis

Using statistical and computational methods to explore and analyze data, uncovering patterns and insights.

Machine Learning

Applying algorithms to build models that can make predictions, classify data, and identify patterns.

Data Visualization

Creating visual representations of data to communicate findings and insights effectively.

Business Analytics (BA)

The skills, technologies, and practices for continuous iterative exploration and investigation of past and current business performance to gain insight and drive strategic decision-making.

Usefulness

In information science, a criterion that determines how effective the interaction between a user and an information object (data) is in accomplishing the task or goal of the user.

Computational Thinking

An iterative process of using abstraction and decomposition when attacking a large complex task or designing a large complex system, consisting of problem formulation, solution expression, and solution execution/evaluation.

Decomposition

The process of breaking down a large, complex problem into a set of smaller, manageable steps or problems.

Abstraction

The practice of treating an actual object of interest as a simplified or generalized series (such as treating a group of specific items as a series of numbers).

Generalization

Devising a process or algorithm that is applicable to an abstracted quantity rather than being restricted to specific given objects.

Data Literacy

The ability to assess a dataset for relevance and suitability, extract meaningful information, perform analysis, and create visualizations to tell valuable data stories.

Structured Data

Highly organized information with clearly defined fields or labels that can be seamlessly included in a database and readily searched via simple operations.

Unstructured Data

Information devoid of any underlying structure or labels, such as natural language text, which requires significant processing to extract structured elements.

Open Data

Data that is freely available in the public domain and can be used, reused, and redistributed by anyone without copyright, patent, or other control restrictions.

Application Programming Interface (API)

A programmatic set of rules and methods used to request, access, and send data between systems or platforms.

Multimodal Data

Data collected across multiple different forms and media, such as text, images, audio, video, gestures, body posture, or spatial patterns.

Comma-Separated Values (CSV)

A text-based file format commonly used for spreadsheets and databases in which individual fields and values are separated by commas.

Tab-Separated Values (TSV)

A text-based file format used for raw data exchange between spreadsheets where data fields are delimited by tab characters.

eXtensible Markup Language (XML)

A plain-text markup language designed to be both human- and machine-readable, providing a software- and hardware-independent format for storing and sharing data.

Really Simple Syndication (RSS)

An XML-based standard format used to publish and deliver frequently updated web content (like news or blogs) directly to RSS aggregators or users.

JavaScript Object Notation (JSON)

A lightweight text-interchange format based on JavaScript standard ECMA-262, constructed on name-value pairs and ordered lists of values for efficient client-server data exchange.

Data Pre-processing

The overarching workflow of scrubbing, transforming, and organizing dirty or raw real-world data to make it complete, consistent, and ready for computational analysis.

Data Munging

Also called data wrangling; the practice of manually, automatically, or semi-automatically manipulating raw data to convert it into a structure more suitable for automated analysis.

Noisy Data

Datasets containing random errors, corrupted values, or extreme outliers that interfere with accurate data processing.

Outlier

A specific data point that is markedly different in value from the other data points in a sample.

Data Integration

The process of combining data and metadata from multiple disparate files or databases into a single coherent storage location while resolving schema and value conflicts.

Data Transformation

Converting data into consistent and system-readable forms using methods such as smoothing, aggregation, generalization, normalization, and feature construction.

Data Reduction

Techniques used to obtain a reduced representation of a dataset that is much smaller in volume yet produces the same or similar analytical results.

Data Cube

A multidimensional dataset stored in two, three, or higher dimensions where each dimension represents an attribute of interest.

Data Cube Aggregation

A reduction method where multidimensional data is aggregated to the smallest sufficient representation required to address a specific task.

Dimensionality Reduction

A data reduction process that removes or combines feature columns in a dataset based on the inherent nature of the data.

Data Discretization

The mapping or dividing of continuous numerical values into discrete intervals or categorical subsets.

Nominal Data

A categorical data type consisting of values from an unordered set with no natural order between them (e.g., colors).

Ordinal Data

A data type whose values originate from an ordered set where relative rank or order is defined (e.g., test grades).

Continuous Data

A numerical data type consisting of real numbers that can take on an infinite number of possible values within a given range.

Feature Space Selection

A method of data reduction that chooses a specific subset of relevant features or columns from a dataset for analysis.