Computer Science Vocabulary

Understanding Computer Science Vocabulary

This vocabulary set covers several major parts of computer science. It begins with how computers store information, moves through algorithms that process that information, explains how devices communicate, then reaches modern machine learning. These areas may seem separate at first.

They are connected by one main idea. A computer represents information in a form it can handle, follows precise steps to transform it, sends it when needed, then may learn useful patterns from large collections of it. Knowing the words helps you describe each stage clearly.

Bits, bytes, place value, bases, and hexadecimal explain the language of digital storage. Computers use two possible states at the lowest level, so binary is natural for electronic circuits. A bit holds one of those states.

Groups of bits form larger values, including bytes. Place value still works in binary and hexadecimal, but each position has a value based on that number system.

Hexadecimal is useful because one hex digit matches a small group of binary digits. This makes long binary values easier for people to read when working with memory, colors, machine code, or network addresses.

Algorithm terms focus on efficiency. Two programs can produce the same answer while taking very different amounts of time or memory as the input grows. Big O gives a way to compare that growth.

Time complexity concerns the work done. Space complexity concerns extra storage used. Worst case thinking prepares you for difficult inputs, while amortized complexity explains why a sequence can be efficient even if one step is occasionally expensive.

Sorting vocabulary shows these ideas in action. A stable sort preserves the order of equal items.

An in place sort avoids much extra memory. Divide and conquer breaks a large task into smaller tasks, and a pivot can guide how a sorting method splits data.

Networking terms describe how information leaves one device and reaches another. Data is divided into packets so it can travel through a network. An IP address identifies a destination.

Routers choose paths between networks. Protocols are shared rules that let different devices interpret messages in the same way. DNS connects human friendly site names to numerical addresses.

These terms matter whenever you troubleshoot a slow connection, build a web service, or think about privacy. A request is not magic. It follows layers of rules and passes through real hardware.

The machine learning terms describe how a model turns examples into predictions or generated text. Layers contain adjustable weights and biases. Activation functions shape the signal moving through a network.

Backpropagation changes internal values after the model measures its error. Pretraining builds broad knowledge from large data, while fine tuning adapts a base model for a narrower task. Embeddings turn words, images, or other items into vectors in a vector space.

Distance or cosine similarity can then represent related meaning. In attention based models, query, key, and value vectors help each token focus on useful context.

Study this deck by grouping terms into these systems. Draw small flow diagrams, explain each process aloud, then use the terms to compare real programs, networks, and models.