LR Parsing Classification: Top-Down, Bottom-Up, Recursive, and Predictive

Verified Sources

Nov 4, 2026

LR parsing is a bottom-up, shift-reduce, deterministic parsing strategy constructed from LR parsing tables (typically ACTION and GOTO) that simulate a rightmost derivation in reverse while scanning input left-to-right. In the standard definition, the “L” means left-to-right scanning of the input, and the “R” indicates use of a rightmost derivation handle concept in the parsing process.
Therefore, among your options, (ii) Bottom-up parsing is the correct choice.

The other options are not the defining characteristics of LR parsing:

(i) Top-down parsing: that’s the family associated with LL parsing (predictive, recursive-descent style).
(iii) Recursive parsing: LR parsers are typically table-driven automata; recursion is not the core defining mechanism.
(iv) Predictive parsing: “predictive” is the hallmark of LL(1)-style lookahead selection; LR parsing uses LR lookahead sets inside ACTION/GOTO table entries rather than an LL-style single nonterminal expansion prediction.

Key terms

LR parsing
shift-reduce parser
ACTION table
GOTO table

⚠️ Important: I attempted web research as required, but external search calls are currently failing in this environment (tool usage limit). I’m still providing a correct conceptual answer based on standard compiler-construction definitions.

type="tip" title="How to recognize LR parsing quickly" content="If a question mentions ACTION/GOTO tables, shift/reduce moves, or “handle” recognition, it’s almost certainly LR = bottom-up shift-reduce parsing."

How LR parsing builds a parse (conceptual lifecycle)

Start state

1. Initialization

Create a stack with the initial state; input pointer at the start."

Move input into stack

3. Shift

If ACTION says shift, push the symbol and advance input."

Apply a production

4. Reduce

If ACTION says reduce by $A \to \beta$ , pop $|\beta|$ symbols."

Transition after reduction

5. GOTO

Use GOTO with the new top state and $A$ to push the nonterminal."

Successful parse

6. Accept

When ACTION is accept, the parse is complete."

Core mechanics of an LR shift-reduce parser (ACTION/GOTO simulation)

1
Step 1
Push initial state onto the stack. Set input pointer to the first token.
2
Step 2
Let $a$ be the current lookahead token from the input.
3
Step 3
Let $s$ be the state on top of the stack. Use ACTION[ $s$ , $a$ ] to decide.
4
Step 4
If ACTION[ $s$ , $a$ ] = shift $t$ , push $a$ and state $t$ , then advance the input pointer.
5
Step 5
If ACTION[ $s$ , $a$ ] = reduce by $A \to \beta$ , pop $|\beta|$ stack symbols.
6
Step 6
Let $s'$ be the new top state. Push $A$ , then push GOTO[ $s'$ , $A$ ].
7
Step 7
"If ACTION[ $s$ , $a$ ] = accept, stop successfully; if no valid action exists, parsing fails."

Key terms

lookahead

handle

reduce

Why the correct option is (ii) Bottom-up parsing

LR parsers work by reducing toward the start symbol. Reductions correspond to recognizing handles (RHS patterns) in the current partially built structure, which is characteristic of bottom-up parsing. This is the opposite direction of top-down parsing, which predicts and expands from the start symbol.

A compact conceptual contrast:

Multiple-choice reasoning (i)–(iv)

Which statement best matches LR parsing?

Binary qualitative alignment (1 = matches, 0 = does not match)

Knowledge Check

Question 1 of 4

Q1Single choice

LR parsing is fundamentally characterized by which parsing direction?

Bottom-up (shift-reduce) parsing

Top-down (predictive expansion) parsing

Recursive-only parsing (must use recursion)

Predictive parsing in the LL(1) sense

Explore Related Topics

Similarity Between SLR(1), LALR(1), and LR(1): Correct Choice and Learning Path

Handle Pruning: Its Role in Predictive Parsing (and How It Differs from Other Parsing Paradigms)

Top-Down and Bottom-Up Parsing Techniques

Browse all research articles