VulnTech Which Data Structure Should I Use? | DSA with Java

Start from the question you need to answer fast

If you mostly need to…	Use	Cost
read/write by position	array, `ArrayList`	O(1)
add/remove at the end	`ArrayList`	O(1)
add/remove at both ends	`ArrayDeque`	O(1)
undo / last-in-first-out	`ArrayDeque` as a stack	O(1)
serve in arrival order	`ArrayDeque` as a queue	O(1)
check “have I seen this?”	`HashSet`	O(1) average
look up a value by key, count things	`HashMap`	O(1) average
keep things sorted, find nearest	`TreeSet`, `TreeMap`	O(log n)
always get the smallest/biggest next	`PriorityQueue`	O(log n)
model connections	adjacency list `ArrayList<ArrayList<Integer>>`	O(n + m) memory
track merging groups	union-find (two small methods)	≈ O(1)

A worked example

“A website gets page visits as a stream. At any moment, tell me the most visited page; when two tie, the one that comes first alphabetically.”

“How many visits per page?” → a HashMap<String, Integer> for counting.
“The most visited” → keep a running best while counting (or a PriorityQueue if pages can also lose visits).

import java.util.HashMap;

public class Main {
    public static void main(String[] args) {
        String[] visits = {"home", "blog", "shop", "blog", "home", "blog", "shop", "shop"};
        HashMap<String, Integer> count = new HashMap<String, Integer>();
        String top = null;
        for (String page : visits) {
            int c = count.getOrDefault(page, 0) + 1;
            count.put(page, c);
            int best = top == null ? 0 : count.get(top);
            if (c > best || (c == best && page.compareTo(top) < 0)) top = page;
            System.out.println("after " + page + ": top is " + top + " (" + count.get(top) + ")");
        }
    }
}

Output

after home: top is home (1)
after blog: top is blog (1)
after shop: top is blog (1)
after blog: top is blog (2)
after home: top is blog (2)
after blog: top is blog (3)
after shop: top is blog (3)
after shop: top is blog (3)

Tip: Real solutions often combine two structures – a map plus a heap, a set plus a queue. Ask what each part of the problem needs to do quickly.

Remember: Java 8 collections never take primitives: ArrayList<Integer>, not <int>. For huge amounts of numbers, a plain int[] is faster and smaller.