0

If i'm writing an algorithm that requires storage of data that will be sequentially accessed from the first item in the list to the last item in the list. And that Data will regularly need to be inserted or deleted at random locations anywhere in the list.

Would I be correct in saying that LinkedList would be more appropriate then ArrayList, Binary Tree and Circular Buffer?

ArrayLists are Random accesss so it's not them, Binary Trees aren't a linear data structure so it's not them as they can't be sequentially accessed in a single run, It can't be a circular buffer as Data needs to be inserted or deleted at random locations but they just enqueue and dequeue at 2 select positions only.

compsci3289294
  • 29
  • 2
  • It depends on how you will determine where to insert an item. Is it based on some sort order that must be maintained? Is it based on a node you already have, and you just want to insert just after or before it? Is it based on a certain index that the new element should have? – trincot Oct 31 '22 at 16:21
  • Items are inserted at random locations – compsci3289294 Oct 31 '22 at 20:35
  • How do you define "location"? What is the random thing... and index? Or a node reference? Examples in your question would be welcome ;-) – trincot Oct 31 '22 at 20:41
  • I'm not sure, this is just how the textbook phrased the question. And there's no mark scheme. – compsci3289294 Oct 31 '22 at 20:44

1 Answers1

0

If all you got is the head of a linked list, and the index ("location") where an insertion must be made, then a linked list is not very handy: you'll have to traverse from the head of the list to the desired location and that takes O() time, where is the desired location. As can in principle be anything between 0 and (where is the current number of nodes in the list), this means the worst and average time complexity is O().

A better structure will be a self-balancing binary search tree (such as AVL, red-black, B-tree, ...), where each node is "augmented" with a property that gives the number of values in the subtree it roots. This allows the values to be stored and traversed (inorder traversal) according to their index. Insertion and deletion then have O(log) time complexity. A particular handy tree data structure for traversal is the B+-tree.

Another one that is as efficient is the skip list. This is a linked list that is augmented with several layers of linked lists, so to allow O(log) access to any index.

If however with "location" it is meant that you get a node reference in a linked list, where the new node must be inserted, then a linked list is fine. But then the question is: how did you get that location? In the end, that is a node reference that must be retrieved somehow, and there is not much else you can do than traverse a list from its head to get such node references... so then we're back to the start of my answer.

trincot
  • 317,000
  • 35
  • 244
  • 286