Insertion & Selection Sort - using Priority Queues

Editor's Notes

• #3 In this session we'll learn how to implement a priority queue by storing its entries in a list S , where the list can be sorted or unsorted. for unsorted list, //simple way of performing insert operation is to add the element at the end of list , by executing method addLast().// This implementation of method insert takes O(1) time as inserting entries at the end of list does not take into account the ordering of the keys. To perform removeMin method , we have to inspect all the elements of list to find an entry with minimum k. So it taken O(n) time, where n is the number of entries, or data size.// So thats why implementation with an unsorted list is also called as "fast insertion and slow removal".
• #4 An alternative implementation of a priority queue is in a sorted list. ,// here insertion requires a complete scan of the list to find the appropriate position to insert the new entry so it takes O(n) time, while// removeMin method just needs to access the first element of the list as that will be the one which is least in the sorted list.and so it takes O(1) times.// it is also called as fast removal and slow insertions,
• #6 There are basically two most popular techniques for sorting, selection-sort and insertion-sort. Lets talk about each of this one by one.
• #7 In selection -sort ,// lets Consider we have to sort n numbers stored in array A by first finding the smallest element of A and exchanging it with the element in A[0], the first element. Then finding the second smallest element of A, and exchanging it with A[1], the second element. Continue in this manner for the first n−1 elements of A. This method of sorting is known as selection sort.
• #8 In last session I mentioned about sorting using priority queus, through insert and removeMin() . Lets see how priority queues help in sorting, lets say we have some data in sequence S, and an empty priority queue, // Selection sort occurs in two phases, in first phase we insert all the elements in priority queue P using simple insert() method,// we ll pick first item// from the sequence insert into P,then second element// , third and so on, this way we have complete data in prority queue now, now in Phase 2 //we repeatedly remove the elements from P using the removeMin() method. we ll find the least element in the qyueue, and put it back in sequence S, // //, we ll do this till the last element in priority queue , when all the elements of queue are moved to sequence S in order. In this whole process bottleneck is in phase 2, where we repeatedly remove an entry with smallest key from Queue P. We start with priority queue size equals n and incrementally decreases with each removeMin until it becomes 0.
• #9  //Thus, the first removeMin operation takes time O(n) as there are n entries, then the second one// takes time O(n − 1) as one entry has already been moved to sequence S and so number of elements are (n-1) now and so on, //until the last (nth) operation takes time O(1). Therefore, //the total time needed for the second phase is //
• #10 Next is insertion sort: In insertion sort the first entry is always taken as sorted. // Insertion of the second entry requires one comparison with the first entry, in order to find its proper position. //Insertion of the third entry requires two comparisons, and // fourth entry requires three comparisons and so on. This method is known as insertion sort, where in each iteration, one entry is added to the sorted list. Inserting after comaprison
• #11 Insertion sort using priority queue is again a two phase process, : here we have a data set in sequence S, and // an empty priority queue P, In Phase 1, we remove the// first element of list and insert it into Priority queue P,, then second element// is removed from the sequence and we scan the data in priority queue , until we find the correct place for this element. , same step is repeated for //third element, it is compared with existing two elements in priority queue and placed at right position, and this is repeated for the whole data set.// // // In Phase 2, we call removeMin method on Priority queue repeatedly,// here because of phase 1 every time it returns the first element of the list and// // we add that element at the end of Sequence S, and this way we get sorted sequence S.
• #12  Analyzing the running time of this whole process, we see that phase 1 is the bottleneck of this sorting ,//as for placing first element , there is no comparison , //while for second element , there is 1 comparison and so on. and like selection sort, //insertion sort also runs in O(n2) time