In the example above, it is as though the elements are in two-dimensions such a table

Transcription

1 Multiple-Dimensional Arrays Elements of arrays may have more than one index, i.e. In example above, it is as though elements are in two-dimensions such a table with rows and columns Array Additional Searching dimensions and Sorting that go beyond two may be included as is needed CST242 A Sample 2-Dimensional Array Multiple-Dimensional Arrays Accessing Array Elements Elements of arrays may have more than one index, i.e. An array element is accessed by using row and column integers in two sets of brackets In example above, it is as though elements are in two-dimensions such a table with rows and columns Additional Integer variables dimensions may that be used go beyond to replace two eir may be or included both of as indexes, is needed i.e. mileage[start][end] Accessing Array Elements Declaring a 2-Dimensional Array An array element is accessed by using row and column integers in two sets of brackets The new array object is created using two sets of brackets in array declaration First bracketed value is conceptually number of rows Integer Second bracketed variables may value be is used conceptually to replace eir number or both of columns of indexes, i.e. mileage[start][end] int[][] mileage = new int[4][4]; Declaring a 2-Dimensional Array Initializing a 2-Dimensional Array in Declaration The new array object is created using two sets of brackets in array declaration First Each bracketed row is assigned value is in conceptually a separate subset number of command-delimited of rows brackets Second bracketed value is conceptually number of columns int mileage[][] = int[][] { mileage = new int[4][4]; { 0, 160, 390, 40}, Initializing {160, a 0, 2-Dimensional 240, 200}, Array in Declaration Each {390, row 240, is assigned 0, 440}, in a separate subset of command-delimited brackets { 40, 200, 440, 0} int }; mileage[][] = { MultidimensionalArrays.java (Page 1) { 0, 160, 390, 40}, MultidimensionalArrays.java {160, 0, 240, 200}, (Page 2) {390, 240, 0, 440}, MultidimensionalArrays.java (Page 3) { 40, 200, 440, 0} MultidimensionalArrays.java }; (Page 3) Searching and Sorting Searching data involves determining wher a value (referred to as search key) is present in data and, if so, finding its location. Two popular search algorithms are simple linear search and faster but more complex binary search Sorting places data in ascending or descending order, based on one or more sort keys The Arrays Class Class Arrays contains methods for manipulating arrays (such as sorting and searching) Also contains a static methods that allow arrays to be viewed as lists Found in java.util package, i.e. import java.util.arrays; The sort() Method of Class Arrays The sort method is a static method of Arrays class that sorts elements in an array in ascending order by default Overloaded version lets sort order to be changed Format: Arrays.sort(arrayObject);

2 import java.util.arrays; The sort() Method of Class Arrays The sort method is a static method of Arrays class that sorts elements in an array in ascending order by default Overloaded version lets sort order to be changed The Format: sort method is a static method of Arrays class that sorts elements in an array Arrays.sort(arrayObject); in ascending order by default Overloaded version lets sort order to be changed Format: Arrays.sort(value); Arrays.sort(arrayObject); The tostring() Method of Class Arrays The tostring method is a static method of class Arrays that displays array object as Arrays.sort(value); a list The Format: tostring() Method of Class Arrays The Arrays.toString(arrayObject); method is a static method of class Arrays that displays array object as a list Format: System.out.println( Arrays.toString(value) ); Arrays.toString(arrayObject); ArraysSort.java Linear System.out.println( Search Arrays.toString(value) ); Linear The linear Search search algorithm searches each element in an array sequentially. If search key does not match an element in array, algorithm tests each The linear search algorithm searches each element in an array sequentially. element, and when end of array is reached, informs user that search If key is search not present. key does not match an element in array, algorithm tests each element, and when end of array is reached, informs user that search If search key is in array, algorithm tests each element until it finds one key is not present. that matches search key and returns index of that element. If search key is in array, algorithm tests each element until it finds one If re are duplicate values in array, linear search returns index of first that matches search key and returns index of that element. element in array that matches search key. If re are duplicate values in array, linear search returns index of first element in array that matches search key. Searching algorithms all accomplish same goal finding an element that matches a given search key, if such an element does, in fact, exist. The major difference is amount of effort y require to complete search. Big O notation indicates worst-case run time for an algorithm that is, how hard an algorithm may have to work to solve a problem. For searching and sorting algorithms, this depends particularly on how many data elements re are. If an algorithm is completely independent of number of elements in array, it is said to have a constant run time, which is represented in Big O notation as O(1). An algorithm that is O(1) does not necessarily require only one comparison. O(1) just means that number of comparisons is constant it does not grow as size of array increases. An algorithm that requires a total of n 1 comparisons is said to be O(n). An O(n) algorithm is referred to as having a linear run time. O(n) is often pronounced on order of n or simply order n. Constant factors are omitted in Big O notation. Big O is concerned with how an algorithm s run time grows in relation to number of items processed. O(n 2 ) is referred to as quadratic run time and pronounced on order of n-squared or more simply order n-squared. When n is small, O(n 2 ) algorithms (running on today s computers) will not noticeably affect performance.

3 of items processed. O(n 2 ) is referred to as quadratic run time and pronounced on order of n-squared or more simply order n-squared. When n is small, O(n 2 ) algorithms (running on today s computers) will not noticeably affect performance. But as n grows, you ll start to notice performance degradation. An O(n 2 ) algorithm running on a million-element array would require a trillion operations (where each could actually require several machine instructions to execute). A billion-element array would require a quintillion operations. You ll also see algorithms with more favorable Big O measures. The linear search algorithm runs in O(n) time. The worst case in this algorithm is that every element must be checked to determine wher search item exists in array. If size of array is doubled, number of comparisons that algorithm must perform is also doubled. Linear search can provide outstanding performance if element matching search key happens to be at or near front of array. We seek algorithms that perform well, on average, across all searches, including those where element matching search key is near end of array. If a program needs to perform many searches on large arrays, it s better to implement a more efficient algorithm, such as binary search Binary Search The binary search algorithm is more efficient than linear search, but it requires that array be sorted. The first iteration tests middle element in array. If this matches search key, algorithm ends. If search key is less than middle element, algorithm continues with only first half of array. If search key is greater than middle element, algorithm continues with only second half. Each iteration tests middle value of remaining portion of array. If search key does not match element, algorithm eliminates half of remaining elements. The algorithm ends by eir finding an element that matches search key or reducing subarray to zero size Binary Search (cont.) In worst-case scenario, searching a sorted array of 1023 elements takes only 10 comparisons when using a binary search. The number 1023 (2 10 1) is divided by 2 only 10 times to get value 0, which indicates that re are no more elements to test. Dividing by 2 is equivalent to one comparison in binary search algorithm. Thus, an array of 1,048,575 (2 20 1) elements takes a maximum of 20 comparisons to find key, and an array of over one billion elements takes a maximum of comparisons to find key.

4 array, which is represented as log 2 n. All logarithms grow at roughly same rate, so in big O notation base can be omitted. This results in a big O of O(log n) for a binary search, which is also known as logarithmic run time. Sorting A difference Algorithms between an average of 500 million comparisons for linear search Sorting and a data maximum (i.e., placing of only data comparisons into some for particular binary order, search! i.e. ascending or descending) Binary is Search one of (cont.) most important computing applications The An important maximum item number to understand of comparisons about needed sorting for is that binary sorted search array of will any sorted be array same is no matter exponent which of algorithm first power you use of to 2 sort greater than array number of elements in The array, choice which of algorithm is represented affects only as log 2 n. run time and memory use of program All Selection logarithms Sort grow at roughly same rate, so in big O notation base can be omitted. The selection sort is a simple, but inefficient, sorting algorithm This results in a big O of O(log n) for a binary search, which is also known as Its first iteration selects smallest element in array and swaps it with first logarithmic run time. element. Sorting The second Algorithms iteration selects second-smallest item (which is smallest item of Sorting remaining data (i.e., elements) placing and swaps data into it with some particular second element order, i.e. ascending or Selection descending) Sort is one of most important computing applications An important item to understand about sorting is that sorted array will be The algorithm continues until last iteration selects second-largest element and same no matter which algorithm you use to sort array swaps it with second-to-last index, leaving largest element in last index The choice of algorithm affects only run time and memory use of program After th i iteration, smallest i items of array will be sorted into increasing Selection order in Sort first i elements of array: The After selection first sort iteration, is a simple, smallest but inefficient, element sorting is first algorithm position Its After first iteration second iteration, selects smallest two smallest element elements in array are in and order swaps in first it with two positions, first element. etc. The Selection second Sort iteration selects second-smallest item (which is smallest item of remaining elements) and swaps it with second element After first iteration, smallest element is in first position; after second iteration, Selection two smallest Sort elements are in order in first two positions, etc. The algorithm selection sort continues algorithm until runs in last O(n iteration 2 ) time. selects second-largest element and swaps it with second-to-last index, leaving largest element in last index After th i iteration, smallest i items of array will be sorted into increasing order in first i elements of array: After first iteration, smallest element is in first position After second iteration, two smallest elements are in order in first two positions, Insertion etc. Sort Selection The insertion Sort sort is anor simple although inefficient, sorting algorithm The first iteration takes second element in array and, if it s less than first After first iteration, smallest element is in first position; after second iteration, element, swaps it with first element two smallest elements are in order in first two positions, etc. The second iteration looks at third element The selection sort algorithm runs in O(n 2 and inserts it into correct position ) time. with respect to first two, so all three elements are in order Insertion At ithsort iteration of this algorithm, first i elements in original array will be The sorted insertion sort is anor simple although inefficient, sorting algorithm The first iteration takes second element in array and, if it s less than first element, swaps it with first element The second iteration looks at third element and inserts it into correct position InsertionSort.java with respect to first two, so (Page all three 2) elements are in order At ith iteration of this algorithm, first i elements in original array will be InsertionSort.java sorted (Page 3) Merge Sort (Page 1) The merge sort is an efficient sorting algorithm that conceptually is more complex than selection sort and insertion sorts Sorts an array by splitting it into two equal-sized sub-arrays, sorting each sub-array, n merging m into one larger array Merge Sort (Page 2)

5 41 selection sort and insertion sorts Sorts an array by splitting it into two equal-sized sub-arrays, sorting each sub-array, n merging m into one larger array Merge Sort (Page 2) The implementation of merge sort in this example is recursive The base case is an array with one element (which of course is sorted already), so merge sort immediately returns in this case Recursion step splits array into two approximately equal pieces, recursively sorts m, n merges two sorted arrays into one larger, sorted array