Skip to main content

Hash-based Search

An example of a hash-based search, although using dictionaries.  The code is designed to manage lists of Person objects, with unique Id's as the key value, with a rudimentary hash function to assign to various lists.

Salient Characteristic(s)
  • Useful for managing large unsorted lists
  • Divides the items into smaller hash-based buckets (divide and conquer)
  • Generally excellent performance
  • Sensitive to the choice of hash function and data structures chosen
Note
  • The Id key is not rigorous enough to guarantee uniqueness, and a better key might need to be selected, e.g., SSN, to prevent the same person being entered twice.
Code

using System;
using System.Collections.Generic;

namespace Algorithms
{
    //Example object to add and find in Hash Search example
    public class Person
    {
        private int _Id;
        public int Id 
        {
            get { return _Id; }
            set { _Id = value; } 
        }
        
        private string _LastName;
        public string LastName 
        {
            get { return _LastName; }
            set { _LastName = value; } 
        }
        
        private string _FirstName;
        public string FirstName 
        {
            get { return _FirstName; }
            set { _FirstName = value; } 
        }
        
        private DateTime _DateOfBirth;
        public DateTime DateOfBirth 
        {
            get { return _DateOfBirth; }
            set { _DateOfBirth = value; } 
        }
    }

    public class HashSearch
    {
        Dictionary<int, Dictionary<int, Person>> _HashList = new Dictionary<int, Dictionary<int, Person>>();
        int _BucketCount = 10;
        public HashSearch()
        {
            //create example list of persons
            IList<Person> arrayToSort = CreatePersonArray();
            
            //create bucket using hash, x/3
            CreateHashedArray(arrayToSort);
        }

        //code to create artificail list for example
        public IList<Person> CreatePersonArray()
        {
            IList<Person> arrayToSort = new List<Person>();
            int[] idNumbers = { 11, 7, 22, 2, 33, 3, 17, 44, 4, 55, 5, 66, 6, 1, 77 };
            foreach (int id in idNumbers)
            {
                Person tempPerson = new Person();
                tempPerson.Id = id;
                arrayToSort.Add(tempPerson);
            }
            return arrayToSort;
        }

        /// <summary>
        /// Helper for this example: Loads predefined array into internal lists
        /// </summary>
        /// <param name="arrayToSort"></param>
        private void CreateHashedArray(IList<Person> arrayToSort)
        {
            foreach (Person item in arrayToSort)
            {
                Add(item);
            }
        }

        /// <summary>
        /// Given a person Id, return person or null
        /// </summary>
        /// <param name="num"></param>
        /// <returns></returns>
        public Person Find(int num)
        {
            Person itemFound = null;
            int hash = CommonMethods.hashFunction(num, _BucketCount);
            Dictionary<int, Person> tempList;
            if (_HashList.TryGetValue(hash, out tempList))
            {
                foreach (KeyValuePair<int, Person> item in tempList)
                {
                    if (item.Key == num)
                    {
                        itemFound = item.Value;
                    }
                }
            } 
            return itemFound;
        }

        /// <summary>
        /// Given an Id, attempts to add, if item already exist, returns an error
        /// </summary>
        /// <param name="item"></param>
        public void Add(Person item)
        {
            int hash = CommonMethods.hashFunction(item.Id, _BucketCount);
            Dictionary<int, Person> tempList;
            if (!_HashList.TryGetValue(hash, out tempList))
            {
                tempList = new Dictionary<int, Person>();
                _HashList.Add(hash, tempList);
            }
            _HashList[hash].Add(item.Id, item);
        }       
    }
}

Helper Code

        public static int hashFunction(int num, int buckets)
        {
            return num / buckets;
        }

Example Usage

using System;

namespace Algorithms
{
    class Program
    {
        static void Main(string[] args)
        {
            HashSearch list = new HashSearch();

            //Verification, code to find items, and if not found, add
            Person find;
            for (int counter = 0; counter < 50; counter++)
            {
                find = null;
                find = list.Find(counter);

                if (find != null)
                {
                    Console.WriteLine(String.Format("Found ID: {0}", find.Id));
                }
                else
                {
                    Person newItem = new Person();
                    newItem.Id = counter;
                    list.Add(newItem);
                    Console.WriteLine(String.Format("Added ID: {0}", counter));
                }
            }

            //Error example: Adding an existing Person item
            try
            {
                Person errorItem = new Person();
                errorItem.Id = 11;
                list.Add(errorItem);
            }
            catch (Exception ex)
            {
                Console.WriteLine(String.Format("Error: {0}", ex.Message));
            }
        }
    }
}

Comments

Popular posts from this blog

Heap Sort

Heap Sort is interesting because the use of the 'heapify' method that creates a binary tree as a flat array. First, a binary tree, also known as a heap, is created, and then the same function is used to sort the elements. In a heap, the first node in the array at zero (0) is the top node of the binary tree. The next two (2) items are the two (2) subnodes of the top node, and so on. For each node at a position (positionIndex), its 2 subnodes are in the following positions:

int left = 2 * positionIndex + 1;int right = 2 *  positionIndex + 2;
The CommonMethods.Swap() function is a reusable class, since swapping values by position is a common action during these example sorts.

Class

using System;

namespace Algorithms
{
    class HeapSort
    {
        public int[]  Sort()
        {
            //creates array
            int[] arrayToSort = { 11, 1, 22, 2, 33, 3, 44, 4, 55, 5, 66, 6, 7, 77, 88 };

            //creates a heap
            BuildHeap(arrayToSort);

            //sorts a heap
         …

Binary Search

This is the basic divide-and-conquer algorithm, requiring a sorted array. This repeatedly halves the remaining positions to search, until the value is found, or it determines that the value cannot be found. As mentioned, it requires a presorted array, or if unsorted, requires it is sorted. The benefits accrue in instances where the array is sorted once, but searched many times.

On a hypothetical array of 100 items, a linear search has an average performances of O(n/2), or a search of half the positions, 50, to find the item. By comparison, the average performance of Binary Search is O(log n), which for 100 items translates into 4.6, a radically reduced number of iterations, but with greater requirements for logic and memory.

Done in both C#, as an iterative while loop, and F#, as a recursion.

C#
using System; namespace Algorithms { class BinarySearch { public BinarySearch() { } public int Search(int num, int[] arrayToSearch) { …

Bucket Sort

This following uses the divide-and-conquer method of resolution, by dividing the problem into smaller arrays, sorting each smaller array, then reinserting the subarrays back into the original array.  This is not a pure solution, in that it uses the .NET List, rather than a linked list struct.

using System;
using System.Collections;
using System.Collections.Generic;

namespace Algorithms
{
    class BucketSort
    {
        public int[] Sort()
        {   
            //creates arbitrary array
            int[] arrayToSort = { 11, 7, 22, 2, 33, 3, 17, 44, 4, 55, 5, 66, 6, 1, 77 };
            //create bucket using hash, x/10
            List<int>[] bucketList = CreateHashedArray(arrayToSort, 10);
            //sort the bucket list back into the original array
            ReorderList(bucketList, arrayToSort);
            return arrayToSort;
        }

        private int hashFunction(int num, int buckets)
        {
            return num / buckets;
        }

        private List<int>[] CreateHashed…