抽獎隨機算法的技術探討與C#實現

本文轉載自查看原文 2016-12-27 14:15 5216 經驗積累

一、模擬客戶需求

1.1 客戶A需求：要求每次都按照下圖的概率隨機，數量不限，每個用戶只能抽一次，抽獎結果的分布與抽獎概率近似。

1.2 客戶B需求：固定獎項10個，抽獎次數不限，每個用戶只能抽一次，抽完為止，抽獎結果必須是固定的那幾個獎項。

二、需求分析

算法1：對於客戶A，由於抽獎次數無限次，出獎分布遵守中獎概率設定值。所以必須在用戶每一次的抽獎行為之前都按照中獎概率隨機出就可以了，這樣隨着樣本的增多，概率分布越來越趨近於設定的抽獎概率。

算法2：對於客戶B，由於獎項固定，需要將獎項硬編碼（或者存儲到DB），獎項抽出來一個就將該獎項從列表中刪除。獎項就抽完后就直接返回未中獎。

三、算法實現

3.1先定義個類，用於存儲獎項和概率的關系：

//算法1.每次都完全隨機的算法
//作者：deepleo
//博客：http://www.deepleo.com/
//郵箱：yemor@qq.com
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;

namespace RandomAlgorithm
{
    /// <summary>
    /// 獎品
    /// </summary>
    public class Prize
    {
        /// <summary>
        /// 名稱
        /// </summary>
        public string Name { set; get; }
        /// <summary>
        /// 中獎概率
        /// </summary>
        public decimal Probability { set; get; }
    }
}

3.2 實現算法1

//算法1.每次都完全隨機的算法
//作者：deepleo
//博客：http://www.deepleo.com/
//郵箱：yemor@qq.com
using System;
using System.Collections.Generic;

namespace RandomAlgorithm
{
    public class EveryRandomAlgo
    {
        public List<Prize> Prizes
        {
            private set;
            get;
        }
        public EveryRandomAlgo(List<Prize> prizes)
        {
            this.Prizes = prizes;
        }


        /// <summary>
        /// 隨機下一次獎品
        /// </summary>
        /// <returns></returns>
        public string Next()
        {
            long tick = DateTime.Now.Ticks;
            Random ran = new Random((int)(tick & 0xffffffffL) | (int)(tick >> 32));
            var rnum = ran.Next(0, 9999);//將0~100映射到0~10000，提高精度到小數點后2位；生成隨機數rnum；
            //Console.WriteLine(rnum);
            var randomProbability = (decimal)rnum / 100;//再換算為0~100范圍
            var target = 0;//命中index
            var end = Prizes[0].Probability;
            for (int k = -1; k < Prizes.Count - 1; k++)
            {
                var min = k < 0 ? 0 : Prizes[k].Probability;//最小中獎概率
                if (randomProbability >= min && randomProbability <= end)//隨機出來的中獎概率位於中獎概率區間內
                {
                    target = k + 1;
                    break;
                }
                end += Prizes[k + 1].Probability;//最大中獎概率累計值
            }
            return Prizes[target].Name;
        }
    }
}

3.3 實現算法2

//算法2.從數據表中隨機取出一條記錄的偽隨機算法
//作者：deepleo
//博客：http://www.deepleo.com/
//郵箱：yemor@qq.com

using System.Collections.Generic;

namespace RandomAlgorithm
{
    public class DefineByDBAlgo
    {
        public List<Prize> Prizes
        {
            private set;
            get;
        }
        private Queue<string> _dbRecords;//存儲在數據表中的獎項記錄數據
        public DefineByDBAlgo(List<Prize> prizes)
        {
            this.Prizes = prizes;
            _dbRecords = new Queue<string>();
            //按照中獎概率初始化獎項數據
            _dbRecords.Enqueue(Prizes[0].Name);
            _dbRecords.Enqueue(Prizes[0].Name);
            _dbRecords.Enqueue(Prizes[0].Name);
            _dbRecords.Enqueue(Prizes[0].Name);
            _dbRecords.Enqueue(Prizes[0].Name);
            _dbRecords.Enqueue(Prizes[1].Name);
            _dbRecords.Enqueue(Prizes[1].Name);
            _dbRecords.Enqueue(Prizes[1].Name);
            _dbRecords.Enqueue(Prizes[1].Name);
            _dbRecords.Enqueue(Prizes[2].Name);
         
        }

        /// <summary>
        /// 隨機下一次獎品
        /// </summary>
        /// <returns></returns>
        public string Next()
        {
            if (_dbRecords != null && _dbRecords.Count > 0)
            {
                lock (_dbRecords)
                {
                    var random = _dbRecords.Dequeue();//這里直接取出最上面的那一條記錄。保證出獎順序與數據庫記錄一致，如果要隨機，只需要更改數據庫記錄順序或者這里使用隨機Index
                    return random;
                }
            }
            return "";
        }
    }
}

3.4 調用代碼

//摘要：抽獎隨機算法
//作者：deepleo
//博客：http://www.deepleo.com/
//郵箱：yemor@qq.com

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading;
using System.Threading.Tasks;

namespace RandomAlgorithm
{
    class Program
    {
        static void Main(string[] args)
        {

            var prizes = new List<Prize>();
            prizes.Add(new Prize { Name = "三等獎50%", Probability = 50 });
            prizes.Add(new Prize { Name = "二等獎40%", Probability = 40 });
            prizes.Add(new Prize { Name = "一等獎10%", Probability = 10 });
            var algo1 = new EveryRandomAlgo(prizes);
            var result1 = new Dictionary<string, int>();
            var count = 5000;
            for (int i = 0; i < count; i++)
            {
                var next = algo1.Next();
               // Console.WriteLine("【{0}】算法1隨機結果：{1}", i + 1, string.IsNullOrEmpty(next) ? "獎項已發完" : next);
                if (!result1.Keys.Any(x => x == next))
                {
                    result1.Add(next, 1);
                }
                else
                {
                    result1[next]++;
                }  
                Thread.Sleep(1);
            }
            foreach (var key in result1.Keys)
            {
                Console.WriteLine("算法1隨最終結果：{0}:{1}個", key, result1[key].ToString());
            }
            Console.WriteLine("=====================================");
            var algo2 = new DefineByDBAlgo(prizes);
            var result2 = new Dictionary<string, int>();
            for (int i = 0; i < count; i++)
            {
                var next = algo2.Next();
                //Console.WriteLine("【{0}】算法2隨機結果：{1}", i + 1, string.IsNullOrEmpty(next) ? "獎項已發完" : next);
                if (!result2.Keys.Any(x => x == next))
                {
                    result2.Add(next, 1);
                }
                else
                {
                    result2[next]++;
                }
            }
            foreach (var key in result2.Keys)
            {
                Console.WriteLine("算法2隨最終結果：{0}:{1}個", key, result2[key].ToString());
            }
            Console.ReadKey();
        }

    }

}