Question

我已编写此代码以生成两个项目集。

我正在编写list<list<>>列表列表以生成两个项集，并希望优化代码以生成长度为k的项集。

using System.Data.OleDb;
using Microsoft.AnalysisServices.AdomdClient;

namespace HW3
{
    class Program
    {
        private static int Min_support = 360;
        private static List<String> itemSet = new List<string>();
        private static List<List<String>> C_itemSet = new List<List<string>>();

        static void Main(string[] args)
        {
            List_Movies();
            //for (int k = 1; k < 3; k++)
            //{ }
            C_itemSet = Geneate_Sets(itemSet);

            for (int i = 0; i < C_itemSet.Count; i++)
            {
                for (int j = 0; j < C_itemSet[i].Count; j++)
                {
                    Console.Write(C_itemSet.ElementAt(i).ElementAt(j));
                }
                Console.WriteLine();
            }

            Console.ReadKey();
        }

        private static void List_Movies()
        {
            string connectionString = @"Provider=SQLOLEDB;Server=.;AttachDbFilename=D:\DBs\MovieClick.mdf;Database=dbname; Trusted_Connection=Yes;";
            OleDbConnection myConn = new OleDbConnection(connectionString);
            myConn.Open();
            string str = "select Movie,COUNT(Movie) from Movies group by Movie HAVING COUNT(MOVIE) > "+Min_support;
            OleDbCommand myComm = new OleDbCommand(str, myConn);
            OleDbDataReader DR = myComm.ExecuteReader();
            while (DR.Read())
            {
                if (DR[0] != null)
                    itemSet.Add(DR[0].ToString()); // Movie name
            }

            myConn.Close();
        }

        private static List<List<String>> Geneate_Sets(List<String> fSet )
        {
            List<List<String>> finalSet= new List<List<string>>();
            for (int i = 0; i < fSet.Count; i++)
            {
                for (int j = i+1; j < fSet.Count; j++)
                {
                    if (!fSet[i].Equals(fSet[j]))
                    {
                        List<string> temp = new List<string>();
                        temp.Add(fSet[i]);
                        temp.Add(fSet[j]);

                            finalSet.Add(temp);
                    }
                }
            }
            return finalSet;
        }

Answer 1

除非您的数据确实非常庞大，否则我建议您将其加载到已经正确实施Apriori的良好库中。

因为根据您的代码判断，您实际上远非Apriori实现远，尤其是实际表现良好的实施，抱歉。就是这样，你似乎没有APRIORI的任何东西。不要指望我们为你编写算法。

如何为apriori算法生成数据集？

1 个答案: