查找匹配度最高的字符串（转）

要求在3W多个中,找出与给定的字符串匹配度最高的字符串.

匹配度计算规则:

匹配度=最长的相同字串的长度 / 两个字符串中最长的字符串的长度.

 

代码如下：

定义的一个主体类:

  public   delegate   decimal  CompareCondition(String value);

     public   class  StringCompare
     {
        private List<IntraString> intraList = new List<IntraString>();

        public StringCompare(List<String> strlist)
        {
            for (int i = 0; i < strlist.Count; i++)
            {
                intraList.Add(new IntraString(strlist[i]));
            }
        }

        public List<IntraString> Compare(CompareCondition condition)
        {
            for (int i = 0; i < intraList.Count; i++)
            {
                intraList[i].Compare(condition);
            }
            intraList.Sort(delegate(IntraString a, IntraString b)
            {
                if (a.CompareResult > b.CompareResult)
                {
                    return 1;
                }
                else if (a.CompareResult == b.CompareResult)
                {
                    return 0;
                }
                else
                {
                    return -1;
                }
            }
            );
            return intraList;
        }
    }

     public   class  IntraString
     {
        private String value;

        public IntraString(String value)
        {
            this.value = value;
        }

        public String Value
        {
            get { return value; }
            set { this.value = value; }
        }

        private decimal compareResult;

        public decimal CompareResult
        {
            get { return compareResult; }
            set { compareResult = value; }
        }

        public void Compare(CompareCondition condition)
        {
            CompareResult = condition(Value);
        }
    }

比较规则的实现:

    /**/ /// 

        ///定义比较的条件
        /// 

        /// 
        /// 
         private   static   decimal  Compare(String value)
         {
            string str = Guid.NewGuid().ToString().Replace("-", "");
            int MaxLength = str.Length;
            if (str.Length < value.Length)
            {
                MaxLength = value.Length;
            }
            bool find = false;
            decimal sameCount = 0;
            for (int i = 0; i < value.Length; i++)
            {
                string newStr = value.Substring(i, value.Length - i);
                for (int j = newStr.Length - 1; j >= 0; j--)
                {
                    if (str.IndexOf(newStr.Substring(0,j+1)) >= 0)
                    {
                        newStr = newStr.Substring(0, j + 1);
                        find = true;
                        break;
                    }
                }
                if (find)
                {
                    sameCount = newStr.Length;
                    break;
                }
            }
            return sameCount / MaxLength;
        }

转载于:https://www.cnblogs.com/yylqinghao/archive/2010/06/24/1764228.html