Go语言统计中文词语的个数,分词.

主要目的是统计一篇文章中的出现词语最多的个数,目前只是出个大概的框架,统计出来的结果只有两种,

一种是两个字的一种是三个字的,而且准确率不高,需要优化的地方太多,

先记下来,回头闲了继续优化一下,把准确率和处理速度提上来.

菜鸟写的不好,望大家多多指教.

</pre><pre name="code" class="plain">package main

import (
	"fmt"
	"io/ioutil"
	"strings"
)

var Tri map[string]int = make(map[string]int)
var Dou map[string]int = make(map[string]int)
var Num int = 2
var Path string = "test.txt"

func main() {
	Buf, _ := ioutil.ReadFile(Path)
	n := len(Buf)
	Str := string(Buf)
	Get_slice(n, Buf, &Str)
	List := Paixu(Dou)
	List_Tri := Paixu(Tri)
	Getvalue(Rm_duplicate(&List), Dou)
	Getvalue(Rm_duplicate(&List_Tri), Tri)
}

func Get_slice(n int, Buf []byte, Str *string) {
	var i, x, y int = 0, 0, 0
	var Three string
	for i < n-2 {
		if Buf[i] > 128 {
			x = i + 3
			for x < n-2 {
				if Buf[x] > 128 {
					s := string(Buf[i:i+3]) + string(Buf[x:x+3])
					if i := strings.Count(*Str, s); i > 1 {
						Dou[s] = i
					}
					y = x + 3
					for y < n-2 {
						if Buf[y] > 128 {
							Three = s + string(Buf[y:y+3])
							break
						} else {
							y++
						}
					}
					break
				} else {
					x++
				}
			}
			if i := strings.Count(*Str, Three); i > 1 {
				Tri[Three] = i
			}
			i = i + 3
		} else {
			i++
		}
	}
}

func Paixu(Map map[string]int) []int {
	var list []int
	for _, k := range Map {
		list = append(list, k)
	}
	num := len(list)
	for i := 0; i < num; i++ {
		for j := i + 1; j < num; j++ {
			if list[i] < list[j] {
				list[i], list[j] = list[j], list[i]
			}
		}
	}
	return list
}

func Getvalue(list []int, Map map[string]int) {
	for _, i := range list[:Num] {
		for k, v := range Map {
			if i == v {
				fmt.Println(k, "出现次数:", i)
			}
		}
	}
}

func Rm_duplicate(list *[]int) []int {
	var x []int = []int{}
	for _, i := range *list {
		if len(x) == 0 {
			x = append(x, i)
		} else {
			for k, v := range x {
				if i == v {
					break
				}
				if k == len(x)-1 {
					x = append(x, i)
				}
			}
		}
	}
	return x
}