GO语言利用K近邻算法实现小说鉴黄
Usuage:
go run kNN.go --file="data.txt"
关键是向量点的选择和阈值的判定
样本数据来自国家新闻出版总署发布通知公布的《40部淫秽色情网络小说名单》
package main
import (
"bufio"
"flag"
"fmt"
"io"
"log"
"math"
"os"
"path"
"path/filepath"
)
var debug bool = false
var data_dir string = "./moyan" //文件存放目录
var limen float64 = 0.1159203888322267 //阈值
const (
MIN_HANZI rune = 0x3400
MAX_HANZI rune = 0x9fbb
)
var labels []rune = []rune{
0x817f, 0x80f8, 0x4e73, 0x81c0,
0x5c41, 0x80a1, 0x88f8, 0x6deb,
}
func errHandle(err error) {
if err != nil {
log.Fatal(err)
}
}
func load(name string) (m map[rune]int, err error) {
f, err := os.Open(name)
if err != nil {
return nil, err
}
defer f.Close()
buf := bufio.NewReader(f)
m = make(map[rune]int)
var r rune
for {
r, _, err = buf.ReadRune()
if err != nil {
if err == io.EOF {
break
}
return nil, err
}
if r >= MIN_HANZI && r <= MAX_HANZI {
m[r] += 1
}
}
return m, nil
}
func classify(m map[rune]int) (idv []float64, dis float64) {
len_m := len(m)
for i, v := range labels {
if debug {
fmt.Println(i, m[v], string(v), float64(m[v])/float64(len_m))
}
idv = append(idv, float64(m[v])/float64(len_m))
}
for _, v := range idv {
dis += math.Pow(v, 2)
}
dis = math.Sqrt(dis)
return
}
func check(fp string, dis float64) {
switch {
case dis >= limen:
fmt.Println(fp, dis, "涉黄")
case dis == 1.0:
fmt.Println(fp, dis, "你在作弊吗")
case dis == 0:
fmt.Println(fp, dis, "检查一下文件字符编码是不是utf8格式吧")
default:
fmt.Println(fp, dis, "正常")
}
}
func walkFunc(fp string, info os.FileInfo, err error) error {
if path.Ext(fp) == ".txt" {
m, err := load(fp)
errHandle(err)
_, dis := classify(m)
check(fp, dis)
}
return err
}
var file string
func init() {
_, err := os.Stat(data_dir)
if err != nil {
err = os.Mkdir(data_dir, os.ModePerm)
errHandle(err)
}
flag.StringVar(&file, "file", "", "file read in,if you don't give the file read in,"+
"it will create a data dictionary,just pust your files in it")
}
func main() {
flag.Parse()
if file == "" {
filepath.Walk(data_dir, walkFunc)
return
}
m, err := load(file)
errHandle(err)
_, dis := classify(m)
check(file, dis)
}
以上所述就是本文的全部内容了，希望大家能够喜欢。
赞 (0)