%all2data.m
%將UCI數據集中各種數據統一為數值型數據，符號型變為自然數，連續型進行標準化，決策屬性轉化為自然數排在最后一列
%暫不考慮缺失值
%一列總含有一個符號型數據則全部作符號型處理
%data_cell0:m*1
%data_cell1:m*n和data_array結構一樣
%sn_feature_d:決策屬性位置為空表示最后一行
%filename=‘te.txt‘;
%讀入數據到data_cell0然后將data_cell0拆分為data_cell1
%?data_cell0=importdata（filename）;
function?data_array=all2data（filenamesn_feature_d）
data_cell0=textread（filename‘%s‘）;
%?clear;clc；data_cell0=textread（‘ecoli.data‘‘%s‘）;sn_feature_d=[];
%?data_cell0=data_cell0（45:50:）%診斷原文件中數據是否格式有錯
data_cell1={};
for?i=1:length（data_cell0）
????dot_sequence=find（data_cell0{i}==‘‘）;
????interval=[0dot_sequencelength（data_cell0{i}）+1];
????for?j=1:length（interval）-1
????????data_cell1{ij}=data_cell0{i}（interval（j）+1:interval（j+1）-1）;
????end
end
%將data_cell1轉化為數字判斷是否為符號型是則轉換為1?2?3...不是則進行標準化
[mn]=size（data_cell1）;
%將決策屬性移到最后一行，其他屬性順序向前平移
if?length（sn_feature_d）==0
????sn_feature_d=n;
end
if?sn_feature_d~=n
????data_cell1（:n+1）=data_cell1（:sn_feature_d）;
????data_cell1（:sn_feature_d）=[];
end
for?j=1:n%按列處理
????%判斷是否為數值型
????isnum=1;%是否為數值型數據的標志
????i=1;
????value_num=[];%用來存放數值性數據
????while?i<=m????
????????i
????????j
????????a=str2num（data_cell1{ij}）;
????????if?length（a）==0
????????????isnum=0;
????????????i=m+1;
????????else
????????????value_num（i）=a;
????????????i=i+1;????????????
????????end
????end
????%對不同數據類型分別處理
????if?isnum==0%符號型數據
????????onefordiff={};
????????onefordiff（1）=data_cell1（1j）;%不同數據只存一個
????????for?i=1:m
????????????t=1;
????????????p=length（onefordiff）;??
????????????position=0;
????????????while?t<=p???????
????????????????a=strcmp（onefordiff{t}data_cell1{ij}）;
????????????????if?a==0
????????????????????t=t+1;
????????????????else
????????????????????position=t;
????????????????????t=p+1;?
????????????????end
????????????end
????????????if?a==0%說明是新數
????????????????onefordiff{p+1}=data_cell1{ij};
????????????????data_array（ij）=p+1;
????????????else
????????????????data_array（ij）=position;
????????????end
????????end
????else%數值型數據
????????%決策屬性為數值型時不進行標準化處理
????????if?j~=n
????????????value_max=max（value_num）;
????????????value_min=min（value_num）;
????????????if?（value_max-value_min）==0
????????????????data_array（:j）=0;%‘數據標準化有誤‘
????????????else
????????????????for?i=1:m
????????????????????data_array（ij）=（value_num（i）-value_min）/（value_max-value_min）;
????????????????end
????????????end
????????else
????????????data_array（:j）=value_num;
????????end????????
????end
end?
data_array

?屬性????????????大小?????日期????時間???名稱
-----------?---------??----------?-----??----

?????文件??????73289??2015-10-05?14:46??chengxu_hu\abalone.mat

?????文件???????2806??2013-01-24?17:39??chengxu_hu\all2data.m

?????文件?????235082??2015-10-05?16:37??chengxu_hu\biodegxiugai.mat

?????文件???????5599??2013-01-24?17:39??chengxu_hu\C4_5.m

?????文件???????5239??2013-07-15?10:35??chengxu_hu\cleve.mat

?????文件???????1535??2016-09-22?20:59??chengxu_hu\clsf_dpd.m

?????文件???????1568??2013-07-12?11:28??chengxu_hu\clsf_dpd_fast.m

?????文件???????1788??2015-02-13?21:14??chengxu_hu\clsf_dpd_fast2.m

?????文件???????1439??2013-07-12?11:40??chengxu_hu\clsf_dpd_fast_3.m

?????文件??????10719??2015-10-14?17:10??chengxu_hu\credit.mat

?????文件???????3984??2015-02-15?13:28??chengxu_hu\crossvalidate.m

?????文件???????3935??2015-02-24?00:34??chengxu_hu\crossvalidate1.m

?????文件???????3382??2015-11-19?20:42??chengxu_hu\featureselect_FW_fast.m

?????文件???????5109??2015-10-15?10:36??chengxu_hu\horse.mat

?????文件??????58276??2015-09-27?17:08??chengxu_hu\iono.mat

?????文件???????1054??2015-10-15?13:15??chengxu_hu\Iris.mat

?????文件???????1817??2015-02-20?19:51??chengxu_hu\KNN.m

?????文件???????1809??2013-04-12?21:00??chengxu_hu\NEC.m

?????文件???????2058??2013-07-15?11:13??chengxu_hu\nec1.m

?????文件???????1833??2013-07-15?10:35??chengxu_hu\osu_svm3.00\cardiag.asv

?????文件???????1837??2013-07-15?10:35??chengxu_hu\osu_svm3.00\cardiag.m

?????文件???????1728??2013-07-15?10:35??chengxu_hu\osu_svm3.00\cmap.mat

?????文件???????2905??2013-07-15?10:35??chengxu_hu\osu_svm3.00\Contents.m

?????文件???????1265??2013-07-15?10:35??chengxu_hu\osu_svm3.00\createset.m

?????文件???????8222??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\car.txt

?????文件???????5239??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\cleve.mat

?????文件??????18770??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\cleve.txt

?????文件??????33408??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\credict.txt

?????文件??????14916??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\crx.mat

?????文件?????177330??2013-07-15?10:35??chengxu_hu\osu_svm3.00\data\crx_fcm.dat

............此處省略141個文件信息