Tuesday 30 May 2017

Validating a Journal

To Validate a Journal
--go to journal and take issn number : number unique to the journal

ISSN 1947 5500 and paste it in Scopus.com and paste it without spaces
ex: 19475500 and select ISSN and press search

Ex: 21853118
256 documents found
DOCUMENTS HAVE SCOPUS NOT ONLY JOURNALS..SO 256 DOUCMENTS OF THAT JOURNAL ARE SCOPUS INDEXED

u will get document title , authors, years, source and cited
click on source International Journal of Intelligent Engineering and Systems

You can see a message as seen below indicating the journal is still in scopus
    Scopus coverage years:from 2008 to Present

Ex: 09745572 
wen u click on source u get
International Journal of Control Theory and Applications

    Scopus coverage years:from 2012 to Present(coverage discontinued in Scopus)

To check the list of journals in scopus go to below link
https://www.elsevier.com/solutions/scopus/content

if a paper is published in scopus journal, it will be checked by scopus in probation period of 4 weeks to 3 months and in this 3 months they will check and then they will put your paper in scopus. SCOPUS WILL CHECK UR PAPER WITH THEIR PLAGIARISM TOOLS

IF UR PAPER IS IN SCOPUS, IT WILL BE IN SCOPUS FOR LIFE TIME. NO MATTER IF JOURNAL IS REMOVED BY SCOPUS

if ur paper is found by plagiarism, and many authors are reporting against ur paper, the paper will be removed without intimation
================================================================
journals can also blacklist authors
================================================================
first author and corresponding author are different
first author is the name that will be displayed first in the paper
================================================================
Thomsan reuters gives impact factor and also provides indexing----

https://www.ijese.org/ has impact factor of 4.72
and elsevier https://www.journals.elsevier.com/computers-and-electrical-engineering/ has impact factor 1.084

https://sites.google.com/site/ijcsis/
has thomsan reuters and next to that it has researcher id...

u can create researcher id and can put it for your journal ..hence the journal is not CORRECT(Role:     Researcher (Academic))

===u can create researcher id in thomson reuters and can put your papers in it===

================================================================
TO CHECK WHETHER THE JOURNAL IS INDEXED IN THOMAS REUTERS
select the search type: ISSN
and type the 0045-7906 (ISSN of jounal) in the below isi master list link

http://ip-science.thomsonreuters.com/mjl/
================================================================
if u have published a paper in some journal and now want to publish your paper in some other journal, u can withdraw your paper from the journal and wait for 2 months and then can send ur paper to another journal.
================================================================

EVERY JOURNAL MUST BE IN 3 OR 4 YEARS TO GET IMPACT FACTOR



Friday 19 May 2017

Converting Excel data to learn Machine learning using Scikit-learn python

If you have suprevised data in EXCEL and you want to do machine learning on that data as given in the below link
http://scikit-learn.org/stable/auto_examples/model_selection/plot_roc.html

u need to convert the excel data into a format required in the link X, y arrays.
here in the link X is a 2 dimensional numpy float array and y is one dimensional numpy array


For ex: if u have data as shown below

fishlength       fishwidth   classlabel
1.2                   1.4                 1
1.5                   4.3                 2

We require X=array([[ 1.2 ,   1.4],[ 1.5,4.3]]) and y=array([1,2])
inorder to convert that data into the above format as required in the link

In the excel sheets remove the headers (fishlength ,fishwidth, classlabel) and cut the classlabel entire column and save it as .csv(comma delimited ) in the SAVE AS dialog box (for ex: here the filename is finaldata.csv)


and paste the classlabel column in seperate excel workbook using PasteSpecial and choosing transpose option in PasteSpecial.
Save this workbook with .csv(comma delimited ) in the SAVE AS dialog box(for ex: here the filename is label.csv)

type the below code:

import csv
import numpy as np
f=open("finaldata.csv")
X=np.genfromtxt(f,delimiter=",")
f.close()
f=open('label.csv')
csv=np.genfromtxt(f,delimiter=",")
y=csv.astype(int)
f.close()

#the below code is copied and pasted by removing nosiy features in above link

y = label_binarize(y, classes=[1, 2, 3])#array([[1, 0, 0],[1, 0, 0],...
n_classes = y.shape[1]#shape[1] gives columns (3 here) and y.shape[0] gives rows
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.5)#50% are taken as training data(X_train and those labels as y_train)
classifier = OneVsRestClassifier(svm.SVC(kernel='linear', probability=True))#creating classifier
y_score = classifier.fit(X_train, y_train).decision_function(X_test)#ysocre=array([[ -2.49503189e+00,   4.13933465e-01,   9.99997811e-01],...X_test value belongs to class 3
fpr = dict()
tpr = dict()
roc_auc = dict()
for i in range(n_classes):
    fpr[i], tpr[i], _ = roc_curve(y_test[:, i], y_score[:, i])#y_test has actual label information and y_score is predicted fpr=false positive rate
    roc_auc[i] = auc(fpr[i], tpr[i])

# Compute micro-average ROC curve and ROC area
fpr["micro"], tpr["micro"], _ = roc_curve(y_test.ravel(), y_score.ravel())
roc_auc["micro"] = auc(fpr["micro"], tpr["micro"])
plt.figure()
lw = 2
plt.plot(fpr[2], tpr[2], color='darkorange',
         lw=lw, label='ROC curve (area = %0.2f)' % roc_auc[2])
plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver operating characteristic example')
plt.legend(loc="lower right")
plt.show()

Tuesday 2 May 2017

Google Code Challenge : Counting Sheeps Solution in C


#include <stdio.h>
int main()
{
               int n, a[10],i,flag,n1,nc,r,j,t1,t;
               scanf("%d",&t);
               for(t1=1;t1<=t;t1++)
               {
                              nc=0;
                              for(i=0;i<10;i++)
                                             a[i]=-1;
                              scanf("%d",&n);
               for(i=0;i<45;i++)
               {
                              n1=n*(i+1);
                              while(n1>0)
                              {
                                             flag=0;
                                             r=n1%10;
                                             for(j=0;j<10;j++)
                                             {
                                                            if(a[j] == r)
                                                            {
                                                                           flag=1;
                                                                           break;
                                                            }
                                             }
                                             if(flag == 0 && nc < 10)
                                             {
                                                            a[nc]=n1%10;
                                                            nc++;
                                             }
                                             n1=n1/10;
                              }
                              if(nc == 10)
                              {
                                             printf("Case #%d:%d",t1,n*(i+1));
                                             break;
                              }
                                            
               }
               if(nc < 10)
                              printf("Case #%d:%s",t1,"INSOMNIA");
               }
              
}