Dataset available at https://www.kaggle.com/nsharan/h-1b-visa (see "Download(106 MB)")
numpy scipy matplotlib sklearn pandas
python h1bSystem.py (ensure h1b_history.csv in the same level folder )
-
def showCASE_STATUS(self,H1Info)
Analyze the H1Bs by the status of their visa applications and show the plot
-
def K_meanAnalyze(self,H1Info)
K-Means Clustering to seperate the h1b location
-
def showWORKSITE(self,dense,H1LatLong)
show the plot of the h1b WORKSITE after k-mean dense is the parameter to control distance between point, when gainning the info from the H1LatLong
-
def showSALARY_table(self,salaryMin,salaryMax,salaryMean,salaryMedian,salaryStd)
show the salary detail after k-mean process
-
def showSALARY_plot(self,salaryMedian,salaryMean)
Plotting and comparison to the median US salary (2015)
-
def showTOP6com_table(self,H1Info)
show top 6 company who has apply to the h1b for employee
-
def showYearTrend_plot(self,H1Info)
show the h1b number in everyear's change
-
def showJOBTITLE_plot(self,H1Info)
show top20 popular Jobtitle and top10 Worksites for H1-B Visa holders
-
def showAVGSalary_plot(self,H1Info)
show top20 salary mean Jobtitle
-
def showFullvsPart_plot(self,H1Info)
show the difference between fulltime job and part time job.
-
def predictCASE_STATUS(self,H1Info,job_TitleName,WORKSITE,EMPLOYER_NAME,PREVAILING_WAGE,dense)
CASE_STATUS situation predict. using decision Tree algorithm from sklearn
-
def testDECISION_ACURACY(self,H1Info,job_TitleName,dense=10)
test one specify job's accuracy in 2016 base on 2011-2015 data
-
def testDECISION_ACURACY(self,H1Info,job_TitleName,dense=10)
-
def testTOP10JOB_acuracy(self,H1Info)
get top10 popular job's accuracy in 2016 base on 2011-2015