Placement Data Analysis
Prediction of the most important aspects of the recruitment process.
Introduction: No two companies are the same. So it makes sense that when it comes to recruitment, every company does things slightly differently. Nevertheless, there are a few things that every recruiter should try to incorporate into their strategy.
Here we will explore some of the most important aspects of any recruitment process based on my Data Analysis and my own Machine Learning classifier’s prediction.
Dataset Overview: The dataset was obtained from Kaggle.It’s consists of Placement data of students in a college in India. It includes secondary and higher secondary school percentages and specialization. It also includes degree specialization, type, and Work experience, and salary offers to the placed students.
Part I: Data Visualization :
Firstly we going to find that if Gender affects the placements?
From the above plot, we can say that Male students are offered a slightly greater salary than females on average.
From the DataAnalysis Process, I explore that, We have samples of 139 Male students and 76 Female students.30 Female and 40 Male students are not placed. Male students have a comparatively higher placement. Male students are getting high CTC jobs.
Does Secondary Education affect placements?
We clearly see from the plot above that: All students with Secondary Education Percentage above 90% are placed, All students with Secondary Education Percentage below 50% are not-placed, Students with good Secondary Education Percentage are placed on average.
Board Of Education does not affect Placement Status much?
Outliers on both, but students from Central Board are getting highly paid jobs.No specific pattern (correlation) between Secondary Education Percentage and Salary. Board of Education is Not Affecting Salary.
Does Higher Secondary School affect Placements?
We can see that Overlap here too. More placements for percentage above 65%A straight drop below 60 in placements -> Percentage must be at least 60 for a chance of being placed Education Board again doesn’t affect placement status much.
what’s the effect of Specialization on Placements?
We can see those very fewer students with Arts specialization. Around 2:1 unplaced ratio for both Science and Commerce students.
Science&Tech students getting more salary on average, Management students are getting more highly paid dream jobs.
Does work Experience affect placements?
We can see that from both the plots that Very few students with work experience not getting placed!
Outliers (High salary than average) on both end but students with experience getting dream jobs, Average salary as well as base salary high for students with work experience.
HeatMap of Attributes affecting Placements and Salary :
Part II: Prediction from Machine Learning Models
Now we going to predicting If Students get placed or not based on the Decision Tree & Random Forest Classifier.
From the model we get that :
- Academic performance affects placement.
- Work Experience Effects Placement.
From Feature Importance of Tree-based Algorithms and MDA, we can say that:
- Academic performance affects placement (All percentages had importance)
- Work Experience Effects Placement
- Gender and Specialization in Commerce (in higher-secondary and undergraduate) also has an effect on placements.
Conclusion :
In from the article, we can conclude that :
- Having higher grades can be helpful while finding a job, but not crucial.
2. Work Experience Effects Placement.
3. Placements also depend on Specializations. STEM students get much better placements.
4. Gender doesn't matter in Placements. But we can say that In Commerce specialization it has little effect.
The findings here are observational, not the result of a formal study.
To see more about this analysis, see the link to my Github available here.
