@article{MAKHILLJEAS2020151419370,
    title = {Recommendation System for Predicting the Placement Percentage for an Educational Institute},
    journal = {Journal of Engineering and Applied Sciences},
    volume = {15},
    number = {14},
    pages = {2817-2826},
    year = {2020},
    issn = {1816-949x},
    doi = {jeasci.2020.2817.2826},
    url = {https://makhillpublications.co/view-article.php?issn=1816-949x&doi=jeasci.2020.2817.2826},
    author = {Varul and},
    keywords = {Data mining,machine learning,neural network,data cleaning,data set,data distribution,Training,Logistic Regression,Support Vector Machine (SVM),K-Nearest Neighbors (KNN),Decision Tree,Artificial Neural Network (ANN),T-Distributed Stochastic Neighbor Embedding},
    abstract = {Engineering students are dubious about what
they need to pursue after graduation. With extensive
options available, starting from campus recruitments to
Masters, students are perplexed, adding factors like
salaries and different job opportunities makes it even
worse. There aren’t any reliable platforms where a student
can predict the outcomes from the beginning of
engineering and take action to bridge this gap for a far
better future. Placement of students is one of the vital
activities in academic establishments. Admission
primarily depends on placements. Admission is directly
proportional to the offer letters received by an institute.
Hence all institutions strive to strengthen the placement
department. Students studying in engineering colleges feel
the difficulty to understand where they substitute
comparison to others and what quite a placement they
might get. The training and placement offices are
available to an image when a student enters the final year
but they’re of no use to a student planning for future
studies. Prediction about the student’s performance is an
integral part of an education system because the overall
growth of the scholar is directly proportional to the
success rate of the scholars in their examinations and
extra-curricular activities. Therefore, there are many
situations where the performance of the scholar must be
predicted for instance in identifying weak performing
students and taking actions for his or her betterment. The
students do not have any platform to see their current
position and repose on their strengths. The platforms
currently available haven’t been trained on real and
complete data sets and don’t learn from their wrong
predictions which reduces the accuracy within the future.
To realize far better efficiency and a system that
determines with every wrong prediction it’s made, so it
uses algorithms like Logistic Regression, Support Vector
Machine (SVM), K-Nearest Neighbors (KNN) which can
cause endless accuracy growth. The model is going to
train on a real data set and a massive number of
qualitative as well as quantitative parameters are going to
consider. This study aims to study the previous year’s
student’s data and predict the placement possibilities of
current students and aids in increasing the situation
percentage of the institutions. This study presents a
recommendation system that predicts whether the present
student is going to be placed or not with a percentage
value. This study helps the placement cell at intervals to
identify potential students and concentrate on and
improve their technical and social skills.}
    }