Академический Документы
Профессиональный Документы
Культура Документы
Result
Prediction Training
Evaluation
Website
Designing
DATA COLLECTION & CLEANING
Final
Dataset :
Data Visualisation
Support Vector Machine ?
• SVM or Support Vector Machine is a
Supervised learning algorithm for
classification and regression problems.
• The idea of SVM is simple: The
algorithm creates a line or a hyperplane
which separates the data into classes.
• In two dimentional space this
hyperplane is a line dividing a plane in
two parts.
Naïve Bayes ?
• Naive Bayes is a Supervised machine learning algorithm based
on the Bayes Theorem, used in a wide variety of classification
tasks.
• It is a classification technique with an assumption of
independence among predictors.
• Applications : filtering spam, classifying documents, sentiment
prediction etc.
• The name naive is used because it assumes the features that
go into the model is independent of each other.
K-Nearest Neighbors ?
• The k-nearest neighbors (KNN) is a
supervised machine learning algorithm
that can be used to solve both
classification and regression problems.
• The KNN algorithm assumes that
similar things exist in close proximity.
• application : pattern recognition,
recommendation systems and
decision-making models.
Artificial Neural Network(ANN) ?
Artificial Neural Networks(ANN) is
a computational model that is
based on human neural network.
ANN consists of multiple layers.
This network architecture includes
an input layer, one or more hidden
layer(s) and the output layer.
Faster in processing information.
Applications : Medical Diagnosis,
Voice Recognition, Financial
Forecasting etc.
Result
Machine Learning Accuracy
Algorithm
Support Vector Machine 85.5%
As we have seen throughout our studies, that the problem statements we have
approached are student, college, and corporate centric. The solution to all of
these problem statements, is based on the model we are going to build, the
output of which will be a number between 0-1, which will determine, the
prediction of a student being placed. During this process, a lot of other
dependent variables will be predicted which will help solve the problem
statements. The expected outputs of the system for student end, is the prediction
about their placement, and the statistics of how they can fair well. College end
will have the analysis of every student, and will have the opportunity to focus
more on the improvement of students.
FUTURE SCOPE
• When the research work and the different models combined together and
given a web application to be accessed the project will be of use to all the
people.
• The continual additions of data will cause the models to work more efficiently
but fine tuning of the models with more qualitative parameters is necessary.
• The models are exposed to the risk of overfitting in the future and hence the
parameters can change their co relations and hence retraining the models is
important.
• A complete application where in a centralised database and the internship
data of the student can be kept for mining processes to better understand
the students and provide valuable guidance to students.
Thank You