Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 18

VISHWAKARMA INSTITUTE OF TECHNOLOGY

SUB - DATA SCIENCE

CH-A
BATCH-4
GROUP-3
PROJECT GUIDE:-
03.Darshan Khamat
06.Priyanka Nagpure Prof.Ranjeet.S.Suryavanshi
14.Sankalp Salunke
19.Jayshri Vishwakarma
"STUDENT
MENTAL HEALTH
ANALYSIS AND
PREDICTION"
Student mental health analysis

Perform exploratory data analysis (EDA) to understand the characteristics


and distribution of mental health-related variables in the dataset.

Identify patterns and trends in student mental health based on demographic


and academic factors.

Objective Build an interactive Shiny web application to allow users (potentially your
professor and peers) to explore the dataset visually and interactively.

Include features for selecting different parameters and observing the


predicted probability of depression.

Train a logistic regression model to predict the likelihood of students


experiencing depression.
PROBLEM STATEMENT

• Limited understanding of mental health trends in


university students.
• Absence of personalized tools for mental health
assessment.
• Challenges in identifying and addressing
depression factors.
.
• Tidyverse: Comprehensive suite for data
manipulation and visualization.

IMPORT • Dplyr: Data manipulation and


summarization tool.
LIBRARIES
• Ggplot2: Data visualization package.
imports essential R libraries

• Readr: Efficient CSV reading utility.


• Caret: Streamlined machine
learning and model evaluation.

• Shiny: R package for creating


IMPORT interactive web applications.

LIBRARIES • plotly - Creating interactive plots.


imports essential R libraries
• scales - Additional scale functions
for graphics.
LOAD DATA
Load data from
.csv
file
Write a note here

Write a
Assign Yes and note here
NO with 0 and
1 Value

• Load data from 'Student Mental health.csv'.


• Fill missing values in the dataset.
• Create a new 'ID' column .
• This preprocessing prepares the data for further analysis and model training.
steps for data cleaning:
Reading the Dataset: read.csv( )

 Checked for missing values apply( ) and is.na( )


Handling Missing Values  Replaced missing values median( )

Renaming Columns: names( )

Adding ID Column: tibble::rowid_to_column( )

Data Summary: summary( )


EXCEPTED
OUTCOME
• Gain insights into the prevalence of mental health issues
among students.

• Develop a reliable predictive model for identifying


potential cases of depression.

• User-friendly Shiny app for interactive exploration and


prediction.
EDA
1.Gender distribution

Fig. Pie chart of gender distribution


Fig. Depression VS Gender
2.Age distribution:

FIig.Age distrubtiont
3.Course and Year distribution:

Fig.Year distributionn Fig.Depression amongst courses


LOGISTIC
REGRESSION
MODEL
CHOOSE RELEVANT VARIABLES FOR HANDLED MISSING VALUES,
PREDICTING DEPRESSION, SUCH OUTLIERS, AND IRRELEVANT
AS AGE, GENDER, COURSE, YEAR FEATURES.
OF STUDY, CGPA

UTILIZES A LOGISTIC Trained the logistic regression model on th


key variables: REGRESSION MODEL FOR training dataset.
1.Gender TRAINING.

2.Age
3.Course Converted categorical variables (e.g.,

4.Year course and year) into factors for model


compatibility
5.CGPA
6.Depression
SHINY APP
Offers a user-friendly interface for easy interaction with the student mental health exploration.

Shiny app interface Output:graph showing student


having depression .
Depression distribution

FIg.pie chart for depressiont


FUTURE SCOPE

1.Model Improvement:

2.Data Enrichment:

3.User Feedback Integration:

4.Real-time Data Monitoring:

5.Mental Health Professionals:


CONCLUSION
Insightful Data Exploration:

Shiny App Impact:

Data Preprocessing Impact:

Contributions to Mental
Health Awareness:

Model Performance:
THANK YOU

You might also like