You are on page 1of 3

CBD 3356 Data Mining and Analysis

Project Proposal

Submitted By –

Tom James Madolil

Hardik Solanki

Katleen Ezekiel Orata

Mohammad Imran Uddin


Problem Statement

With unprecedented industrialization, there seems to be an every increasing demand

for higher productivity, better scores in schools and better overall performance. Such

demands put a massive toll on performances resulting in unbearable mental stress. However,

the availability of mental health practititioners are scanty at one hand, and difficult to afford

on the other.

Method

Text Analysis will be done on the Textual Data to check for keywords that denote

heavy mental stress and possible aggression. The pulled tweets will be subjected to various

NLP pre-processing techniques like word tokenization, stemming/lemmatization, stop word

removal, POS tagging. Words collected and cleaned will be used to train the model which

will be able to successfully predict the mental health status of a person for his/her tweet.

Intended experiments

Experiments intended to be performed during the length of the project includes Data

Pre-processing, Textual Analysis, Model Training using available tweets and Testing of

trained model using new tweets available.


Planning and Milestones.

High Level Timeline

Key milestones Begin Date End Date

Start of project 13 July 2022

Data 13 July 2022 20 July 2022

Gathering/Cleaning

Model Building 20 July 2022 27 July 2022

Model Deployment 27 July 2022 23 July 2022

Project Presentation 23 July 2022 06 August 2022

Project Revision 06 August 2022 20 August 2022

Project close 20 August 2022

You might also like