Leading the Way in ML, GenAI, and Research as SWE

Nimra Waqar • Peerlist

In this project, I developed a Cognitive Therapy and Routine Recommendation System, focusing on mental health support through personalized interventions. Leveraging NLP techniques and the Gemini chatbot API, the system engages users in cognitive therapy by analyzing responses to the PHQ-9 questionnaire, commonly used to assess depression severity.

Features:

✔️Utilized NLP to interpret user responses and provide tailored feedback, supporting cognitive restructuring.
✔️Offered personalized daily activities, videos, and articles, based on the user's mental health status and interests.
✔️Implemented a weekly tracking system to monitor mental health improvements, refining recommendations through data-driven insights.

Cognitive Therapy & Routine Recommender

✔️In this project, I developed an Email Spam Classification System to enhance security by filtering out unwanted and potentially harmful emails. 
✔️The system utilizes machine learning algorithms to accurately differentiate between legitimate emails and spam, protecting users from phishing attacks, malware, and other email-based threats.

Email Spam Classification

I developed a data-driven project focused on advancing gender equality in alignment with SDG 17. This project aims to harness the power of data analytics to address gender disparities and drive actionable strategies for improvement.

Key Features:

✔️Utilized comprehensive datasets to analyze gender disparities in education, employment, and other critical areas.
✔️Developed models to forecast gender equality indices based on socio-economic factors, providing insights into future trends and potential impact areas.
✔️Created dynamic dashboards for real-time visualization of gender-related metrics, allowing users to explore and understand data trends interactively.
✔️Generated reports and recommendations to guide policy-making and strategic initiatives towards closing gender gaps and promoting equality.

Equalize360

developed a comprehensive solution for fake news detection using both deep learning and traditional machine learning models. I began by preprocessing the dataset, which involved handling missing values, combining relevant text fields, and preparing the text for model training through tokenization and padding. I implemented an LSTM model to capture the temporal dependencies in the text, achieving a high accuracy of 98.37%. In addition to deep learning, I explored various traditional machine learning models like Random Forest, SVM, Naive Bayes, and Gradient Boosting to compare their effectiveness in detecting fake news. The Random Forest model achieved the highest accuracy among traditional models with 91.10%.

Fake News Detection

In this project, I developed an interactive IPL Data Analysis Dashboard using Python and its powerful libraries such as Plotly and Dash. The dashboard provides insightful visualizations on IPL matches, including metrics like matches per season, top batsmen, winner distribution, and venue analysis. The project involved extensive data manipulation with pandas, creating interactive charts using Plotly, and building a responsive web application with Dash.

IPL Data Analysis

In this project, I Implemented K-Means clustering on the well-known Iris dataset to classify iris flowers into distinct groups based on their sepal and petal measurements. 
✔️I utilized the Elbow Method to determine the optimal number of clusters, enhancing model performance. 
✔️To visualize the clusters effectively, I employed Principal Component Analysis (PCA) to reduce the dataset to two dimensions. This approach provided clear visual separation of the clusters and highlighted the centroids.

Iris Flower Clustering Using K-Means Algorithm and PCA

I developed a machine learning model to predict student academic performance using Random Forest Regression. The project involved comprehensive data preprocessing, including handling missing values and encoding categorical variables to prepare the dataset for analysis. I employed Random Forest Regression due to its robustness and ability to handle complex datasets with high dimensionality. The model was evaluated using Mean Squared Error (MSE) to ensure accurate predictions of students' final grades based on various features, such as attendance and prior academic records.

Predicting Student Academic Performance

In this project, I developed a predictive model to estimate student performance based on the number of study hours. 
✔️Using Python and scikit-learn, I implemented a Linear Regression model that predicts a student's score based on the hours they have studied. 
✔️The dataset was split into training and testing sets to evaluate the model's accuracy. 
✔️The model achieved a high R² score, indicating a strong correlation between study time and student scores.

Student Performance Prediction

In this project, I performed a comprehensive analysis of a Superstore's sales and profit data to uncover key insights and trends. 
✔️Using Python's powerful data manipulation and visualization libraries, I cleaned the dataset, handled missing values, and removed duplicates to ensure data integrity. 
✔️I then explored the distribution of sales and profit, analyzed relationships between key variables, and visualized total sales and profit across different categories and regions.