Pradyoth S P

This project involves building a big data pipeline to source, process, and visualize data. The pipeline consists of multiple steps, including data sourcing using Python, Kafka for data streaming, Apache Spark for ETL processing, and Tableau for data visualization.

The "World Energy Consumption Visualized" project uses interactive visualizations to explore the dynamic connection between global energy consumption and economic prosperity, offering insights through D3.js and Our World in Data's dataset.

1. Developed a Backorder Prediction system using machine learning techniques, achieving high performance with accuracy of 86.33%, recall of 80.65%, and precision of 6.05%.
2. Trained and evaluated three models: Decision Tree, Random Forest, and Light GBM, with the Random Forest model yielding the best results, achieving accuracy of 90.03%, recall of 80.77%, and precision of 18.78% on the test dataset.
3. Implemented batch prediction capability, allowing for efficient prediction on… Show more

1. Developed a Backorder Prediction system using machine learning techniques, achieving high performance with accuracy of 86.33%, recall of 80.65%, and precision of 6.05%.
2. Trained and evaluated three models: Decision Tree, Random Forest, and Light GBM, with the Random Forest model yielding the best results, achieving accuracy of 90.03%, recall of 80.77%, and precision of 18.78% on the test dataset.
3. Implemented batch prediction capability, allowing for efficient prediction on multiple product records using a CSV file.
4. Provided a user-friendly form interface for single product prediction, streamlining the prediction process for individual products.
5. Built the application using Django, scikit-learn, and LGBM, utilizing industry-standard tools and technologies for efficient development and deployment. Show less

This project aims to use k-means and Agglomerative clustering to segment customers into different groups based on their characteristics and purchasing habits. The goal is to understand the similarities and differences between the customer segments, which can help inform marketing strategies and target specific groups of customers.

Skills: Python, K-Means Clustering, Agglomerative Clustering, PCA, Elbow Method, Feature Engineering.

1. This project aims to demonstrate the use of embarrassingly parallel computing techniques for detecting persons in video frames.
2. In this project, we will use the parallel processing capabilities of a multi-core processor to speed up the process of detecting persons in video frames. We will be using a pre-trained deep learning model for object detection, specifically the YOLOv3 model.

– Responsible for the coding and circuit design of a solar mobile charger using NodeMCU.
– Improvised the design and functionality of the device to automatically change and track the angle of solar panels.

– Analyzed dataset (containing over 200 thousand rows) to determine acceptance rate, using NLP vectorizers such as BOW (Bag of Words) and TF-IDF (Term Frequency & Inverse Term Frequency), and Naive Bayes.
– Achieved AUC (Area Under the Curve) score of 0.71.

– Trained dataset and implemented face detection system using OpenCV. This system can unlock a phone or door using face identification.
– Achieved an accuracy of ~86%.

Dataset used is "Haberman's Survival Data Set"
Link - https://www.kaggle.com/gilsousa/habermans-survival-data-set

Description of Dataset:
The dataset contains cases from a study that was conducted between 1958 and 1970 at the University of Chicago's Billings Hospital on the survival of patients who had undergone surgery for breast cancer.