WeatherWizard

This project implements a weather prediction machine learning algorithm using the K-Nearest Neighbors (K-NN) algorithm in Python. The dataset for training the model was curated from the Wunderground Repository using a RESTful API. The algorithm predicts weather conditions based on historical data and features such as temperature, humidity, and wind speed.

Key Features

• Accurate weather prediction using K-NN algorithm.
• Data curated from the Wunderground Repository using a RESTful API.
• Manual data cleaning and preprocessing for improved prediction accuracy.
• Visualization and analysis of the dataset using R.

Project Overview

Data Collection

The dataset used for training the model was collected from the Wunderground Repository via a RESTful API. This dataset includes historical weather data such as temperature, humidity, wind speed, and more.

Data Cleaning and Preprocessing

1. Manual Data Cleaning: The dataset underwent manual data cleaning to address missing values, outliers, and inconsistencies.
2. Feature Selection and Extraction: Features were selected based on their relevance to weather prediction. This process involved careful analysis and resulted in a 15% improvement in prediction accuracy.

Model Implementation

The K-NN algorithm was implemented using the scikit-learn library in Python. The model was trained on the preprocessed dataset to predict weather conditions accurately.

Visualization and Analysis

Before data preprocessing, the dataset was visualized and analyzed using R. This step provided insights into the distribution of data, patterns, and relationships between variables.