Analysis of Household Dataset

Overview

This project analyzes the "Individual Household Electric Power Consumption" dataset from the UCI Machine Learning Repository. The analysis involves statistical methods, hypothesis testing, regression, clustering, and dimensionality reduction techniques to gain insights into household electricity consumption patterns.

Dataset

• Source: UCI Machine Learning Repository

• Timeframe: 4 years of electric power consumption data

• Attributes: Various electrical parameters such as active power, reactive power, voltage, and current

Analysis Phases

Phase 1: Statistical Analysis

• Population Sampling & Hypothesis Testing:

  • Creating a normal population from the dataset

  • Extracting samples and comparing variances between attributes

• Regression Analysis:

  • Identifying linear relationships between variables

Phase 2: Machine Learning Techniques

• Dimensionality Reduction & Clustering:

  • Applying Principal Component Analysis (PCA)

  • Implementing clustering techniques for better data interpretation

• Analysis of Variance (ANOVA):

  • Comparing means of specific characteristics across different groups

Resources:

• GitHub Link

• Report 1

• Report 2

• Dataset