This repository contains a diverse collection of machine learning and data science datasets covering various domains including sports analytics, environmental monitoring, energy efficiency, and sensor data.

Files: ARM-Metric-train-TS.csv, ARM-Metric-test-TS.csv

Description: Time series data containing RSS (Received Signal Strength) measurements from multiple sensors for activity recognition tasks.

Features:

• avg_rss12, var_rss12: Average and variance of RSS between sensors 1-2
• avg_rss13, var_rss13: Average and variance of RSS between sensors 1-3
• avg_rss23, var_rss23: Average and variance of RSS between sensors 2-3
• ATYPE: Activity type (e.g., "bending")
• Multiple time windows with suffix _1, _2, _3, _4, _5

Use Cases: Activity recognition, IoT sensor analysis, time series classification

File: CPP_base_ampliado.csv

Description: Comprehensive power plant dataset with environmental and operational parameters for energy efficiency prediction.

Features:

• AT: Atmospheric Temperature
• V: Exhaust Vacuum
• AP: Atmospheric Pressure
• RH: Relative Humidity
• PE: Net Electrical Energy Output (target variable)
• Interaction features: AT-V, AT-AP, AT-RH, AT-PE, AP-RH, AP-PE

Use Cases: Energy efficiency prediction, power plant optimization, regression analysis

File: DM-Metric-missing-3.csv

Description: Multi-sensor environmental monitoring data with intentional missing values for data imputation research.

Features:

• Indoor Sensors: Temperature, CO2, Humidity, Lighting (Comedor & Habitacion)
• Weather Data: Precipitation, Wind, Solar radiation, Exterior temperature/humidity
• Temporal: Date, Time, Day of Week
• Environmental: Crepusculo, Piranometro, Entalpic measurements

Use Cases: Data imputation, environmental monitoring, time series analysis, missing data research

File: ENB2012_data-Y1.csv

Description: Building energy efficiency dataset from the ENB2012 study, focusing on heating and cooling load prediction.

Features:

• X1 to X8: Building characteristics (compactness, surface area, wall area, roof area, height, orientation, glazing area, glazing distribution)
• Y2: Heating load (target variable)
• VAL: Validation indicator

Use Cases: Building energy efficiency, heating load prediction, architectural optimization

Files: football.train2-r.csv, football.test2-r.csv

Description: Comprehensive football match statistics and results data for sports analytics and prediction modeling.

Features:

• Match Info: Division, Date, Home/Away teams
• Scores: Full-time and half-time goals (FTHG, FTAG, HTHG, HTAG)
• Results: Full-time and half-time results (FTR, HTR)
• Statistics: Shots, shots on target, fouls, corners, cards
• Performance Metrics: Recent form indicators (res1, res5, res20)
• Target Variables: HomeWin, ScoreDraw

Use Cases: Sports betting, team performance analysis, match outcome prediction

• Python 3.7+
• pandas, numpy, matplotlib, seaborn (for data analysis)
• scikit-learn (for machine learning tasks)

# Clone the repository
git clone <repository-url>
cd datasets

# Install required packages
pip install pandas numpy matplotlib seaborn scikit-learn

import pandas as pd
import matplotlib.pyplot as plt

# Load a dataset
df = pd.read_csv('ENB2012_data-Y1.csv')
print(f"Dataset shape: {df.shape}")
print(f"Columns: {df.columns.tolist()}")
print(f"First few rows:\n{df.head()}")

# Basic statistics
print(df.describe())

# Check for missing values
print(df.isnull().sum())

# Visualize distributions
df.hist(figsize=(12, 8))
plt.tight_layout()
plt.show()

• ARM-Metric datasets for activity recognition
• DM-Metric for environmental monitoring trends
• Football data for performance over time

• CPP_base_ampliado for energy efficiency prediction
• ENB2012_data for building heating load prediction

• ARM-Metric for activity type classification
• Football data for match outcome prediction

• DM-Metric-missing-3 for testing missing value strategies

• Interaction features in CPP_base_ampliado
• Temporal features in football and DM-Metric datasets

• Missing Values: DM-Metric-missing-3 contains intentional missing values for research purposes
• Data Types: Most datasets contain mixed numerical and categorical variables
• Scaling: Some features may require normalization (e.g., RSS values, energy measurements)
• Temporal Aspects: Football and DM-Metric datasets include time-based features

• ARM-Metric: Research on RSS-based activity recognition
• CPP_base_ampliado: Power plant efficiency optimization studies
• ENB2012: Building energy efficiency standards and research
• Football Data: Sports analytics and betting research

This repository contains datasets for research and educational purposes. Please check individual dataset licenses and cite original sources when using in publications.