Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/rooshmica/concrete-compressive-strength-prediction

Concrete Dataset for Regression Model training
https://github.com/rooshmica/concrete-compressive-strength-prediction

linear-regression machine-learning polynomial-regression python rmse-score scikit-learn

Last synced: 3 days ago
JSON representation

Concrete Dataset for Regression Model training

Awesome Lists containing this project

README 

        
#### Concrete-Compressive-Strength-Prediction



Concrete DateSet Path -> https://archive.ics.uci.edu/dataset/165/concrete+compressive+strength



### Project Overview

This project aims to predict the compressive strength of concrete using machine learning models. The dataset contains 1,030 observations with 9 attributes, where 8 are quantitative input variables (cement, water, fine aggregate, etc.) and 1 is the target variable (concrete compressive strength). The goal is to identify key factors that influence concrete strength and build accurate prediction models.



### Key Features

Attributes: The dataset includes cement, water, coarse aggregate, fine aggregate, age, fly ash, and other factors.

Target: Concrete compressive strength.



Machine Learning Models:

Multiple Linear Regression

Polynomial Regression



### Objective



The objective of the project is to:



Analyze the relationships between features and concrete compressive strength.

Build multiple machine learning models to predict the compressive strength.

Compare model performance and select the best model.



### Approach



Data Understanding: The dataset is first explored using correlation matrices and visualizations (e.g., scatter matrix).

Data Preparation: The dataset is split into training and testing sets for model building.

Modeling: Different machine learning models are applied and compared for their performance.

Evaluation: Model performance is evaluated using metrics such as R² and RMSE.



### Data Information

The dataset contains 1030 samples  



| idx | column                        | non-null count | dtype   |

|-----|-------------------------------|----------------|---------|

| 0   | Cement                        | 1030 non-null  | float64 |

| 1   | Blast Furnace Slag            | 1030 non-null  | float64 |

| 2   | Fly Ash                       | 1030 non-null  | float64 |

| 3   | Water                         | 1030 non-null  | float64 |

| 4   | Superplasticizer              | 1030 non-null  | float64 |

| 5   | Coarse Aggregate              | 1030 non-null  | float64 |

| 6   | Fine Aggregate                | 1030 non-null  | float64 |

| 7   | Age                           | 1030 non-null  | int64   |

| 8   | Concrete Compressive Strength | 1030 non-null  | float64 |

  



### Checking for missing values  



![DataSet Describe](images/Concrete_data.png)



In some components such as *Blast Furnace Slag*, *Fly Ash*, *Superplasticizer*, the min value (minimum value in that column) contains a value of 0.

Usually, the value 0 will be considered as *missing values*, however in this project **the value 0 will be assumed that the component was not used in the mixing process**



### Data Preprocessing Steps



Loaded the dataset and handled any inconsistencies (no missing values).

Performed exploratory data analysis (EDA) to understand feature relationships.

Split the data into training and testing sets.



## Correlation Matrix

The following correlation matrix illustrates the relationships between different features in the dataset:



![Correlation Matrix](images/Concrete_Correlation_Matrix.png)



## Scatter Matrix

Here is the scatter matrix showing the pairwise relationships between the features:



![Scatter Matrix](plots/scatter_plot_concrete.png)



### Models Used



Linear Regression: A basic model for capturing linear relationships.

Polynomial Regression: Enhances the model's ability to handle non-linear relationships.



### The models are evaluated based on:



R² (Coefficient of Determination): Measures how well the model explains the variance in the data.

RMSE (Root Mean Squared Error): Measures the average error in the model's predictions.

Key Findings:

Polynomial regression significantly improved model performance compared to linear regression (from 54% to 77% accuracy).



The evaluation metric used is the **root_mean_squared_error (RMSE)** loss function, implemented using the mean_squared_error loss function from sklearn and then taking the square root using numpy.sqrt(). The result is the RMSE loss function.



RMSE or Root Mean Squared Error is a loss function obtained from the process of squaring the error (y_true - y_prediction) and dividing by the count, then taking the square root.



### Dependencies



pandas

numpy

matplotlib

seaborn

scikit-learn



### Conclusion



In this project, we focused on predicting the compressive strength of concrete using machine learning techniques, specifically **Linear Regression** and **Polynomial Regression**. After thoroughly exploring the dataset and preprocessing the data, we built both models to assess their predictive capabilities.



**Key Findings:**

- **Linear Regression**: The linear regression model served as a baseline model. It provided a reasonable initial prediction but had limitations in capturing non-linear relationships within the data, resulting in a lower accuracy.

  

- **Polynomial Regression**: To improve the model’s ability to capture non-linear trends, we applied polynomial regression. This enhanced the model's performance significantly, increasing the accuracy from **54%** with linear regression to **77%**. The polynomial model was able to better capture the complexity of the relationship between the features and the target variable, thus reducing the model error and improving predictions.



### Overall Impact:

- Polynomial regression yielded a much better predictive performance than linear regression, demonstrating that some degree of non-linearity is present in the relationship between the input features and concrete compressive strength.

- The findings from this study can help in making more accurate predictions for concrete strength, which is crucial for civil engineering applications like construction and infrastructure development.



### Future Work:

- Further improvements could be made by experimenting with other machine learning models like Decision Trees or Random Forest, which might perform even better for this type of dataset.

- A deeper exploration into feature engineering and hyperparameter tuning could also be beneficial for optimizing model performance.



This study successfully demonstrated that a simple polynomial regression model can significantly improve predictive accuracy when dealing with complex relationships in concrete strength prediction.