Dark Mode
Concrete Strength Prediction Dataset
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset is designed to provide a starting point for individuals new to deep learning, serving as a benchmark for concrete strength prediction [1]. It is structured as a simple dataset for regression tasks using neural networks [1]. The primary goal is to not only achieve improved results but also to facilitate understanding of the deep learning process and learning throughout the journey [1].
Columns
The dataset includes nine columns, with eight features and one target variable:
- Cement: The amount of cement used. This column ranges from 102.00 to 540.00, with a mean of 281 and a standard deviation of 104 [2, 3].
- Blast Furnace Slag: The amount of slag produced in a blast furnace. Values range from 0.00 to 359.40, with a mean of 73.9 and a standard deviation of 86.2 [3].
- Fly Ash: The quantity of ash produced. This ranges from 0.00 to 200.10, with a mean of 54.2 and a standard deviation of 64 [4].
- Water: The amount of water required. This column has values from 121.80 to 247.00, with a mean of 182 and a standard deviation of 21.3 [4, 5].
- Super-plasticizer: Describes the rigidity of cement after drying. Values are between 0.00 and 32.20, with a mean of 6.2 and a standard deviation of 5.97 [5, 6].
- Coarse Aggregate: Represents the coarse nature of the cement particles. This ranges from 801.00 to 1145.00, with a mean of 973 and a standard deviation of 77.7 [6].
- Fine Aggregate: Describes the fineness of the cement. Values are from 594.00 to 992.60, with a mean of 774 and a standard deviation of 80.1 [7].
- Age: The age or time before the concrete requires repairing, measured in days. This ranges from 1.00 to 365.00, with a mean of 45.7 and a standard deviation of 63.1 [7, 8].
- Strength: The target variable, representing the strength of concrete in kilonewtons (/kN) per kilonewton. This ranges from 2.33 to 82.60, with a mean of 35.8 and a standard deviation of 16.7 [8].
Distribution
The dataset is provided in a CSV format, specifically
concrete_data.csv, with a file size of 58.99 kB [2]. It contains 9 columns and consists of 1030 valid records, with no mismatched or missing values detected across any of the columns [2-8].Usage
This dataset is ideal for:
- Regression analysis: Particularly for predicting concrete strength based on its constituent materials and age [1].
- Deep learning studies: Serving as a foundational dataset for those beginning their journey in deep learning [1].
- Neural network model development: Suitable for training and benchmarking neural network models [1].
- Educational purposes: Excellent for demonstrating the process of data preparation and model building in machine learning courses [1].
Coverage
The sources do not provide specific geographic, time range, or demographic scope for this dataset. It focuses purely on the chemical and physical properties of concrete mixtures and their resulting strength.
License
CC0: Public Domain
Who Can Use It
This dataset is particularly useful for:
- Beginners in deep learning: Those who are just starting out and need a straightforward dataset to practise with [1].
- Data scientists: For developing and testing regression models, especially with neural networks [1].
- Machine learning engineers: To benchmark the performance of different deep learning architectures for predictive tasks [1].
- Researchers and students: Engaging in studies related to materials science, civil engineering, or applied machine learning [1].
Dataset Name Suggestions
- Concrete Strength Prediction Dataset
- Concrete Mix Regression Data
- Deep Learning Concrete Properties
- Neural Network Concrete Strength
Attributes
Original Data Source: Concrete Strength Prediction Dataset
QUALITY