I am working on an academic project on 'predicting fuel consumption using artificial intelligence (AI) in the automotive field' and I need a large dataset for training the model. I haven't found one yet.
For predicting fuel consumption using AI, you'll need a dataset that includes relevant variables such as engine size, vehicle weight, fuel type, driving patterns, and environmental conditions. Here are some potential sources for such datasets:
1. UCI Machine Learning Repository: This repository hosts a variety of datasets, including some related to vehicle performance and fuel consumption. For example, the "Automobile" dataset contains information on car characteristics and fuel efficiency.
2. EPA Fuel Economy Data: The Environmental Protection Agency (EPA) provides detailed fuel economy data for vehicles sold in the United States. This dataset includes attributes like vehicle make and model, engine specifications, and fuel economy ratings.
3. Kaggle: Kaggle hosts numerous datasets, and you can find specific ones related to fuel consumption and vehicle data. Examples include the "Vehicle Fuel Consumption" dataset, which contains information on different car models and their fuel consumption rates.
4. Open Data Portals: Various government and research institutions provide open access to transportation and vehicle data. For instance, the UK government's data portal might have relevant datasets on vehicle emissions and fuel consumption.
5. Canadian Vehicle Survey: This dataset includes detailed information about vehicle use in Canada, including fuel consumption, distance traveled, and vehicle characteristics.
### Specific Dataset Examples
1. Fuel Economy Dataset (EPA): Available [here](https://www.fueleconomy.gov/feg/download.shtml)
- This dataset includes variables like make, model, engine displacement, fuel type, combined MPG, city MPG, and highway MPG.
2. Automobile Dataset (UCI): Available [here](https://archive.ics.uci.edu/ml/datasets/automobile)
- This dataset includes variables like symboling, normalized losses, make, fuel type, aspiration, number of doors, body style, drive wheels, engine location, wheel base, length, width, height, curb weight, engine type, number of cylinders, engine size, fuel system, bore, stroke, compression ratio, horsepower, peak rpm, city mpg, highway mpg, and price.
3. Vehicle Fuel Consumption Dataset (Kaggle): Available [here](https://www.kaggle.com/datasets/anderas/car-consume)
- This dataset includes variables like model, fuel consumption in L/100 km, CO2 emissions in g/km, engine size, and more.
These datasets should provide a solid foundation for developing a predictive model for fuel consumption using AI techniques. If you need help with specific details on how to use these datasets, feel free to ask!
You can try using a synthetic data-set using ctgan or something similar
Government Databases: Many governments publish datasets related to vehicle fuel efficiency and emissions. For example, in the United States, the Environmental Protection Agency (EPA) maintains the Fuel Economy website (https://www.fueleconomy.gov/) which includes downloadable datasets on fuel economy ratings for different vehicles.
You can alternately request dataset from the researchers who have already worked in this field. Below are title of some such research papers
1. Prediction of Fuel Efficiency Using Machine Learning
2. Fuel Consumption Prediction Model using Machine Learning
3. Utilizing Artificial Intelligence to Identify An Optimal Machine Learning Model for Predicting Fuel Consumption in Diesel Engines
4. Use of Artificial Neural Networks to Predict Fuel Consumption on the Basis of Technical Parameters of Vehicles
5. FUEL EFFICIENCY PREDICTION
A suitable dataset for predicting fuel consumption is the "UCI Machine Learning Repository - Auto MPG Dataset." This dataset includes various car attributes such as horsepower, weight, and engine size, along with fuel consumption (miles per gallon). Another option is the "EPA Fuel Economy Data," which provides detailed fuel economy information for different vehicles. Both datasets can be used to train AI models for fuel consumption prediction.