Dataset for "Low-Cost, Multi-Sensor Non-Destructive Banana Ripeness Estimation Using Machine Learning"
Processed datasets containing all numerical sensor data used for training and testing the ML algorithms discussed in the associated publication. Data from temperature, pressure, humidity, VOC and spectral sensors is included. The data is split into four datasets (as defined in Table V of the associated publication), each containing a different combination of sensor data and each subdivided into data ("x") and labels ("y") for both testing and training data. 30% of the cleaned data is randomly taken to form the testing data, while the remaining 70% forms the training data. Each data subset is balanced, as discussed in section 3.E.3 in the associated publication.
Cite this dataset as:
Callaghan, K.,
Martinez Hernandez, U.,
2025.
Dataset for "Low-Cost, Multi-Sensor Non-Destructive Banana Ripeness Estimation Using Machine Learning".
Bath: University of Bath Research Data Archive.
Available from: https://doi.org/10.15125/BATH-01459.
Export
Data
ds_34.zip
application/zip (2MB)
Creative Commons: Attribution 4.0
"ds_34" dataset, containing data from all sensors for tests 3 & 4 only.
Spect.zip
application/zip (723kB)
Creative Commons: Attribution 4.0
"Spect" dataset, containing all spectral data from tests 2, 3, 5 only.
TPHiS.zip
application/zip (1MB)
Creative Commons: Attribution 4.0
"TPHiS" dataset, containing data from the internal environmental sensors (temperature, pressure, humidity) and all spectral data for tests 2, 3, 4, 5 only.
VOC.zip
application/zip (592kB)
Creative Commons: Attribution 4.0
"VOC" dataset, containing VOC data from all three VOC sensors (TGS2620, TGS2602, SGP40) for tests 3 & 4 only.
Creators
Kristian Callaghan
University of Bath
Uriel Martinez Hernandez
University of Bath; University of Sheffield
Contributors
University of Bath
Rights Holder
Coverage
Collection date(s):
From 11 September 2023 to 15 March 2024
Documentation
Data collection method:
The data collection methodology can be found in the associated publication.
Data processing and preparation activities:
The data preparation & processing methodology can be found in the associated publication.
Technical details and requirements:
The datasets were created with Python 3.10.13, with libraries Numpy 1.26.0 and Pandas 2.1.2. The data is saved in CSV format and does not require specialist software to read.
Additional information:
Data organisation and encoding is described in the associated ReadMe files.
Documentation Files
ds_34-ReadMe.txt
text/plain (5kB)
Creative Commons: Attribution 4.0
ReadMe documentation for "ds_34.zip" dataset files.
Spect-ReadMe.txt
text/plain (4kB)
Creative Commons: Attribution 4.0
ReadMe documentation for "Spect.zip" dataset files.
TPHiS-ReadMe.txt
text/plain (4kB)
Creative Commons: Attribution 4.0
ReadMe documentation for "TPHiS.zip" dataset files.
VOC-ReadMe.txt
text/plain (3kB)
Creative Commons: Attribution 4.0
ReadMe documentation for "VOC.zip" dataset files.
Funders
Engineering and Physical Sciences Research Council
https://doi.org/10.13039/501100000266
Manufacturing in Hospital: BioMed 4.0
EP/V051083/1
Publication details
Publication date: 16 January 2025
by: University of Bath
Version: 1
DOI: https://doi.org/10.15125/BATH-01459
URL for this record: https://researchdata.bath.ac.uk/id/eprint/1459
Related papers and books
Callaghan, K. M. S., and Martinez-Hernandez, U., 2025. Low-Cost, Multi-Sensor Non-Destructive Banana Ripeness Estimation Using Machine Learning. IEEE Sensors Journal, 1-1. Available from: https://doi.org/10.1109/jsen.2025.3528250.
Contact information
Please contact the Research Data Service in the first instance for all matters concerning this item.
Contact person: Uriel Martinez Hernandez
Faculty of Engineering & Design
Electronic & Electrical Engineering
