Find research datasets worth reusing

MACHINE LEARNING APPROACHES FOR DEMAND FORECASTING: THE IMPACT OF CUSTOMER SATISFACTION ON PREDICTION ACCURACY

This study investigates the effectiveness of various machine learning models in predicting product demand based on customer satisfaction data. Four models—Linear Regression, Random Forest, Gradient Boosting, and Support Vector Machine (SVM)—were evaluated using performance metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R² score. The results indicate that Gradient Boosting achieved the highest accuracy, with an MAE of 2.56, MSE of 12.75, RMSE of 3.57, and R² score of 0.82, effectively capturing the complex, non-linear relationships inherent in customer satisfaction factors. Random Forest also demonstrated strong performance, while Linear Regression and SVM showed limitations in handling intricate datasets. These findings underscore the importance of utilizing advanced machine learning techniques for accurate demand forecasting, highlighting the critical role of customer satisfaction data in enhancing predictive capabilities. The insights gained from this research can guide organizations in optimizing inventory management and improving customer satisfaction in a rapidly evolving market.

Dataset of the manuscript: Predictive design to determine optimal absorber placement in colloidal photonic crystals

This data publication is based on the metadata and datasets underlying the manuscript "Predictive design to determine optimal absorber placement in colloidal photonic crystals". The Data is roughly organized by the figure of appearance. Figure 1 contained no result data Figure "Figure 2" contains: <ul> <li>Simulated and experimental reflectance spectra of bare PS colloidal crystal and CIELab color coordinates of simulated bare PS colloidal crystal.</li> </ul> Figure "Figure 3+4" contains: <ul> <li>Data for particle based and layer based designs for chroma optimization according to Eq. 2 </li> <ul> <li>convergence history J(steps)</li> <li>Optimized design absorber distributions (average of layers)</li> <li>CIELab color coordinates</li> <li>Spectra</li> </ul> <li>CIELab color coordinates and Chroma of all predictive designs sorted by threshold L value according to Eq. 3 + comparative designs: bottom absorber, top absorber, homogeneous.</li> </ul> Figure "Figure 5" contains: <ul> <li>Chemdraw File containing chemical structures</li> <li>Pendant drop surface tension measurements </li> <li>Surface pressure increase on Langmuir-Blodgett trough</li> </ul> Figure "Figure 6" contains: <ul> <li>SEM images of mono- and multilayers labeled in accordance to design and composition</li> </ul> Figure "Figure 7+8" contains: <ul> <li>Photographs of fabricated multilayers</li> <ul> <li>Homogeneous designs labeled in accordance to composition</li> <li>Layered designs labeled in accordance to design type (XBA: bottom absorber with X absorbing numbers; XTA: bottom absorber with X absorbing numbers; Ld_XX: predictive design with L threshold of XX)</li> </ul> <li>Spectra of all samples including their error determined from 2 measurements</li> <li>Average spectra of all designs (averaged from all samples of that design) including their error estimated using gaussian error propagation</li> <li>Color data of all samples calculated from spectra </li> <ul> <li>CIELab coordinates and Chroma</li> <li>xyz values</li> <li>RGB values</li> </ul> </ul> Figure "Figure 9" contains: <ul> <li>Optimized design absorber distributions, spectra and CIELab color coordinates and chroma for colloidal crystals of varying primary particle size</li> </ul>

Four spatial prediction datasets of susceptibility to gully erosion, comparing machine learning models, in the Piraí drainage basin, southeastern Brazil

The data in this repository refer to the article published in the journal Land, entitled: Machine Learning Models for the Spatial Prediction of Gully Erosion Susceptibility in the Piraí Drainage Basin, Paraíba do Sul Middle Valley, Southeast Brazil.

opencc-by-4.0Sep 2024View details →

Development and Comparison of Model-Based and Data-Driven Approaches for the Prediction of the Mechanical Properties of Lattice Structures

This dataset comes from the following paper: Chiara Pasini, Oscar Ramponi, Stefano Pandini, Luciana Sartore, Giulia Scalet, Development and Comparison of Model-Based and Data-Driven Approaches for the Prediction of the Mechanical Properties of Lattice Structures, J. of Materi Eng and Perform, 2024. <a href="https://doi.org/10.1007/s11665-024-10199-x">https://doi.org/10.1007/s11665-024-10199-x</a> It contains: <ul> <li>"Notes.pdf" describing all the files uploaded</li> <li>. m of the neural network</li> <li>. inp of the Abaqus finite element simulations</li> </ul>

The raw data for the research "Comparing Neural Network Models Based on Macro Perspective Economic and Environmental Indicators with ARIMA Model in predicting Construction Cost Index in UK"

The raw data for the research "Comparing Neural Network Models Based on Macro Perspective Economic and Environmental Indicators with ARIMA Model in predicting Construction Cost Index in UK". Data collector: Runda Zheng

opencc-by-4.0Sep 2024View details →

Online example database generated representing nuclear astrophysics models predictions of correlations between stable/stable abundances of specific isotopes.

Library of figures created using the SIMPLE code (Stellar Interpretation for Meteoritic data and PLotting). The SIMPLE stellar database includes 18 core-collapse supernova models with 3 different initial masses of 15, 20, 25 solar masses, all of solar metallicity and non-rotating stars. The 6 sets are the following: <ul> <li>Rauscher et al. 2002 [Ra02] (<a href="https://ui.adsabs.harvard.edu/abs/2002ApJ...576..323R/abstract">https://ui.adsabs.harvard.edu/abs/2002ApJ...576..323R/abstract</a>),</li> <li>Pignatari et al. 2016 [Pi16] (<a href="https://ui.adsabs.harvard.edu/abs/2016ApJS..225...24P/abstract">https://ui.adsabs.harvard.edu/abs/2016ApJS..225...24P/abstract</a>),</li> <li>Sieverdin et al. 2018 [Si18] (<a href="https://ui.adsabs.harvard.edu/abs/2018ApJ...865..143S/abstract">https://ui.adsabs.harvard.edu/abs/2018ApJ...865..143S/abstract</a>),</li> <li>Limongi & Chieffi 2018 [LC18] (<a href="https://ui.adsabs.harvard.edu/abs/2018ApJS..237...13L/abstract">https://ui.adsabs.harvard.edu/abs/2018ApJS..237...13L/abstract</a>),</li> <li>Ritter et al. 2018 [Ri18] (<a href="https://ui.adsabs.harvard.edu/abs/2018MNRAS.480..538R/abstract">https://ui.adsabs.harvard.edu/abs/2018MNRAS.480..538R/abstract</a>),</li> <li>Lawson et al. 2022 [La22] (<a href="https://ui.adsabs.harvard.edu/abs/2022MNRAS.511..886L/abstract">https://ui.adsabs.harvard.edu/abs/2022MNRAS.511..886L/abstract</a>)</li> </ul> The figures can be divided into two types. The first shows the structure of the ejecta and the abundance of the selected isotopes. The layers are automatically detected using SIMPLE based on the abundances of the main fuels (H-1, He-4, C-12, O-16, Ne-20, Si-28) from the supernova model ejecta. The code names the different layers based on the schematic diagram in Schofield et al. 2022 (<a href="https://ui.adsabs.harvard.edu/abs/2022MNRAS.517.1803S/abstract">https://ui.adsabs.harvard.edu/abs/2022MNRAS.517.1803S/abstract</a>).  Ni and Fe isotopes are plotted in the figures. The abundances shown include the radiogenic contribution from unstable isotopes. The SIMPLE code is designed to compare stellar data with measurements from meteorites. To achieve this, abundances in mass fractions need to be converted into isotopic ratios using specific units. See Lugaro et al. 2023 (<a href="https://ui.adsabs.harvard.edu/abs/2023EPJA...59...53L/abstract">https://ui.adsabs.harvard.edu/abs/2023EPJA...59...53L/abstract</a>) for details. For the specific case of Ni64 ratios, in comparison with model data we report the measured meteoritic anomaly by Steele et al 2012 (<a href="https://ui.adsabs.harvard.edu/abs/2012ApJ...758...59S/abstract">https://ui.adsabs.harvard.edu/abs/2012ApJ...758...59S/abstract</a>) as a continuous horizontal line. The same is done for the Fe54 ratios, with reference measurements by Hopp et al 2022 (<a href="https://ui.adsabs.harvard.edu/abs/2022E%26PSL.57717245H/abstract">https://ui.adsabs.harvard.edu/abs/2022E%26PSL.57717245H/abstract</a>).  In the database the abundance plots are identified as structure_<model reference>_<initial mass>_<element>_<decayed or undecayed>.png. In particular, the available reference model options are Ra02, Pi16, Si18, LC18, Ri18, La22;  the initial mass of the progenitors are 15, 20 or 25 (solar masses). The third part of the filenames are the plotted elements (in this case Ni or Fe) and then if they are decayed or undecayed. In this database we only consider the decayed species, which means that the radioactive isotopes whose decay can add to the abundance of the selected isotopes were considered. In this plots the x-axis represents the total mass from the core and the y-axis is the mass fraction on a logarithmic scale. For the slopes the same name scheme applies, but they are identified as slopes_<model reference>_<initial mass>_<element>_<decayed or undecayed>.png, and on the y-axis there are the slope values insteas of abundances.

openmit-licenseOct 2024View details →

Supporting data for RCANE: A Deep Learning Algorithm for Whole-genome Pan-Cancer Somatic Copy Number Aberration Prediction using RNA-seq Data.

This is the data repository for RCANE: A Deep Learning Algorithm for Whole-genome Pan-Cancer Somatic Copy Number Aberration Prediction using RNA-seq Data. To use this dataset, please refer to <a href="https://github.com/HowardGech/RCANE" target="_blank" rel="noopener">https://github.com/HowardGech/RCANE</a>.

TRANSFORMING CUSTOMER RETENTION IN FINTECH INDUSTRY THROUGH PREDICTIVE ANALYTICS AND MACHINE LEARNING

In recent years, the fintech industry has experienced rapid growth, driven by technological advancements and evolving consumer expectations. Fintech companies offer innovative financial services, such as digital banking, investment platforms, and payment solutions, catering to the needs of a tech-savvy customer base. However, as competition intensifies, customer retention has emerged as a critical challenge for these companies. According to a study by Ransom (2021), acquiring a new customer can cost five times more than retaining an existing one, making it imperative for fintech organizations to focus on strategies that enhance customer loyalty. The financial technology (fintech) sector has experienced unprecedented growth in recent years, fundamentally transforming how individuals and businesses access and manage financial services. Characterized by the integration of technology with financial services, fintech encompasses a wide array of offerings, including digital banking, peer-to-peer lending, robo-advisory services, and payment processing. As of 2023, the global fintech market was valued at approximately $309 billion and is projected to reach around $1.5 trillion by 2030, according to a report by Fortune Business Insights. This remarkable growth is largely attributed to advancements in digital technology, increasing smartphone penetration, and a growing consumer preference for online financial solutions. Moreover, the COVID-19 pandemic accelerated the adoption of digital financial services, as consumers sought contactless transactions and remote banking options.

opencc-by-4.0May 2024View details →

ECOCLIMAP-SG-ML: an ensemble land cover map for numerical weather prediction

This dataset contains ensemble land cover maps for numerical weather prediction at 60 m resolution over Europe. As they were generated thanks to machine learning, the weights and the training data are also provided.

opencc-by-4.0Nov 2024View details →

Transfer learning and DNA language models enhance transcription factor binding predictions

This is the dataset for replicating the results of the paper called "Transfer learning and DNA language models enhance transcription factor binding predictions" by Ekin Deniz Aksu and Martin Vingron. See https://github.com/ekinda/tfbs_prediction_paper

dryad40/100

Data from: The comparative biogeography of Philippine geckos challenges predictions from a paradigm of climate-driven vicariant diversification across an island archipelago

A primary goal of biogeography is to understand how large-scale environmental processes, like climate change, affect diversification. One often-invoked but seldom tested process is the so-called ''species-pump'' model, in which repeated bouts of co-speciation is driven by oscillating climate-induced habitat connectivity cycles. For example, over the past three million years, the landscape of the Philippine Islands has repeatedly coalesced and fragmented due to sea-level changes associated with the glacial cycles. This repeated climate-driven vicariance has been proposed as a model of speciation across evolutionary lineages codistributed throughout the islands. This model predicts speciation times that are temporally clustered around the times when interglacial rises in sea level fragmented the islands. Given the significance and conceptual impact the model has shown, surprisingly few tests of this prediction have been provided. We collected comparative genomic data from 16 pairs of insular gecko populations to test the prediction of temporally clustered divergences. Specifically, we analyze these data in a full-likelihood, Bayesian model-choice framework to test for shared divergence times among the pairs. Our results provide support against the species-pump model prediction in favor of an alternative interpretation, namely that each pair of gecko populations diverged independently. These results suggest the repeated bouts of climate-driven landscape fragmentation has not been an important mechanism of speciation for gekkonid lizards on the Philippine Islands. Interpretations of shared mechanisms of diversification historically have been pervasive in biogeography, often advanced on the basis of taxonomy-based depictions of species distributions. Our results call for possible re-evaluation of other, classic co-diversification studies in a variety of geographic systems.

opencc-zeroDec 2018View details →

opencc-by-4.0Jun 2020View details →

Metabolic pathway prediction using non-negative matrix factorization with improved precision

We include samples of various data types used in the work "Metabolic pathway prediction using non-negative matrix factorization with improved precision" More information about the software package and instructions are provided in <a href="https://github.com/hallamlab/triUMPF">hallamlab/triUMPF</a>

opencc-by-4.0Jun 2021View details →

Relevance of predictive and postdictive error information in the course of motor learning.

Dataset associated with the following publication: Maurer, LK, Joch, M, Hegele, M, Maurer, H, & Müller, H (2021). Relevance of predictive and postdictive error information in the course of motor learning. Neuroscience doi:10.1016/j.neuroscience.2021.05.007

openafl-3.0Jun 2021View details →

Protein language model embeddings and predictions of the human proteome

Residue and sequence embeddings of the human proteome (SwissProt for organism Human, downloaded on 2021.06.09) computed using bio_embeddings (bioembeddings.com) using the ProtT5 embedder at full precision (https://www.biorxiv.org/content/10.1101/2020.07.12.199554v3). Additionally: - Sequence-level predictions of subcellular localization in 10 classes using LA (https://www.biorxiv.org/content/10.1101/2021.04.25.441334v1) - Residue-level three state secondary structure prediction (alpha, sheet or other) using models reported in the ProtTrans paper (https://www.biorxiv.org/content/10.1101/2020.07.12.199554v3)   Files included: - human.fasta --> FASTA-formatted sequences of human from SwissProt - DSSP3_human_ProtT5Sec.fasta --> Secondary structure predictions in three states for each residue of each protein in human.fasta. "H" stands for Helix; "E" stands for Sheet; "C" stands for Other. - subcell_human_LA_ProtT5.csv --> Subcellular location (10 states) and memrane-boundness (2 states) for each protein in human.fasta - embeddings_file.h5 --> per-residue embeddings of sequences in human.fasta. Each dataset in the .h5 file represents a protein sequence and contains a matrix of length Lx1024, with L being the length of the protein sequence. Datasets are indexed using integers. The original sequence identifier (from the FASTA header) can be accessed through the "original_id" attribute. See https://docs.bioembeddings.com/v0.2.0/notebooks/open_embedding_file.html for information on how to open the file - reduced_embeddings_file.h5 --> per-sequence embeddings of sequences in human.fasta (obtained by mean-pooling the residue-embeddings along the length dimension of the protein sequence). Each dataset in the .h5 file represents a protein sequence and contains a vector of size 1024 (meaning, each sequence has the same dimension).

Data set: Can Stephen Curry really know? - Conscious access to outcome prediction of motor actions

Data set associated with the following pre-print: Can Stephen Curry really know? - Conscious access to outcome prediction of motor actions. bioRxiv: 2021.03.30.437477

opencc-by-4.0Jun 2018View details →

Regression Model to Predict the Higher Heating Value of Poultry Waste from Proximate Analysis.

The response variable is High Heating Values (HHV), while the independent variables are Fixed Carbon (FC), Volatile Matter (VM), and Ash (A).

opencc-by-4.0Jan 2021View details →

transynergy: synergistic prediction and pathway deconvolution of drug combinations

Dataset used for transynergy: synergistic prediction and pathway deconvolution of drug combinations

Pseudomonas aeruginosa predicted prophages from publicly available genomes

Through the Genome Information by Organism section of the NCBI Genome database, P. aeruginosa bacterial genomic assemblies were downloaded (September 2020). Genome quality was assessed totaling 5,383 genomes total. All 5,383 genomes were then entered into VirSorter v.1 (https://github.com/simroux/VirSorter). The data set provided here includes all category 1 and category 4 predicted prophage sequences.

R code to accompany 'A novel initialisation technique for decadal climate predictions'

R code for the diagnostic published in 'A novel initialisation technique for decadal climate predictions' <a href="https://doi.org/10.3389/fclim.2021.681127">https://doi.org/10.3389/fclim.2021.681127</a> The data loaded follow the cmor standard.

Data for publication of "Determining the sensitive parameters of WRF model for the prediction of tropical cyclones in the Bay of Bengal using Global Sensitivity Analysis and Machine Learning"

The data are made available as part of the paper "Determining the sensitive parameters of WRF model for the prediction of tropical cyclones in the Bay of Bengal using Global Sensitivity Analysis and Machine Learning", submitted to Geoscientific Model Development. This data set incorporates selected post-processed files needed to reproduce the results presented in the paper. The data contains six zip files, that are: <ul> <li>Namelist.input files for the WRF model simulations of ten tropical cyclones</li> <li>WRF model simulation outputs using the default parameter values</li> <li>WRF model simulation outputs using the optimal parameter values (which give minimum RMSE value)</li> <li>IMDAA surface observations and IMERG precipitation data</li> <li>IMD observed tracks of ten tropical cyclones</li> <li>Ipython notebooks of sensitivity analysis and machine learning codes</li> </ul> The remaining files are the ncl scripts that were used to obtain the figures. The ncl scripts used the data in the zip files.