Find research datasets worth reusing

opencc-by-4.0Jan 2024View details →

Evaluating the Robustness of Deep-learning Algorithm-selection Models by Evolving Adversarial Instances - Code and Data

This repository contains the code and data for reproducibility of the paper 'Evaluating the Robustness of Deep-learning Algorithm-selection Models by Evolving Adversarial Instances'.  The following files are included: <ul> <li>Data.zip : contains the original instances in the datasets;</li> <li>Models.zip : trained Deep Neural Networks models used in the paper;</li> <li>New_instances.zip : generated instances using the approach;</li> <li>Parsed_data.zip : results and statistics of the experiments;</li> <li>script_adversarial_v3.py : Python script used to generate the results</li> </ul>

opencc-by-4.0Jan 2024View details →

Tyche Algorithm and Learning Elements Preferences

This dataset contains the result of the survey to learning styles and their learning element preferences. The survey was executed in summer term of 2023 in German universities.   The zip folder "implementation.zip" contains the code to the Tyche algorithm with some examples. The Python script runs with Python 3.10.5. Before executing, you need to install the packages of requirements.txt (pip install -r requirements.txt) For executing static and generic Tyche run "run_tyche.py"   Refering to Data_local_view.xlsx Learning element01[1] means probability for first learning element after the lecture for Learning Goal. Coding for the numbers in brackets is the following: <table> <tbody> <tr> <td>1</td> <td>Learning goal</td> </tr> <tr> <td>2</td> <td>Brief overview</td> </tr> <tr> <td>3</td> <td>Manuscript</td> </tr> <tr> <td>4</td> <td>Quiz</td> </tr> <tr> <td>5</td> <td>Exercise</td> </tr> <tr> <td>6</td> <td>Summary</td> </tr> <tr> <td>7</td> <td>Auditory additional material</td> </tr> <tr> <td>8</td> <td>Textual additional material</td> </tr> <tr> <td>9</td> <td>Visual additional material</td> </tr> </tbody> </table>   The presented work is supported by the ‘German Federal Ministry of Research, Technology and Space’ (BMFTR) through the granting of the funding project HASKI (FKZ: 16DHBKI035).

opencc-by-nc-4.0Mar 2024View details →

Screener and Enumerator with Force-Field Optimization (SEFFO): algorithm for searching adsorption sites and configurations on 2D materials

See ref:   (submission stage)

LAEA: A 2D LiDAR Assisted UAV Exploration Algorithm for Unknown Environment

Open the record for dataset details and reuse information.

Example solution of Waste collection routing for benchmarking routing algorithms

Set of ESRI shapefiles: <ul> <li>Cluster_Punti: set of points covered by each trip</li> <li>Cluster_Inviluppi: envelopes for each set of points, for visualization purposes e.g. to highlight intersection between the areas covered by different trips</li> <li>Trips_arc_id: sequence of arcs composing the trip</li> </ul> The folder also contains export_giri.xls which summarizes the trips and their KPIs. The solution is related with the instance shapefiles available at: https://doi.org/10.5281/zenodo.10875680

Example instance of Waste collection services for testing routing algorithms

Set of ESRI shapefiles: <ul> <li>bins\bin2: position of bins (that should be emptied e.g. twice a week)</li> <li>bins\PdR: grouping of bins per arc and direction</li> <li>depot\depot: depot for trucks (where the trucks have to start at the beginning of the shift and return at the end of the shift)</li> <li>plant\plant: waste treatment plant (where the trucks have to be emptied before returning to the depot)</li> </ul>

Impact of Training Instance Selection on Automated Algorithm Selection Models for Numerical Black-box Optimization -- Reproducibility Files

This repository contains the files to reproduce the results from the paper "Impact of Training Instance Selection on Automated Algorithm Selection Models for Numerical Black-box Optimization". Data Collection In this folder, all files used to generate the raw performance data for the set of algorithms are included, as well as the code used to generate the ELA features. For the performance, the packages 'ioh', 'nevergrad', 'modde' and 'modcma' are essential, while 'pflacco' is used for ELA. For all scripts in this folder, the number of parallel threads and the folders for reading function settings (included as 3 csv files in this folder) and storing data should be set before execution.  Note that the full performance data exceeds 50GB, so it is not included in this repository. Instead, the results of processing it (using the aocc_extraction script) are included in the 'auc_MABBOB' folder (spread across multiple csv-files, with a version using a different budget factor included as well). The ELA data is included as 'ELA' and 'ELA_BBOB' for the affine combinations and component functions respectively.   Data Processing, Analysis and Visualization The remaining reproducibility files can be found in the Reproducibility folder. Within this folder are several notebooks which handle various steps in the pipeline, starting with preprocessing the data collected in the previous steps. This results in some csv-files, which are also included for convenience. Afterwards, the remaining notebooks deal with correlation analysis, instance selection methods, and all included plots from the paper. To match the environment used during our execution of these scripts. a yml-file (to be used with conda or mamba) is available as well.

Finite element dataset and Artificial Neural Networks algorithms to predict the mechanical properties of innovative CLT

This folder includes the data collected from the finite element simulations of the innovative CLT to compute its mechanical properties, the error of the closed-form solutions predicting the bending stiffness in the minor direction D22, the variation of the distance between the Reissner Mindlin and Bending Gradient theory in terms of spacing between lateral lamellas, the hyperparameters tuning of several Artificial Neural Networks algorithms with or without prior knowledge, the ML evaluations, the saved artificial neural network algorithms to predict each mechanical property of innovative CLT, and the ML application to use it.

AD/AE of SC24 Submission: A sparsity-aware distributed-memory algorithm for sparse-sparse matrix multiplication

AD/AE for SC24 submission: A sparsity-aware distributed-memory algorithm for sparse-sparse matrix multiplication

Supporting datasets for the article Using rare event algorithms to understand the statistics and dynamics of extreme heatwave seasons in South Asia

<h3>Supporting datasets for the article Using rare event algorithms to understand the statistics and dynamics of extreme heatwave seasons in South Asia, submitted to Environmental Research: Climate</h3>   The dataset contains all intermediate data used for the article. The raw outputs of the model may be available upon reasonable request to clement.lpr@gmail.com The climate model Plasim can be downloaded, together with its documentation, from the webpage of  the « Theoretische Meteorologie » group at the University of Hamburg: <a href="https://www.mi.uni-hamburg.de/en/arbeitsgruppen/theoretische-meteorologie/modelle/plasim.html">https://www.mi.uni-hamburg.de/en/arbeitsgruppen/theoretische-meteorologie/modelle/plasim.html</a> The three jupyter notebooks <a href="../api/records/10888194/draft/files/Figures_article_archive.ipynb/content" target="_blank" rel="noopener noreferrer">Figures_article_archive.ipynb</a>, <a href="../api/records/10888194/draft/files/Zg500_maps_article_archive.ipynb/content" target="_blank" rel="noopener noreferrer">Zg500_maps_article_archive.ipynb</a>, <a href="../api/records/10888194/draft/files/Correlation_maps_3days_ERA5_Plasim_archive.ipynb/content" target="_blank" rel="noopener noreferrer">Correlation_maps_3days_ERA5_Plasim_archive.ipynb</a> contain the analysis and code used to produce the figures in the article. <a href="../api/records/10888194/draft/files/pyscripts.tar/content" target="_blank" rel="noopener noreferrer">pyscripts.tar</a> contains 4 python utilities, data_proceeding_module.py, plot_routines.py, subseasonal_stats_utilities.py, utilities_REA_analysis.py that are imported by the notebooks.

Machine learning algorithm reveals surface deoxygenation in the Agulhas Current due to warming

This file contains ML – random forest constructed oxygen in the Agulhas Current.

Data set for A Novel VNS-based Algorithm for SVC Allocation in the Brazilian Interconnected Power System

This release includes the 107-bus version of the Brazilian Interconnected Power System (available <a href="https://www.sistemas-teste.com.br/">here</a>). The system consists of 107 buses, 104 lines, and 67 transformers distributed across three areas: South, Southeast, and Mato Grosso. This test system provides extensive applications for problems related to steady-state analysis.

Algorithm Selection with Probing Trajectories: Benchmarking the Choice of Classifier Model - Data

This repository contains the data and additional information for the paper 'Algorithm Selection with Probing Trajectories: Benchmarking the Choice of Classifier Model'.  The following files are included: <ul> <li>accuracy.zip: raw performance files for all models;</li> <li>plots.zip: additional plots;</li> <li>tuning.zip: tuning log files.</li> </ul>

A Big Data Analysis Algorithm for Massive Sensor Medical Images

The smart sensor based big data analysis recommendation system has significant privacy and security concerns when it comes to using sensor medical images for suggestions and monitoring. The danger of security breaches and unauthorized access which might lead to identity theft and privacy violations increases when sending and storing sensitive medical data on the cloud. Insufficient or erroneous patient data can lead to poor treatment decisions, misdiagnoses, and unreliable recommendations. By creating an anomaly detection system based on machine learning specifically for medical image and providing timely treatments and notifications, our effort will improve patient care and well-being. We infer the feature extraction, feature selection, attack detection, and data collection data processing procedures in order to anticipate the anomaly in patient data. We transfer the data, take care of any missing values, and sanitize it using the data pre-processing mechanism. We employed the RFE and DPCA algorithms for feature selection and extraction, respectively. In addition, we applied the AGRNN approach to identify abnormalities. Data arrival rate, resource consumption, propagation delay, transaction epoch, true positive rate, false alarm rate, and RMSE are some of the metrics used to evaluate the proposed task.

Personalize E-Commerce Product Recommendations Based on User Behavior Using Reinforced Learning Algorithms

The development of a personalized and adaptive e-commerce product recommendation system will be developed using the Reinforcement Learning algorithm in this study. Initial data is extremely promising in its ability to raise sales conversion: 30% of the products added to the cart are never purchased. Additionally, there is a strong correlation of 0.8 between viewed versus purchased products. Data collection was from 447 Indonesian respondents over a period of June to July 2024. It was collected using an online questionnaire that measures recommendation quality, satisfaction, and ease of use with purposive sampling. Partial Least Squares Structural Equation Modeling was done on the data analysis. From that, it has been found that system quality is positively related to the accuracy, novelty, and diversity of the recommendation. The results further show how this would lead to an improved user experience, satisfaction, and sales conversion with the reinforcement learning-based system. These findings give insight into developing efficient adaptive recommendation systems on e-commerce platforms and open opportunities for further research.

Competitive Metaheuristic Algorithms for Building a Performance Database of a Dual-Band Combline Bandpass Filter with Microstrip Connection (Version 17)

To run the files, remove the prefixes (e.g. figA- or fig5A-)

Competitive Metaheuristic Algorithms for Building a Performance Database of a Dual-Band Combline Bandpass Filter with Microstrip Connection (Version 17) [Dataset]. Zenodo.

To run the files, remove the prefixes (e.g. figA- or fig5A-)

opencc-by-4.0Dec 2023View details →

Typhoon track tracking and forecasting algorithm based on multi-source information

Typhoon track tracking and forecasting algorithm based on multi-source information

openmit-licenseNov 2024View details →

A Preliminary Investigation on the Usage of Quantum Approximate Optimization Algorithms for Test Case Selection - Online Appendix

QAOA-TCS - Quantum Regression Test Case Selection   This repository contains all the necessary resources to reproduce the results of the QAOA-TCS method.   Dataset Files   The "datasets" folder contains:   - "sir_programs"   SIR Programs   The "datasets/sir_programs" folder contains, for each SIR program considered by this project, all the files needed to gather statement coverage, execution costs, and past fault coverage information.   For example, in the "flex" program folder:   - The file "fault-matrix.txt" contains rows representing flex's test cases. Each row has columns representing different versions of the program. Each cell (i,j) contains a binary value (0 or 1) indicating whether the i-th test case detects a fault in the j-th version. This configuration is called the fault matrix and provides historical fault coverage information.   - The folder "json_flex" contains a folder for each test case, with files like "flexi.gcov.json" to recover statement coverage and execution costs. These files detail which basic blocks were executed and how many times, enabling the calculation of total statement coverage and execution costs for each test case.   Source Code Files   DIVGA.m   The "MATLAB/DIVGA.m" file contains the pipeline for the DIVGA algorithm. For simplicity, the statement coverage, execution costs, and fault coverage data already gathered by "Notebook.ipynb" are written into text files, which DIVGA.m reads to bypass the actual datasets.   DIVGA.m must be reconfigured for each target program. Update parameters like M, N, and gamultiobj routine settings. Ensure H_size in line 104 is less than max{N, M} + 1. Update the denominator in line 53 based on the total number of code lines in the target program. Adjust the result reporting target files as well.   Notebook.ipynb   This file includes pipelines for dataset analysis, algorithm execution, and empirical comparisons.   It has two main sections:   1. QAOA-TCS vs SelectQA and Classical Algorithms      - Pipelines analyze SIR programs and compare QAOA-TCS, SelectQA, and classical algorithms. Manual configuration is needed when changing the target program, including updating file paths for Pareto fronts and configuring frontiers to build the reference.      - Statistical analysis requires populating the variables "algorithm_nondom_sirprogram" with the number of non-dominated solutions found during each of the 10 runs.   Results Files   The "results" folder contains the outcomes of QAOA-TCS, DIV-GA, Additional Greedy, and SelectQA after experiment execution. These files enable empirical evaluations and comparisons between the methods.

Sample data and algorithm implementation for ChiSCAT: Unsupervised Learning of Recurrent Cellular Micromotion Patterns from a Chaotic Speckle Pattern

Sample data and algorithm implementation for the article  <div>Trelin, A., Kussauer, S., Weinbrenner, P., Clasen, A., David, R., Rimmbach, C., & Reinhard, F. (2024). ChiSCAT: Unsupervised Learning of Recurrent Cellular Micromotion Patterns from a Chaotic Speckle Pattern. Nano Letters, 24(40), 12374-12381.</div>