Find research datasets worth reusing

opencc-by-4.0Nov 2018View details →

Integrated Health Monitoring and Reinforcement of Transportation Structures with Optimized Low-Cost Multifunctional Braided Cables

<p>Corresponding data set for Tran-SET Project No. 17STTAM04. Abstract of the final report is stated below for reference:</p> <p>"The objective of this research study is to design, fabricate, and characterize multifunctional high strength and self-sensing braided cables and structures using novel Fe-based shape memory alloys (SMAs). The system exploits unique properties of recently developed low-cost super-elastic FeMnAlNi SMAs, which enables excellent super-elastic properties, high strength, and self-sensing in structural health monitoring (SHM) systems. This novel material technology can be coupled with modeling efforts that allow for accurate prediction of both the materials and structural response during sensing. At the conclusion of the project, we have demonstrated that with careful design of processing parameters, it is possible to control the yield strength and superleastic properties of FeMnAlNi SMAs. We have fabricated, for the first time in the world, a large diameter wires and bars from these inexpensive iron based SMAs which are expected to help with scaling up the fabrication of these materials. In addition, we were able to develop a method for fabricating braided cables from the fabricated Fe-SMA wires and a lab scale experimental setup has been designed and built. Prototype design for the braiding weave was created and tested. Finally, an experimental setup has been designed and manufactured to measure changes in the magnetic response of the SMA braided cables under load in order to directly correlate the magnetic response with deflection and strain. Clearly, the successful fabrication of wires and braided cables of the inexpensive iron based SMAs with superelastic strains comparable to nickel-titanium SMAs and with favorable magnetic sensing capabilities have significant implications for transportation infrastructure as these materials can provide structural health monitoring capability while also mitigating large shape changes during natural disasters. Future work needs to focus on revealing the coupling between mechanical properties (in particular damage) and the changes in magnetic properties in order to provide guidance on how these materials can be utilized in structural health monitoring."</p>

opencc-by-4.0Oct 2019View details →

Reasearch data for publication "On optimizing metaheuristics performance with different error normalization approaches"

<p>These are the results of experiments performed for publication "On optimizing metaheuristics performance with different error normalization approaches" using authors own software -- EvANN -- which is available on github.</p>

opencc-by-4.0Nov 2019View details →

ROS-Specific Huntingtin Interactions: PAR overlay slot blotting optimization

<p>Optimization to find conditions allowing quantification of huntingtin PAR binding in the PAR overlay assay.</p>

Data supporting the publication "Optimized Sandwich and Topological Structures for Enhanced Haptic Transparency"

<p>This dataset includes experimental modal responses of sandwich and topological structures, which are designed to enhance haptic transparency. </p>

ML-Optimized QKD Frequency Assignment for Efficient Quantum-Classical Coexistence in Multi-Band EONs

<p>Abstract: Quantum key distribution (QKD) represents a cutting-edge technology that ensures unbreakable security. Coexisting quantum and classical signals on a multi-band (O+E+S+C+L-band) system offer a viable solution for secure, high-rate networks amidst growing classical traffic and address quantum signal sensitivity. In this study, we assume a dynamic classical traffic load and varying configurations of classical channels (CChs). Considering the varying behavior of Secure Key Rate (SKR) under different classical conditions, solving the integral noise equations are crucial for optimizing QKD implementation and enhancing resource efficiency. The complexity and time-consuming nature of this process challenge infrastructure providers in determining the optimal quantum channel (QCh) frequency in real time. To tackle these challenges, we propose a machine learning (ML) algorithm. By leveraging ML, QKD can be implemented efficiently, optimizing resource utilization while significantly reducing computation and processing time in dynamic classical traffic. We implement three ML algorithms at various fiber intervals, all of which estimate the optimal frequency for QCh with 99\% accuracy and perform computations on average in 0.09 seconds, which is significantly faster compared to integral computational methods that have a mean time of 637 seconds.<br><br>Information: In this file, the Excel sheet contains data for each fiber interval, including inputs such as fiber length in each interval, the overall classical loading factor percentage, the C-band loading factor percentage, the L-band loading factor percentage, the highest active classical frequency (which serves as input to the machine learning model), and the QCh frequency that resulted in the highest SKR.</p>

H2020 ENODISE: VKI Experimental dataset of configuration C for numerical optimization

<p>The database includes acoustic and load measurements performed in the ALCOVES anechoic facility of the von Karman Institute on co-rotating coaxial propellers. This configuration corresponds to Config. C of the ENODISE project. This database has been used for numerical optimization.  </p>

opencc-by-4.0Jul 2024View details →

Optimization and scalability assessment of supercapacitor electrodes based on hydrothermally grown MoS2 on carbon cloth

<p>This dataset contains the measurement data for figures (graphs) published in journal article:</p> <p>Optimization and scalability assessment of supercapacitor electrodes based on hydrothermally grown MoS2 on carbon cloth </p> <p>by Jasna Mannayil, Olli Pitkänen, Minna Mannerkorpi, and Krisztian Kordas</p>

opencc-by-4.0Apr 2024View details →

In vivo optimization of the experimental conditions for the non-invasive optical assessment of breast density

<p>We applied time-domain diffuse optical spectroscopy over a broad spectral range (600-1100 nm) to estimate the breast composition in terms of water, lipids, collagen, oxy- and deoxy-hemoglobin concentrations, together with scattering parameters (scattering amplitude <em>a</em> and scattering power <em>b</em>). These optical parameters are correlated with the density of the breast, which is an important risk factor involved in the development of breast cancer. We performed <em>in vivo </em>measurement on 11 healthy volunteers, using a measurement protocol that involves reflectance and transmittance geometries, different positions of the subject and different measurement locations on the breast. This work has been pubblished in Scientific Reports: https://doi.org/10.1038/s41598-024-70099-x .</p> <p>This page contains the dataset generated during this work, togheter with some tools to read it and the analysis that we performed.</p>

QM9-OR: DFT Optimized Geometries and Optical Rotations for Selected QM9 Molecules

<p>The QM9-OR data set consists of 121,416 molecules extracted from the original QM9 data set. For each molecule, the coordinate geometry is optimized using density functional theory, and the calculated specific rotations at three wavelengths using CAM-B3LYP/6-31G** are provided. The number of chiral centers and the absolute configurations are also provided for each molecule. Data set is a binary file in NumPy (npy) format. Each entry contains the following: index number, InChI identifier, Cartesian coordiantes, one-hot encoded atom type (H, C, N, O, F), chiral centers, and calculated optical rotations at 355 nm, 589.3 nm, and 633 nm.</p>

Online Appendix of "Toward Interactive Optimization of Source Code Differences: An Empirical Study of Its Performance"

<div> <div><strong>Abstract</strong></div> <div>This is the dataset of the paper entitled "Toward Interactive Optimization of Source Code Differences: An Empirical Study of Its Performance", presented at SCAM 2024.  It contains information related to the target commits and their attributes, as well as simulation results pertaining to the research questions in the paper.</div> <div> </div> <div>The target projects and commits in the dataset are based on a prior study: Nugroho, et al.: "How different are different diff algorithms in Git?: Use --histogram for code changes", Empirical Software Engineering, 2020, https://doi.org/10.1007/s10664-019-09772-z</div> <div> </div> </div> <div> <div><strong>Survey Overview</strong></div> <div> <div><em>1. Filtration</em></div> <div> <div>We collected attributes related to the changes for filtering and RQ purposes.</div> <ul> <li>Number of lines</li> <li>Number of changed lines</li> <li>Similarity distance</li> <li>Number of mismatch diff area</li> </ul> <div><em>2. RQ1</em></div> <div> <div>We investigate the minimum number of feedback actions needed to correct the initial diffs to the target diffs. Regarding the simulation, two types of heuristic functions (non-admissible and admissible functions) have been used to reduce costs. When the search with the non-admissible heuristic function of the initial state does not match the ideal optimal result, i.e., when there is room for improvement in the number of feedback actions, we applied another A* search with the admissible heuristic function.</div> <div> </div> <div> <div>We obtained the following results through search:</div> <ul> <li>Number of feedback actions (A* search with non-admissible heuristic)</li> <li>Number of feedback actions (A* search with admissible heuristic)</li> </ul> <div> </div> <div><em>3. RQ2</em></div> <div> <div> <div>We investigated the various effects that feedbacks have on the diffs by examining the diffs at depth 1 of the search tree. The dataset records the maximum, minimum, median, mean, and standard deviation for each search problem.</div> <br> <div>The study yielded the following results:</div> <ul> <li>Similarity distance</li> <li>Number of mismatch diff area</li> </ul> </div> </div> </div> </div> </div> </div> <div> <div> </div> <div> <div><strong>Dataset Columns</strong></div> <div>The following are the contents represented by the columns in the CSV file and their descriptions.</div> <div> </div> <div> <table> <tbody> <tr> <td>Column name</td> <td>Description</td> </tr> <tr> <td>project_name</td> <td>Name of the project associated with the data.</td> </tr> <tr> <td>filename</td> <td>Name of the file being analyzed.</td> </tr> <tr> <td>filepath</td> <td>Path to the file within the project.</td> </tr> <tr> <td>commit_id</td> <td>Commit hash representing the new version of the file.</td> </tr> <tr> <td>parent_commit</td> <td>Commit hash representing the old version of the file.</td> </tr> <tr> <td>error_commit</td> <td>An error occurred when retrieving the commit from the repository.</td> </tr> <tr> <td>error_setup</td> <td>Any error when generating the new and old versions of the file.</td> </tr> <tr> <td>error_analyze</td> <td>An error when collecting information for filtering.</td> </tr> <tr> <td>new_loc</td> <td>Lines of the new version of the source code.</td> </tr> <tr> <td>old_loc</td> <td>Lines of the old version of the source code.</td> </tr> <tr> <td>histogram_len</td> <td>Path length of the diff when using the Histogram algorithm.</td> </tr> <tr> <td>myers_len</td> <td>Path length of the diff when using the Myers algorithm.</td> </tr> <tr> <td>histogram-myers#edge</td> <td>Number of difference edges between Histogram and Myers diff.</td> </tr> <tr> <td>histogram-dp#edge</td> <td>Number of difference edges between Histogram and initial diff.</td> </tr> <tr> <td>myers-dp#edge</td> <td>Number of difference edges between Myers and initial diff.</td> </tr> <tr> <td>histogram-myers#area</td> <td>Number of mismatch diff areas between Histogram and Myers diff.</td> </tr> <tr> <td>histogram-dp#area</td> <td>Number of mismatch diff areas between Histogram and initial diff.</td> </tr> <tr> <td>myers-dp#area</td> <td>Number of mismatch diff areas between Myers and initial diff.</td> </tr> <tr> <td>dp#candidate</td> <td>Number of feedback candidates of initial diff (similarity distance).</td> </tr> <tr> <td>#insert</td> <td>Number of lines added in the change.</td> </tr> <tr> <td>#delete</td> <td>Number of lines deleted in the change.</td> </tr> <tr> <td>#change</td> <td>#insert + #delete.</td> </tr> <tr> <td>error_Asearch</td> <td>An error occurring during A* search with a non-admissible heuristic.</td> </tr> <tr> <td>#feedback_A</td> <td>Number of feedback actions for A* search with a non-admissible heuristic.</td> </tr> <tr> <td>time_A</td> <td>Time taken for A* search with a non-admissible heuristic (ms).</td> </tr> <tr> <td>RQ1_error_iteration</td> <td>An error when iterations exceed the limit (10,000,000) during A* search with an admissible heuristic.</td> </tr> <tr> <td>RQ1_error_timeout</td> <td>An error when the search time exceeds the limit (1,800 seconds).</td> </tr> <tr> <td>RQ1_error_other</td> <td>Other errors encountered during A* search with an admissible heuristic.</td> </tr> <tr> <td>RQ1#feedback</td> <td>Number of feedback actions for A* search with an admissible heuristic.</td> </tr> <tr> <td>RQ1#iter</td> <td>Number of iterations for A* search with an admissible heuristic.</td> </tr> <tr> <td>RQ1_time</td> <td>Time taken for A* search with an admissible heuristic.</td> </tr> <tr> <td>RQ2_error_exceed</td> <td>An error due to exceeding time or iteration limits in RQ2.</td> </tr> <tr> <td>RQ2_error_other</td> <td>Other errors encountered during RQ2.</td> </tr> <tr> <td>RQ2#children</td> <td>Number of children nodes of the initial state (= similarity distance).</td> </tr> <tr> <td>RQ2#candidate_min</td> <td>Minimum similarity distance among the generated diffs.</td> </tr> <tr> <td>RQ2#candidate_max</td> <td>Maximum similarity distance among the generated diffs.</td> </tr> <tr> <td>RQ2#candidate_ave</td> <td>Average of similarity distance among the generated diffs.</td> </tr> <tr> <td>RQ2#candidate_median</td> <td>Median of similarity distance among the generated diffs.</td> </tr> <tr> <td>RQ2#candidate_sd</td> <td>Standard deviation of similarity distance among the generated diffs.</td> </tr> <tr> <td>RQ2#area_min</td> <td>Minimum number of mismatch diff areas among the generated diffs.</td> </tr> <tr> <td>RQ2#area_max</td> <td>Maximum number of mismatch diff areas among the generated diffs.</td> </tr> <tr> <td>RQ2#area_ave</td> <td>Average number of mismatch diff areas among the generated diffs.</td> </tr> <tr> <td>RQ2#area_median</td> <td>Median number of mismatch diff areas among the generated diffs.</td> </tr> <tr> <td>RQ2#area_sd</td> <td>Standard deviation of number of mismatch diff areas among the generated diffs.</td> </tr> <tr> <td>is_used</td> <td>Indicates whether this data is used in the results of RQ1 and RQ2.</td> </tr> </tbody> </table> </div> </div> </div> </div>

Support data for article "The concept of optimal planning of a linearly oriented segment of the 5G network"

<p>Support data for article</p> <p>V. Kovtun, K. Grochla, E. Zaitseva, and V. Levashenko, “The concept of optimal planning of a linearly oriented segment of the 5G network,” PLOS ONE, vol. 19, no. 4. Public Library of Science (PLoS), p. e0299000, Apr. 17, 2024. doi: 10.1371/journal.pone.0299000.</p> <p>This research is part of the project No. 2022/45/P/ST7/03450 co-funded by the National Science Centre and the European Union Framework Programme for Research and Innovation Horizon 2020 under the Marie Skłodowska-Curie grant agreement No. 945339.</p>

Experimental result data for primal-dual contextual Bayesian optimization

<p>This dataset is the result of the paper "Primal-Dual Contextual Bayesian Optimization for Control System Online Optimization with Time-Average Constraints" published in IEEE Conference on Decision and Control 2023. </p>

opencc-by-4.0Feb 2024View details →

ESA 4DMED-SEA - Mediterranean Multivariate Optimal Interpolated Salinity and Density fields

<p>Daily Mediterranean gap-free Level-4 (L4) analyses of the Sea Surface Salinity (SSS) and Sea Surface Density (SSD) at 1/24° of resolution from 2016 to 2022, obtained through a multivariate optimal interpolation algorithm that combines sea surface salinity images from multiple satellite sources as NASA’s Soil Moisture Active Passive (SMAP) and ESA’s Soil Moisture Ocean Salinity (SMOS) satellites with in situ salinity measurements and satellite UHR SST information. </p>

opencc-by-4.0Dec 2023View details →

Optimizing Encephalomyocarditis virus VP1 protein assembly on Pseudorabies Virus envelope via US9 protein anchoring

<p>Live herpesvirus-vectored vaccines play a critical role in veterinary medicine, yet they may sometimes provide insufficient protection due to suboptimal antigen expression or localization. Encephalomyocarditis virus (EMCV) presents a notable zoonotic threat, with VP1 protein as a key immunogen on its capsid. To enhance immunogenicity, we investigated recombinant pseudorabies virus (rPRV) as a vaccine vector against EMCV. Through in silico analysis, we discovered that fusing VP1 with US9 enhances the formation of a type II transmembrane heterodimer. We constructed six rPRV groups expressing VP1 with various forms, revealing that VP1 fused with US9's C-terminal (US9-VP1) enhances VP1's membrane localization and its incorporation into the PRV envelope, unlike wild-type VP1. Immunogold electron microscopy illustrated that rPRV with deleted US8 and US9, supplemented with US8 regulatory sequence (rΔ89-U9VP1), enhanced the incorporation of VP1 into the viral envelope. Post-immunization, only rΔ89-U9VP1 conferred 100% protection against EMCV in mice and induced high levels of virus-neutralizing antibodies in piglets. Furthermore, rPRV expressing VP1 stimulated robust T-cell responses, as demonstrated by flow cytometry and ELISpot. This study marks the debut of rPRV's potential as an EMCV vaccine, underscoring how the selection of the US9 C-terminal domain and US8 regulatory sequence significantly amplifies the presentation of heterologous antigens, enhancing vaccine efficacy.</p>

Process Integration Multi-Actor Multi-Criteria Optimization (PI-MAMCO) Framework: Case Study on Palm Oil Based Complex (POBC)

<p>The dataset includes the complete workflow for the PI-MAMCO method implementation, demonstrated on a POBC integration case study. The files are structured as follows:</p> <ol> <li> <p><strong>NOTEBOOKS Folder</strong>: This folder contains the Jupyter notebooks that constitute the entire workflow of our case study. These notebooks import necessary model equations, data, and functions from the "MODEL" folder. Running these notebooks with the correct Python package versions will reproduce the exact results presented in our submitted article.</p> </li> <li> <p><strong>MODEL Folder</strong>: This folder includes all model equations, data, and functions required by the Jupyter notebooks. The scripts in this folder are integral to executing the PI-MAMCO method as outlined in our research.</p> </li> <li> <p><strong>OUTPUTS Folder</strong>: The results generated by running the notebooks are saved in this folder. This includes all outputs reported in our manuscript as well as additional results not included in the paper.</p> </li> <li><strong>OUTPUTS_GENERATED Folder</strong>: The files contained in this folder are the expected results from executing the case study notebook.</li> </ol> <p><em>Reproducibility</em>: To ensure the reproducibility of our results, please ensure that you are using the correct versions of the required Python packages. Detailed instructions for setting up the Python environment are provided within the notebook and the readme.txt.</p> <p>Contained are also outputs from the <a href="https://www.sciencedirect.com/science/article/pii/S0957417422003931" target="_blank" rel="noopener">IRenE procedure</a>, used for the literature review (<em>01_extracted_keywords.xlsx</em> ; <em>02_sampling_results.csv</em>)</p>

CAABA/MECCA model output for evaluating optimized step size control in Rosenbrock solvers for stiff ODEs

<p>This dataset comprehend the simulation output obtained with the CAABA/MECCA model for the optimization of the step size control in the Rosenbrock integrators available from the Kinetic PreProcessor (KPP) version 2.2.3_rs4. This dataset is made avalaible for the peer-review of the manuscript describing the model and the different options of step size control that was implemented. </p>

Optimal (v, 4, 1) binary cyclically permutable constant weight codes ((v,4,1) optical orthogonal codes (OOC)) and cyclic S(2, 4, v) designs

<p><strong>Optimal (v, 4, 1) binary cyclically permutable constant weight codes ((v,4,1) optical orthogonal codes (OOC)) and cyclic S(2, 4, v) designs.</strong></p> <p><strong>All (v,4,1) OOC</strong> for 26≤v≤76 and v=85: All_oc_26-85_4_1.zip</p> <p><strong>All cyclic 2-(v,4,1) designs</strong> for v=40, 52, 64, 76, 88: d_4_1.zip.</p> <p><strong>Some (v,4,1) OOC: </strong></p> <p>The codes are constructed by backtrack search considering all possibilities for the first two codewords (n=0 and n=1), but only the first 3 possibilities for the n-th codeword (n > 1) are considered for 77≤v≤98: E3oc_77-98_4_1.zip.</p> <p>The codes are constructed by backtrack search considering all possibilities for the first two codewords (n=0 and n=1), but only the first 2 possibilities for the n-th codeword (n > 1) are considered for 99≤v≤136: E2oc_99-136_4_1.zip</p> <p><strong>Some cyclic 2-(v,4,1) designs. </strong>The designs are constructed by backtrack search considering all possibilities for the first two or three baseblocks, but only the first 2 possibilities for the n-th baseblock (n > 1 for strictly cyclic designs, n>2 for designs with one short orbit) are considered for 100≤v≤160: E2d_4_1.zip</p>

Coarse-Grained and Multi-Dimensional Data-Driven Molecular Generation: A Structure-Based Framework for Selective Inhibitor Design and Optimization

<p><span>Many approaches not only fail to consider the intricate binding pocket interactions, leading to molecules with suboptimal properties and stability, but also struggle with designing selective inhibitors. To address this challenge, we have developed an innovative structure-based three-dimensional molecular generation framework named </span><span>Coarse-grained and Multi-dimensional Data-driven molecular generation (CMD-GEN). This framework bridges three-dimensional ligand-protein complex data with two-dimensional drug-like molecule data by utilizing coarse-grained pharmacophore points sampled from diffusion models, thereby enriching the training data for generative models.</span> <span>Through a hierarchical architecture, it decomposes the generation of three-dimensional molecules within the pocket into sampling of coarse-grained pharmacophore points, generating of chemical structures, and alignment of conformations, avoiding the instability issues associated with inherent in deep generative model-based generation of molecular conformations.<br><br>This project provide the source dataset used to train and evaluate the overall model.<br></span></p>

opencc-by-4.0Jul 2021View details →

Efficient Enumeration of the Optimal Solutions to the Correlation Clustering problem

<div><strong>Description. </strong>This is the data used in the experiments presented in the following paper:<br> <div> <ul> <li> <div> <div>N. Arınık, R. Figueiredo, and V. Labatut, “Efficient Enumeration of the Optimal Solutions to the Correlation Clustering problem,” <em>Journal of Global Optimization</em>, vol. 86, pp. 355–391, 2023. DOI: <a href="http://doi.org/10.1007/s10898-023-01270-3">10.1007/s10898-023-01270-3</a> ⟨<a href="https://hal.science/hal-03935831">hal-03935831</a>⟩</div> </div> </li> </ul> <p><strong>Source code. </strong>The related source code is available on GitHub: </p> <ul> <li><a href="https://figshare.com/articles/dataset/Efficient_Enumeration_of_Correlation_Clustering_Optimal_Solution_Space/%3Ci%3Ehttps://github.com/%3C/i%3E%3Ci%3E%3Ci%3ECompNet%3C/i%3E/Sosocc%3C/i%3E">https://github.com/CompNet/Sosocc</a></li> <li><a href="https://figshare.com/articles/dataset/Efficient_Enumeration_of_Correlation_Clustering_Optimal_Solution_Space/%3Ci%3Ehttps://github.com/CompNet/EnumCC%3C/i%3E">https://github.com/CompNet/EnumCC</a></li> </ul> <div><strong>Citation. </strong>If you use these data, please cite the above reference:</div> <div><br><code>@Article{Arinik2021,</code><br><code>  author    = {Arınık, Nejat and Figueiredo, Rosa and Labatut, Vincent},</code><br><code>  title     = {Efficient Enumeration of the Optimal Solutions to the Correlation Clustering problem},</code><br><code>  journal   = {Journal of Global Optimization},</code><br><code>  year      = {2023},</code><br><code>  volume    = {86},</code><br><code>  pages     = {355-391},</code><br><code>  doi       = {10.1007/s10898-023-01270-3},</code><br><code>}</code></div> <div> </div> <div><strong>Funding. </strong>This research benefited from the support of Agorantic FR 3621, as well as the FMJH Program PGMO and from the support to this program from EDF-THALES-ORANGE-CRITEO.</div> <div> </div> <div><strong>Further details. </strong>We describe below the structure of the zip file `article_materials.zip`:</div> <div> <ul> <li>Experiments for Dataset 1 <ul> <li>delay_exec_time: all the results and plots regarding the difference of execution times between EnumCC(3) and OneTreeCC() (i.e., EnumCC(3) minus OneTreeCC()), represented on the log-scaled y-axis of the plots. When such difference takes a negative value, this means our proposed method EnumCC(3) runs faster than OneTreeCC().</li> <li>EnumCC_nb-jumps: all the results regarding the number of jumps related to EnumCC(3), i.e. n<sub><em>jump</em></sub>(EnumCC(3))</li> <li>exec_time: all the results regarding the execution times of EnumCC(3) and OneTreeCC().</li> <li>nb-sols: all the results regarding the number of optimals solutions based on EnumCC(3). Note that we show the results of OneTreeCC() only for those with n=50, since both methods run out of the time limit of 12h for several networks with n=50.</li> </ul> </li> <ul> <li>networks: We generate these complete and incomplete networks through our random signed network generator, which is publicly available online. For complete unweighted signed networks, this model relies on only three parameters: n (number of vertices), l<sub>0</sub> (initial number of modules) and <em>q<sub>m</sub></em> (proportion of misplaced edges, i.e. edges meant to be frustrated by construction). Moreover, we make the assumption that the proportion of misplaced edges is the same inside and between the modules. When it comes to incomplete unweighted signed networks, we introduce two more parameters, which are the density d of the graph and the proportion q neg of the negative edges. The last parameter q<sub>neg</sub> allows to control the ratio of positive to negative edges. For complete unweighted signed networks with d = 1, we generate 20 replications for parameter values l<sub>0</sub> = 3, n <em>∈</em> {32, 36, 40, 45, 50} and <em>q<sub>m</sub></em> <em>∈</em> {0.1, 0.2, 0.3, 0.4, 0.5, 0.6}. In these networks, the value of <em>q<sub>neg</sub> </em>with the considered parameters is approximately equal to 0.7. For incomplete unweighted signed networks with d <em>∈</em> {0.25, 0.50}, we generate 20 replications for parameter values l<sub>0</sub> = 3, n <em>∈</em> {32, 36, 40}, q<sub>m</sub> <em>∈</em> {0.1, 0.2, 0.3, 0.4, 0.5, 0.6} and <em>q<sub>neg</sub> </em><em>∈</em> {0.3, 0.5, 0.7}. In total, we produce 600 and 1,080 instances for complete and incomplete networks, respectively, which makes a total of 1,680 instances.</li> <li>partitions: folder containing the partitioning results of two methods: EnumCC(3) vs. OneTreeCC(). Note that the results of OneTreeCC() are not shown for space considerations, except for those with n=50.</li> <li>Results</li> </ul> <li>Experiments for Dataset 2</li> <ul> <li>benchmark netwoks: We generate these complete and incomplete networks through our random signed network generator, which is publicly available online. The optimal solution for a generated network is known by construction. For a given n, d and l<sub>0</sub>, we first create a perfectly structurally balanced (i.e., internally positive and externally negative) signed network with a built-in module structure. The underlying module structure constitutes the optimal partition. Then, in order to take into account different positive to negative ratio values for internal and external edges we generate several signed networks by perturbing the initial signed network without affecting its underlying optimal partition, thanks to its definition of stability range. We generate signed networks with parameter values n <em>∈</em> {30, 40, 50, 60, 70, 90}, d <em>∈</em> {0.25, 1.00} and l<sub>0</sub> <em>∈</em> {2, 4, 6}. In total, we produce 214 and 184 instances for complete and incomplete networks, respectively, which makes a total of 398 instances.</li> <li>benchmark partitions: folder containing the partitioning results of two methods: CoNS(<em>r<sub>max</sub></em>) with vs. without MVMO pruning, where <em>r<sub>max</sub></em> ∈ {3,4}.</li> <li>results: two `csv` files containing benchmark results between CoNS(<em>r<sub>max</sub></em>) with vs. without MVMO pruning, where r<sub>max</sub> ∈ {3,4}.</li> </ul> </ul> </div> </div> </div>