This paper describes a differentially private post-processing algorithm for learning attribute-aware fair regressors.

In 2018, only 19 percent of Poles declared that they were familiar with the term "fair trade".

Image reconstruction-based anomaly detection models are widely explored in industrial visual inspection. However, existing models usually suffer from the trade-off between normal reconstruction fidelity and abnormal reconstruction distinguishability, which damages the performance.

This research employed a mixed methods online survey to investigate research software funders’ perspectives.

All participants gave informed consent at the start of the online survey. The University of Illinois Urbana-Champaign Institutional Review Board (no. 24374) reviewed the study and determined it exempt.

Data collection took place from December 2023 to May 2024. The mean completion time for the detailed survey was 28 minutes and 13 seconds. The data were cleaned and prepared for analysis by removing any identifiable respondent details.

Survey design

The survey began by collecting profile information, including institutional affiliation and job title. The survey primarily gathered detailed information about initiatives, policies, or programs to support research software but also included a much smaller set of questions about additional topics, such as strategic funding priorities and awareness of key concepts. The data generated from this survey are too extensive to report in a single manuscript. Here, we focus on the results generated via the set of questions asking about FAIR and FAIR4RS, specifically, the following survey items:

In addition, an open-ended question asked for further detail about the respondents’ assessments of FAIR4RS’s relevance to their work.

Sampling

The survey targeted international research funders, including governmental and non-governmental (e.g., philanthropic) organizations. An initial contact list was created based on participation in the Research Software Association (ReSA) and known responsibilities for research software funding among the authors' networks. This list was refined by removing individuals who had moved to unrelated professional roles or were unavailable long-term due to personal issues.

The final contact list comprised 71 people at 37 funding organizations. After excluding individuals when a member of their organization had already provided a complete response or when the person was no longer working on a relevant topic or was otherwise unavailable (total of n=30), 41 people remained. Of these, five did not complete the survey, while 36 individuals (representing 30 research funding organizations) did, yielding a response rate of 87.8% (and representing 81% of the original organizations). Fully completed survey responses were not required for inclusion in the sample, resulting in varied sample sizes across different survey questions.

The respondents represented governmental (n=26), philanthropic (n=6), and corporate (n=1) research funders.

Respondents’ job titles spanned the following categories: Senior Leadership and Executive (e.g., Vice President of Strategy); Program and Project Management (e.g., Senior Program Manager); Planning and Business Development; and Scientific, Technical, and IT roles (e.g., Scientific Information Lead).

Most respondents, 72.7% (n=24), answered “Yes” to the question, “Has your organization established any policies, initiatives, or programs aimed at supporting research software?” Meanwhile, 18.2% (n=6) said “No,” and 9.1% (n=3) were “Unsure.”

Regarding geographic distribution in the achieved sample, most survey respondents were from North America and Europe, with 15 and 12 participants, respectively. The sample also comprised 4 participants from South America, 3 from Oceania, and 1 from Asia, reflecting a global but uneven representation across continents. Some participating funders covered a broad spectrum of disciplines, while others focused on specific domains such as social sciences, health, environment, physical sciences, or humanities.

Around half of Japanese consumers were aware of the term "Fairtrade" as revealed in a survey conducted in October 2024. Young people aged 15 to 19 years showed the highest awareness of fair trade, with over ** percent stating to know the term and its meaning, while another **** percent had at least heard of the term.

This statistic shows the awareness of fair trade products in Great Britain from 2007 to 2015. According to the survey conducted biennially for Bord Bia Irish Food Board in 2013, ** percent of respondents had heard of Fair Trade certified products.

The share of the respondents who knew about the opposition protests that took place in the Russian capital due to non-admission of several candidates to the 2019 local elections increased from 41.2 percent to 62.4 percent between July 20-22 and July 29-31, 2019. In the latter period under consideration, approximately 16 percent of Moscow residents were not aware of the protest actions.

This record describes a survey around the awareness of the FAIR data principles, undertaken in Australia in 2017 by ANDS, Nectar and RDS. ANDS (Australian National Data Service), Nectar (National collaborative tools and resources), and RDS (Research Data Services) are NCRIS facilities. NCRIS is an Australian Federal Government investment in research infrastructure. ANDS(ands.org.au), Nectar(nectar.org.au) and RDS(rds.org.au) have integrated their work in line with proposals laid out in the NCRIS Roadmap (https://docs.education.gov.au/node/43736), early in 2017.

The survey was conducted as a Google Form, and analysed in a 12 page report (see Summary of Full Results - attached). Results of the demographics and quantitative responses are shared attached to this record. The qualitative responses are not shared, for reasons of confidentiality.

Background (from Summary Report)

ANDS/RDS/Nectar undertook a baseline survey to assess level of awareness around FAIR in the research community at eResearch Australasia conference (Oct 2017) and through an online survey. The online survey was closed a few weeks later on 16.11.17. A list of questions is provided (see Are you FAIR aware? Google Form.pdf). There were 249 responses.

According to a survey conducted in October 2024 in Japan, the awareness of the term "Fairtrade" was split among consumers. Around 52 percent of respondents neither knew the meaning nor had they heard the term before, whereas the other 48 percent had at least heard the term before.

This module is a part of the introductory online course “Data Steward Training”. In this course learners will develop the skills and gain knowledge on how to be a successful data steward. By the end of this module, learners will: Be able to explain the difference between FAIR and Open Data to researchers. Be able to make data FAIR by using a set of practical guidelines and tools. Understand the range of skills and knowledge associated with data stewardship. Be aware of different approaches to stewardship service provision. Be able to identify key elements that help make research data discoverable, accessible, interoperable and reusable. Be able to practise making data FAIR. This module is suitable for data stewards and trainers seeking introductory learning material. It will take around 1 hour and 30 minutes to complete the module. The materials include video presentations, full video transcripts, PowerPoint presentations and various learning activities and resources to support learning. This learning material has been developed in collaboration with the FAIRsFAIR and EOSC Synergy projects and has been adapted from the Data Steward Instructor Training Workshops run throughout 2020 and 2021. The The Data Steward Training course contains a total of 5 modules, each of which has been uploaded to Zenodo separately. Please see the related identifiers for the DOI's for each module. Please note: that module 4 (Data Management Plans) has been re-used and was originally published as module 8 of the EOSC synergy course: Bringing synergy to better data management and research in Europe - https://doi.org/10.5281/zenodo.5517465

Questions and Answers are organised in a tabular fashion. The questions act as titles of the columns. Recorded answers are text-based, nominal, and ordinal. The theme of the questions enable the assessment awareness of FAIR principles and data management practices for early career scientists (PhDs) in geography. Some questions were reused from related surveys with permission. No data was collected to identify the individual. Contact details for follow up interviews were removed from the questionnaire results prior to the upload of the dataset.

Dataset Card for m4-bias-eval-fair-faces

This dataset consists of generations made by the 80 Billion and 9 Billion variants of the IDEFICS (Image-aware Decoder Enhanced à la Flamingo with Interleaved Cross-attentionS) model. IDEFICS is an open-access reproduction of Flamingo, a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary sequences of image and text inputs and produces text outputs. In order to evaluate the model's… See the full description on the dataset page: https://huggingface.co/datasets/HuggingFaceM4/m4-bias-eval-fair-face.

A list of 30+ online resources have been identified and curated by the FAIR4RS Subgroup 5: Adoption Guidelines. These resources are available as the supplementary materials of the report (Martinez et al., 2022) and can be downloaded and cited from this landing page.

The list is open for additions by the community via comments directly to this link. We particularly encourgae authors of new and exisiting resources to add as much detail as possible to describe their resource and its relevance to the FAIR4RS Principles. Each of the columns has a description and whether the information is optional or not. Whe plan to add tags to the added resources for each semester and when there is another set of 30 resources we can resealease a new version.

This list reflects the wide spectrum of global contributions supporting the implementation of the FAIR Principles, particularly regarding research software. It is a snapshot of currently available resources, although we expect that new resources will become available in the future and that the contents of the current list will evolve. It is important to note that most of these resources precede the definition of the FAIR4RS Principles; however, these still support their implementation.

The resources were manually collected, analyzed, and categorized according to their type: guidelines, tools, metadata schemas and registries/repositories. For each resource detail is also provided on which of the FAIR4RS Principles that the resource supports.

Data collection

This subgroup initiated a crowdsourcing effort to identify relevant resources. All members had the opportunity to provide and describe existing FAIR research software guidelines and tools. During the first two months of the subgroup operation in 2021, subgroup participants (referred to as data providers) added resources to an online spreadsheet. Data providers were encouraged to list resources that they were aware of, authored, or were supported by their institutions. Subsequently, the subgroup organized virtual calls to discuss the resources, their descriptions and the categorization. Over the next two months each data provider added descriptions to resources they were familiar with. This meant that some resources gained descriptions from different data providers. Before the completion of the list, the subgroup leads checked the list and cleaned it (removing items that lacked information or providing complementary information). The resulting list is the first crowdsourced list of its type and it welcomes your contributions!

The Intellectual Property Office has written daily life questions and answers commonly written in each chapter of the Copyright Act (copyright and author, copyright, assignment of copyright property rights, exercise and extinction, fair use, copyright collective management organizations, remedies for copyright infringement, penalties) in order to provide the public with a basic understanding of copyright concepts. This has been made into a "Understanding Copyright" available for free download and use on the government's open data platform.

A new approach addressing the spectrum scarcity challenge in 6G networks by implementing an enhanced licensed shared access (LSA) framework is considered. The proposed mechanism aims to ensure fairness in spectrum allocation to mobile network operators (MNOs) through a novel weighted auction called the fair Vickery-Clarke-Groves (FVCG) mechanism in which the determination of weights is based on the results of the previous auctions.

Dataset Description

Context and Methodology

• Research Domain/Project:
  This dataset was created for a machine learning experiment aimed at developing a classification model to predict outcomes based on a set of features. The primary research domain is disease prediction in patients. The dataset was used in the context of training, validating, and testing.

• Purpose of the Dataset:
  The purpose of this dataset is to provide training, validation, and testing data for the development of machine learning models. It includes labeled examples that help train classifiers to recognize patterns in the data and make predictions.

• Dataset Creation:
  Data preprocessing steps involved cleaning, normalization, and splitting the data into training, validation, and test sets. The data was carefully curated to ensure its quality and relevance to the problem at hand. For any missing values or outliers, appropriate handling techniques were applied (e.g., imputation, removal, etc.).

Technical Details

• Structure of the Dataset:
  The dataset consists of several files organized into folders by data type:

  • Training Data: Contains the training dataset used to train the machine learning model.

  • Validation Data: Used for hyperparameter tuning and model selection.

  • Test Data: Reserved for final model evaluation.

  Each folder contains files with consistent naming conventions for easy navigation, such as train_data.csv, validation_data.csv, and test_data.csv. Each file follows a tabular format with columns representing features and rows representing individual data points.

• Software Requirements:
  To open and work with this dataset, you need VS Code or Jupyter, which could include tools like:

  • Python (with libraries such as pandas, numpy, scikit-learn, matplotlib, etc.)

Further Details

• Reusability:
  Users of this dataset should be aware that it is designed for machine learning experiments involving classification tasks. The dataset is already split into training, validation, and test subsets. Any model trained with this dataset should be evaluated using the test set to ensure proper validation.

• Limitations:
  The dataset may not cover all edge cases, and it might have biases depending on the selection of data sources. It's important to consider these limitations when generalizing model results to real-world applications.

Additional file 2: Supplementary Table 2. Factors Associated with Knowledge of Alcohol-Cancer Risk by Cancer Type (Straight Liners Excluded). A table formatting the same way as Table 3 in the manuscript, but excluding straight liners. Mean differences and confidence intervals are included.

This German Crop BioGreenformatics Network (GCBN, https://www.denbi.de/gcbn) user training introduces FAIR-aware data management for plant genomic and phenomic research data. It was part of the workshop series at the International Symposium on Integrative Bioinformatics 2018, Rothamsted Research, Harpenden, UK. The first part by Matthias Lange ('Part_1_Lange_FAIR.pdf') give a comprehensive motivation about FAIR research data management and publication. In the second part by Daniel Arend ('Part_2_Arend_eDAL.pdf') an overview and usage example how to set up a e!DAL based repository (https://edal.ipk-gatersleben.de) using the 'bringing the infrastructure to the data' approach is presented.

Context and Methodology

The dataset originates from the research domain of Customer Churn Prediction in the Telecom Industry. It was created as part of the project "Data-Driven Churn Prediction: ML Solutions for the Telecom Industry," completed within the Data Stewardship course (Master programme Data Science, TU Wien).

The primary purpose of this dataset is to support machine learning model development for predicting customer churn based on customer demographics, service usage, and account information.
The dataset enables the training, testing, and evaluation of classification algorithms, allowing researchers and practitioners to explore techniques for customer retention optimization.

The dataset was originally obtained from the IBM Accelerator Catalog and adapted for academic use. It was uploaded to TU Wien’s DBRepo test system and accessed via SQLAlchemy connections to the MariaDB environment.

Technical Details

The dataset has a tabular structure and was initially stored in CSV format. It contains:

• Rows: 7,043 customer records

• Columns: 21 features including customer attributes (gender, senior citizen status, partner status), account information (tenure, contract type, payment method), service usage (internet service, streaming TV, tech support), and the target variable (Churn: Yes/No).

Naming Convention:

• The table in the database is named telco_customer_churn_data.

Software Requirements:

• To open and work with the dataset, any standard database client or programming language supporting MariaDB connections can be used (e.g., Python etc).

• For machine learning applications, libraries such as pandas, scikit-learn, and joblib are typically used.

Additional Resources:

• Source code for data loading, preprocessing, model training, and evaluation is available at the associated GitHub repository: https://github.com/nazerum/fair-ml-customer-churn

Further Details

When reusing the dataset, users should be aware:

• Licensing: The dataset is shared under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

• Use Case Suitability: The dataset is best suited for classification tasks, particularly binary classification (churn vs. no churn).

• Metadata Standards: Metadata describing the dataset adheres to FAIR principles and is supplemented by CodeMeta and Croissant standards for improved interoperability.

Fairtrade Chocolate Market Outlook

The global Fairtrade chocolate market size was valued at approximately USD 1.3 billion in 2023 and is projected to reach USD 2.4 billion by 2032, growing at a compound annual growth rate (CAGR) of 6.7% during the forecast period. This robust growth is primarily driven by increasing consumer awareness about ethical sourcing and sustainability, coupled with rising disposable incomes and changing consumer preferences toward high-quality, premium chocolate products.

One of the key growth factors driving the Fairtrade chocolate market is the rising consumer demand for ethically sourced products. Consumers are increasingly becoming aware of the social and environmental impact of their purchases, leading to a shift toward products that are certified fair trade. This certification ensures that producers receive fair compensation, which supports sustainable farming practices and improves the livelihoods of farmers. Additionally, the growing popularity of organic and natural food products is further propelling the demand for Fairtrade chocolates.

Another significant factor contributing to the market's growth is the expansion of retail channels. With the proliferation of supermarkets, hypermarkets, and specialty stores, Fairtrade chocolates are becoming more accessible to a broader consumer base. Online retailing has also emerged as a vital distribution channel, offering consumers the convenience of purchasing their favorite Fairtrade chocolate products from the comfort of their homes. E-commerce platforms are integrating advanced technologies such as artificial intelligence and machine learning to enhance the shopping experience, which is expected to boost sales further.

Additionally, the increasing use of Fairtrade chocolate in various applications such as confectionery, beverages, and bakery products is driving market growth. As consumers continue to seek out unique and high-quality food experiences, manufacturers are incorporating Fairtrade chocolate into innovative product offerings. This trend is particularly evident in the premium confectionery and artisanal bakery segments, where the use of ethically sourced ingredients can significantly enhance the product's appeal and marketability.

The rise of Organic Chocolate is also playing a crucial role in shaping the Fairtrade chocolate market. As consumers become more health-conscious and environmentally aware, the demand for organic products, including chocolate, has surged. Organic Chocolate, which is produced without synthetic fertilizers or pesticides, aligns with the values of sustainability and ethical sourcing that are central to the Fairtrade movement. This synergy not only enhances the appeal of Fairtrade chocolates but also broadens their market reach, attracting consumers who prioritize both health and ethical considerations in their purchasing decisions.

From a regional perspective, Europe dominates the Fairtrade chocolate market, followed by North America. The Europe market benefits from strong consumer awareness and well-established fair trade certification systems. However, the Asia Pacific region is expected to witness the highest growth rate during the forecast period, driven by increasing disposable incomes, urbanization, and a growing middle-class population with a preference for premium, ethically-sourced products.

Product Type Analysis

The Fairtrade chocolate market is segmented by product type into dark chocolate, milk chocolate, white chocolate, and others. Dark chocolate holds a significant share of the market due to its perceived health benefits and rich flavor profile. Consumers are increasingly opting for dark chocolate as it contains higher cocoa content and lower sugar levels compared to milk and white chocolate, making it a healthier choice. Moreover, the antioxidant properties of dark chocolate, which are linked to various health benefits such as improved heart health and reduced inflammation, are further driving its popularity.

Milk chocolate, on the other hand, remains a favorite among a broad demographic, particularly children and younger consumers. The creamy texture and sweetness of milk chocolate make it a versatile ingredient in various applications ranging from confectionery to beverages and bakery products. The ongoing innovation in flavor combinations and packaging is also contributing to the sustained demand for milk chocolate in the Fairtrade segment.