This dataset collects the slides that were presented at the Data Collaborations Across Boundaries session in SciDataCon 2022, part of the International Data Week.

The following session proposal was prepared by Tyng-Ruey Chuang and submitted to SciDataCon 2022 organizers for consideration on 2022-02-28. The proposal was accepted on 2022-03-28. Six abstracts were submitted and accepted to this session. Five presentations were delivered online in a virtual session on 2022-06-21.

Data Collaborations Across Boundaries

There are many good stories about data collaborations across boundaries. We need more. We also need to share the lessons each of us has learned from collaborating with parties and communities not in our familiar circles.

By boundaries, we mean not just the regulatory borders in between the nation states about data sharing but the various barriers, readily conceivable or not, that hinder collaboration in aggregating, sharing, and reusing data for social good. These barriers to collaboration exist between the academic disciplines, between the economic players, and between the many user communities, just to name a few. There are also cross-domain barriers, for example those that lay among data practitioners, public administrators, and policy makers when they are articulating the why, what, and how of "open data" and debating its economic significance and fair distribution. This session aims to bring together experiences and thoughts on good data practices in facilitating collaborations across boundaries and domains.

The success of Wikipedia proves that collaborative content production and service, by ways of copyleft licenses, can be sustainable when coordinated by a non-profit and funded by the general public. Collaborative code repositories like GitHub and GitLab demonstrate the enormous value and mass scale of systems-facilitated integration of user contributions that run across multiple programming languages and developer communities. Research data aggregators and repositories such as GBIF, GISAID, and Zenodo have served numerous researchers across academic disciplines. Citizen science projects and platforms, for instance eBird, Galaxy Zoo, and Taiwan Roadkill Observation Network (TaiRON), not only collect data from diverse communities but also manage and release datasets for research use and public benefit (e.g. TaiRON datasets being used to improve road design and reduce animal mortality). At the same time large scale data collaborations depend on standards, protocols, and tools for building registries (e.g. Archival Resource Key), ontologies (e.g. Wikidata and schema.org), repositories (e.g. CKAN and Omeka), and computing services (e.g. Jupyter Notebook). There are many types of data collaborations. The above lists only a few.

This session proposal calls for contributions to bring forward lessons learned from collaborative data projects and platforms, especially about those that involve multiple communities and/or across organizational boundaries. Presentations focusing on the following (non-exclusive) topics are sought after:

1. Support mechanisms and governance structures for data collaborations across organizations/communities.

2. Data policies --- such as data sharing agreements, memorandum of understanding, terms of use, privacy policies, etc. --- for facilitating collaborations across organizations/communities.

3. Traditional and non-traditional funding sources for data collaborations across multiple parties; sustainability of data collaboration projects, platforms, and communities.

4. Data workflows --- collection, processing, aggregation, archiving, and publishing, etc. --- designed with considerations of (external) collaboration.

5. Collaborative web platforms for data acquisition, curation, analysis, visualization, and education.

6. Examples and insights from data trusts, data coops, as well as other formal and informal forms of data stewardship.

7. Debates on the pros and cons of centralized, distributed, and/or federated data services.

8. Practical lessons learned from data collaboration stories: failure, success, incidence, unexpected turn of event, aftermath, etc. (no story is too small!).

This dataset contains 30 independent contracts in English, each with realistic structure, signatures, dates, and values. The contracts were generated synthetically to simulate real-world agreements across different domains, including:

Real Estate (house and apartment financing contracts)

Vehicles (car and motorcycle financing contracts)

Education (student loan agreements)

Health (medical and insurance-related contracts)

General Financing (personal and business loan contracts)

Each contract is between 2 and 3 pages, written in professional legal style, and includes fictitious names, addresses, values, and bank standards. They are useful for projects in:

Natural Language Processing (NLP)

Information Extraction (IE)

LegalTech and FinTech AI applications

Training/Testing OCR models

Document classification and summarization research

No personal or sensitive information is included — all data is fully synthetic.

Use this guide to find information on Tempe data policy and standards.Open Data PolicyEthical Artificial Intelligence (AI) PolicyEvaluation PolicyExpedited Data Sharing PolicyData Sharing Agreement (General)Data Sharing Agreement (GIS)Data Quality Standard and ChecklistDisaggregated Data StandardsData and Analytics Service Standard

Forms and Templates

This datasets contains a register of the Information Sharing Agreements (ISA) the Leicester City Council holds with other organisations.Leicester City Council only shares information with outside organisations when it is necessary, for example effective service delivery and/or to protect the welfare of individuals and only when legislation states that sharing for a specific purpose is allowed.Copies of the agreements are available upon request from info.requests@leicester.gov.uk (with any necessary redactions e.g. signatures)

Metadata Portal Metadata Information

A partial list of contracts the State Department of Health started or amended between July 1, 2017 and June 30, 2018. Includes grants and contracts for goods and professional services tracked in the department's primary contract database, the Enterprise Contract Management System (ECMS). It does not include contracts many of the department's contracts with Washington's local health jurisdictions, or contracts for expert witnesses, purchase orders, data sharing agreements, contracts issued by the department but not tracked in ECMS, or contracts exempt from disclosure under state or federal regulation. In 2017, the department added a data element to categorize entities. This is not found in prior years' data sets. Acronyms commonly found in this data set are: CBO=Community Based Organizations/Non-Profits CLH=Local Health Jurisdictions EMS= EMS/Trauma Centers GVF=Government Federal GVL=Government Local (EXCEPT Con-Con/LHJ) GVS=Government State (EXCEPT Higher Ed) HED=Higher Education HSP=Hospitals POP = Period of Performance PRV=Private/For-Profits SCH=Schools, School Districts & Education Institutions (excluding Higher Ed) SOW = Statement of Work TRB=Tribal Entity

Context

In early 2016, The Washington Post wrote that the Justice Department is "resuming a controversial practice that allows local police departments to funnel a large portion of assets seized from citizens into their own coffers under federal law.

The "Equitable Sharing Program" gives police the option of prosecuting some asset forfeiture cases under federal instead of state law, particularly in instances where local law enforcement officers have a relationship with federal authorities as part of a joint task force. Federal forfeiture policies are more permissive than many state policies, allowing police to keep up to 80 percent of assets they seize." (link to the full article can be found here).

This is the raw data from the Department of Justice’s Equitable Sharing Agreement and Certification forms that was released by the U.S. Department of Justice Asset Forfeiture and Money Laundering Section.

Content

• spending_master.csv is the main spending dataset that contains 58 variables.

• notes.csv lists the descriptions for all variables.

Acknowledgements

The original dataset can be found here. The data was originally obtained from a Freedom of Information Act request fulfilled in December 2014.

Inspiration

• Which agency/sector/item received the most amount of funds from the Justice Department?

• How many agencies received non-cash assets from the federal government through Equitable Sharing?

• Are there any trends in the total equitable sharing fund across agencies?

We propose a new platform for user modeling with blockchains that allows users to share data without losing control and ownership of it and applied it to the domain of travel booking. Our new platform provides solution to three important problems: ensuring privacy and user control, and incentives for sharing. It tracks who shared what, with whom, when, by what means and for what purposes in a verifiable fashion. The paper presents a case study of applying the framework for a hotel reservation system as one of the enterprise nodes of Multichain which collects users' profile data and allows users to receive rewards while sharing their data with other travel service providers according to their privacy preferences expressed in smart contracts. The user data from the repository is converted into an open data format and shared via stream in the blockchain so that other nodes can efficiently process and use the data. The smart contract verifies and executes the agreed terms of use of the data and transfers digital tokens as a reward to the user. The smart contract imposes double deposit collateral to ensure that all participants act honestly. The paper also presents a performance evaluation of the new platform by analyzing latency and memory consumption with selected three test-scenarios and measuring the transaction cost for smart contracts deployment. The results show that the node responded quickly in all our cases with a befitting transaction cost.

🇨🇦 캐나다

Dataset

This dataset was created by Panda Xiong

Released under Community Data License Agreement - Sharing - Version 1.0

Contents

This Technical Bulletin provides information about data exchange standards, which can reduce the costs and time required to develop a data exchange. The TB also reviews other factors for agencies to consider when developing data exchanges, such as data quality, data sharing agreements, data governance, and security.

Metadata-only record linking to the original dataset. Open original dataset below.

Dies sind die Data Sharing Agreements (DSAs), die der London Borough of Barnet mit anderen Organisationen geschlossen hat. Meistens wird ein DSA verwendet, wenn es keinen Vertrag zwischen dem Rat und einer anderen öffentlichen Organisation gibt. Personen, die eine DSA unterzeichnen, erhalten keine gesetzlichen Rechte zum Teilen von Daten; die Weitergabe von Daten muss bereits legal sein. In den DSAs werden die Regeln für die Weitergabe personenbezogener Daten aus den genannten Gründen festgelegt.

Personenbezogene Daten, wie die Namen und Unterschriften von Beamten, wurden aus Gründen der DSGVO aus DSA entfernt.

This Toolkit is intended for staff at all levels of government who work within offices and agencies that promote the well-being of children and families and would like to know more about:

This includes information about the:

This toolkit also includes several samples of sharing agreements that offices and agencies can use in their own agreements.

We hope it will help offices and agencies that want to develop and expand responsible sharing activities, and dispel common misconceptions about confidentiality requirements that are often raised when trying to responsibly share records.

We would like to know about the individuals using this toolkit so we can further tailor our future work to you. Please take a moment to send us an email telling us:

Send Email

PAPERWORK REDUCTION ACT OF 1995 (Pub. L. 104-13) STATEMENT OF PUBLIC BURDEN: Through this information collection, ACF is gathering information to learn about the individuals using this toolkit so we can tailor our future work to better assist them. Public reporting burden for this collection of information is estimated to average of 10 minutes per respondent, including the time for reviewing instructions, gathering and maintaining the data needed, and reviewing the collection of information. This is a voluntary collection of information. agency may not conduct or sponsor, and a person is not required to respond to, a collection of information subject to the requirements of the Paperwork Reduction Act of 1995, unless it displays a currently valid OMB control number. The OMB # is 0970-0401 and the expiration date is 06/30/2024. If you have any comments on this collection of information, please contact Joshua.Williams@acf.hhs.gov

Individuals who receive human services and other assistance often receive those services from several independent programs. Providing a case worker records from multiple programs can improve their understanding of a recipient and how to better serve them. Providing a researcher records from multiple programs can improve their understanding of the overall system and how to improve services for all recipients. However, sharing records with other offices and agencies often raises legitimate concerns, such as whether it (a) complies with applicable law, (b) meets individuals’ privacy expectations, and (c) creates a security risk.

This toolkit discusses how to share records collected by human services and related programs. It summarizes the key federal requirements that determine when different records may be shared. It explains how leaders and workgroups can help resolve challenges that arise when developing sharing plans. It advises how offices and agencies might secure electronic records. It also includes success stories, documents used to facilitate record sharing, and links to helpful online resources.

This toolkit will not replace the important practice of consulting legal counsel. However, we hope it will give ideas on what is possible, help resolve concerns that arise, and generally aid and inspire responsible record sharing.

Developing processes to share records with other offices and agencies that balance the many competing priorities is complex but not insurmountable. Human services and related programs are often governed by an overlapping web of requirements designed to protect the confidentiality of the individuals connected to those services. Stakeholders often bring additional concerns and interests to the table.

There are many ways to help ensure a responsible record sharing initiative is successful. These can include:

Building Trust: Stakeholders often approach all new record sharing initiatives with a range of legitimate and unfounded concerns; concerns, and especially unfounded concerns, often dissipate once stakeholders begin to trust the other parties and their intents.

Gabay, Mary et al. (2021). Confidentiality Toolkit, OPRE Report 2021-175, Washington, DC: Office of Planning, Research, and Evaluation, Admini

Forms and Templates

The data files contain information about the preferences of bachelor 1 and 2 students obtained via a discrete choice experiment (12 choice tasks per respondent), demographic characteristics of the sample and population, experiences with free-riding, attitude towards teamwork, and a measure of individualism/collectivism. Students were presented a different grade weight before each choice task (i.e., 10%, 30%, or 100%). The data was collected from mid-June to mid-July 2021.

Access to the data is subject to the approval of a data sharing agreement due to the personal information contained in the dataset.

A summary of the publication can be found below: Reducing free-riding is an important challenge for educators who use group projects. In this study, we measure students’ preferences for group project characteristics and investigate if characteristics that better help to reduce free-riding become more important for students when stakes increase. We used a discrete choice experiment based on twelve choice tasks in which students chose between two group projects that differed on five characteristics of which each level had its own effect on free-riding. A different group project grade weight was presented before each choice task to manipulate how much there was at stake for students in the group project. Data of 257 student respondents were used in the analysis. Based on random parameter logit model estimates we find that students prefer (in order of importance) assignment based on schedule availability and motivation or self-selection (instead of random assignment), the use of one or two peer process evaluations (instead of zero), a small team size of three or two students (instead of four), a common grade (instead of a divided grade), and a discussion with the course coordinator without a sanction as a method to handle free-riding (instead of member expulsion). Furthermore, we find that the characteristic team formation approach becomes even more important (especially self-selection) when student stakes increase. Educators can use our findings to design group projects that better help to reduce free-riding by (1) avoiding random assignment as team formation approach, (2) using (one or two) peer process evaluations, and (3) creating small(er) teams.

The Office of Contracting and Procurement (OCP) is required to post contract awards valued at $100,000 or more for agencies served by OCP. The awarded contracts database includes a caption describing the type of goods or services provided, the contract number, the ordering agency, the contract amount, the period of time covered by the contract award, the contractor receiving the award, and the market type. To obtain a copy of any contract, you may submit a FOIA request online via the DC government Public FOIA Portal. Data has been updated to include agency budget code, name, and acronym attributes. Budget codes were used to assign the agency name and acronym to each record. Agencies that share the same budget code, such as those under the Executive Office of the Mayor, were left blank in PASS records. For questions regarding details within the data, contact the Office of Contracting and Procurement at https://contracts.ocp.dc.gov/contact.

Data Contracts for AI Market Outlook

According to our latest research, the global Data Contracts for AI market size reached USD 1.65 billion in 2024, reflecting robust expansion driven by increasing adoption of AI-powered data management solutions. The market is expected to grow at a remarkable CAGR of 23.7% from 2025 to 2033, positioning the industry to achieve a value of approximately USD 13.45 billion by 2033. This significant growth is propelled by the rising need for robust data governance, regulatory compliance, and secure data sharing frameworks across industries leveraging artificial intelligence.

One of the primary growth factors fueling the Data Contracts for AI market is the escalating demand for data governance and compliance management in an era of stringent data privacy regulations. Organizations are increasingly recognizing the importance of establishing clear, enforceable agreements—known as data contracts—to ensure the responsible use, sharing, and processing of data within AI systems. The proliferation of global data protection laws such as the GDPR, CCPA, and emerging regulations in Asia Pacific and Latin America is compelling enterprises to adopt solutions that facilitate traceability, accountability, and transparency in AI-driven data workflows. As a result, the market is witnessing heightened investment from industries such as BFSI, healthcare, and government, where regulatory compliance is paramount.

Another significant driver of market growth is the rapid advancement and integration of AI technologies across diverse business functions, which in turn amplifies the complexity and volume of data being processed. With AI models relying heavily on high-quality, well-governed datasets, data contracts are becoming essential tools for defining data ownership, access rights, and usage limitations. This is particularly crucial in collaborative environments where data is sourced from multiple internal and external stakeholders. The need to mitigate risks associated with data breaches, unauthorized access, and model bias is prompting organizations to invest in advanced data contract solutions that offer granular control and real-time monitoring capabilities, further accelerating market expansion.

The increasing adoption of cloud-based AI platforms and the emergence of multi-cloud and hybrid environments are also contributing to the growth trajectory of the Data Contracts for AI market. As enterprises transition their data infrastructure to the cloud to enhance scalability and flexibility, they face new challenges related to data integration, interoperability, and security. Data contracts play a pivotal role in addressing these challenges by standardizing data exchange protocols and ensuring compliance across distributed environments. Moreover, the rise of data marketplaces and data-as-a-service models is creating new opportunities for monetizing data assets, thus underscoring the need for robust contractual frameworks to manage data rights and obligations in AI ecosystems.

Regionally, North America continues to dominate the Data Contracts for AI market, accounting for the largest share in 2024, followed closely by Europe and Asia Pacific. The presence of leading AI technology providers, early adoption of digital transformation initiatives, and a mature regulatory landscape are key factors driving market leadership in these regions. Meanwhile, Asia Pacific is emerging as the fastest-growing region, fueled by rapid industrialization, increasing investments in AI research, and evolving data privacy regulations. Latin America and the Middle East & Africa are also witnessing steady growth, albeit from a smaller base, as enterprises in these regions begin to recognize the strategic value of data contracts in AI deployments.

Component Analysis

The Data Contracts for AI market is segmented by component into Software, Services, and Platforms, each playing a distinct role in the overall ecosystem. The software segment encompasses sol

This repository is used for archiving the classification of papers in the following article: Data: to share or not to share? A Semi-Systematic Literature Review in (rational) data sharing in inter-organizational systems.

This archive is supporting the paper for the labels that have been given to the papers found in the Semi-Systematic Literature Review.

The following categories and labels are possible in the classification of the literature:

Category ["Paper"] - Labels: Summarized name(s) of the author(s).

Category ["Knowledge dimension"] - Labels: ["Data", "Information", "Knowledge", "Data/information (paper is ambiguous on whether it is data or information that is shared)", "None", "Data/Knowledge (data is learned and transferred in to knowledge, the knowledge is shared)"]

Category ["Type of industry"] - Labels: ["None", "Supply chain", "Healthcare", "Vehicles", "Agricultural", "Research", "Networks", "Automotive", "Innovation", "Smart Grid", "Social media", "R&D", "e-Governance", "Construction Sector", "Government-enterprise", "Manufacturing", "Power grid", "Smart Cities", "Personal data", "Cyber security", "E-commerce", "Maritime", "Online marketplaces", "Assembly", "Engineering", "Communities of Practice", "Knowledge Management Systems", "B2B commerce", "Fisheries", "Outsourcing", "High-tech firms", "Crisis", "e-Services", "Seaports", "Horticulture", "Data markets", "Media", "Cultural Heritage Institutions", "Fresh Products", "Knowledge market", "Ecological", "Ride Sharing", "Government", "Transit", "Medical", "Virtual Research Organization", "Energy", "Oil and gas", "Education"]

Category ["Game theory approach"] - Labels: ["None", "Non-cooperative game", "Evolutionary", "Cooperative game", "Stackelberg", "Auction", "Unclear defined", "Diffusion kernels", "Markov game", "Negotiation", "(Non-)cooperative game", "Differential game", "Pricing", "Bayesian", "Contract theory", "Non-collusion", "Stochastic differential game", "Fisher’s market", "Hotelling", "Bargaining", "Access control"]

Category ["Technologies of interest"] - Labels: ["blockchain", "None", "smart contracts", "federated learning", "internet of things", "machine learning", "cloud computing", "artificial intelligence", "ethereum", "5G", "data trust", "deep neural networks", "digital twins", "internet of (medical) things", "smart grid", "data governance", "artifical intelligence", "collaborative learning", "smart contract", "encryption", "data escrow", "NFT", "data mining", "semantic technologies", "transfer learning"]

Category ["Level of trust"] - Labels: ["Calculus-Based Trust", "None", "security", "privacy", "integrity", "transparency", "authentication", "confidentiality", "traceability", "verification", "Knowledge-Based Trust", "privacy"]

Category ["Type of contract"> - Labels: ["None", "Smart contracts", "Contract theory", "Linear wholesale price contract", "GMP and IPD contracts", "General contract", "Incentive contract", "Trust-embedded contract", "Wholesale price contract"]

Tree Contract Work Details contains information related to work performed on specific tree work contracts. This dataset can be joined to the Forestry Work Orders dataset (https://data.cityofnewyork.us/Environment/Forestry-Work-Orders/bdjm-n7q4) by joining work_order_id from Tree Contract Work Details to OBJECTID from Forestry Work Orders. There are many work_order_ids per single contract_number. Not all work orders in Forestry Work Orders are represented in this dataset because it is only used to track work on specific contracts. Data Dictionary found here: https://docs.google.com/spreadsheets/d/1rAQL_d8yQ5axRIGl9vLxpr73rDmQUZ23VKQC8QQWIYI/edit?usp=sharing