The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Dataset Card for Dataset Name
Transcriptions of general debates at the UN from 1970 to 2016, chunked, with embedding and reduced vectors.
Dataset Details
This dataset consists of 37307 chunked portions of transcripts, each approximately 250 characters, with metadata consisting of the UN session, country of speaker, year, original full transcript, embedding vectors, (using all-mpnet-base-v2), UMAP reduced vectors and information weight score.
The original dataset was prepared by Alexander Baturo, Niheer Dasandi, and Slava Mikhaylov, and is presented in the paper "Understanding State Preferences With Text As Data: Introducing the UN General Debate Corpus" Research & Politics, 2017.
Preparation
This dataset was prepared in Google Colab, the notebook is publicly available: https://colab.research.google.com/drive/1O--Pc-kYs0e6Jua9FjYE7-21pqH8Hv41?usp=sharing
- Downloads last month
- 13