Model-assisted research draft. The dataset contains 20,000 selected items from Telegram and YouTube, with original text/captions and selected original recording files. Labels have not received independent human validation. All acceptance, reuse and review flags retain their recorded values. Each sentiment has 5,000 items: positive, neutral, negative, and stronglynegative. Sources cover 200 Telegram channels, 200 public Telegram groups, and 57 YouTube channels. This is a selected, balanced sample; these counts do not estimate sentiment prevalence in the source population. The language labels are uzLatn (6,000), uzCyrl (2,388), uzrumixed (3,000), ru (4,000), spoken uz (4,541), en (65), and…
Independent publisher
Mahmud Nosirov
mahmudnosirov
Models in Library0
Datasets in Library1
Models on Hugging Face—
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Text classification.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.