Dataset

market-borrow-rate-short-interest-coherence-squeeze-risk-v0.1

Estimates, Events & Fund Flows Download 1 source

A small text-classification dataset by ClarusC64 that labels whether borrow-rate, short-interest and price signals cohere into a real short squeeze or diverge into a false signal. It is published on a model hub with an MIT license tag and a 10-row train split.

Open on Hugging Face

Data preview

Subset

Loading preview...

Dataset card

What this dataset contains

Each row is a text-classification example for a single ticker, pairing qualitative market signal fields with a yes/no-style label. The publisher describes its purpose as detecting when short-pressure signals align into a real squeeze versus when they diverge into false signals. The visible train split is tiny, at 10 rows, with the overall size category tagged as n<1K.

Structure and fields

Records are strings throughout, with core identifier columns uid, dataset_id and ticker. Per-ticker qualitative inputs include borrow_rate_signal, short_interest_signal, days_to_cover_signal, price_momentum_signal and volume_spike_signal, accompanied by an ops_summary text field. Supervisory fields comprise question, ground_truth_label, ground_truth_rationale, outcome_signal and source_citation, and the only listed split is train.

How analysts use it

Given the 10-row train split, the resource is best treated as a demonstration or evaluation toy rather than a production training set for short-squeeze detection models. An investment-fund workflow would more realistically use it to prototype prompt or rule logic for flagging coherent squeeze setups from borrow-rate and short-interest feeds, with a source_citation column available for audit trails. Any live deployment would require sourcing substantially more labelled examples elsewhere.

Documentation & references 1