DATA_DICTIONARY.md

# Data dictionary

This dataset contains two useful views of the same source data.

## `data/train.jsonl`

Row-level anonymized crowd feedback votes.

| Field | Meaning |
|---|---|
| `id` | Stable public row ID generated for this release |
| `source_record_id` | Original feedback record ID |
| `record_type` | Always `crowd_feedback_vote` |
| `question` | Geopolitical market/question title |
| `category` | Dataset category, currently `geopolitics` |
| `market_id` | Original market/pod ID |
| `epoch` | Epoch associated with the vote |
| `feedback` | Qualitative feedback label/comment |
| `vote` | `up_vote` or `down_vote` |
| `up_vote` | Boolean vote direction |
| `votes` | Numeric vote count from source system |
| `voting_power` | Weight attached to the feedback vote |
| `created_at` | Original creation timestamp |
| `updated_at` | Original update timestamp |
| `deleted` | Whether the source row was deleted |
| `version` | Source schema/version field |
| `split` | Dataset split |

## `data/market_summary.jsonl`

One row per geopolitical question/market, aggregated from `data/train.jsonl`.

| Field | Meaning |
|---|---|
| `id` | Stable public market-summary ID |
| `record_type` | Always `market_feedback_summary` |
| `question` | Geopolitical market/question title |
| `category` | Dataset category |
| `market_id` | Original market/pod ID |
| `num_feedback_votes` | Number of feedback votes attached to the question |
| `up_votes` | Number of up-votes |
| `down_votes` | Number of down-votes |
| `up_vote_share` | Up-votes divided by total feedback votes |
| `total_voting_power` | Sum of voting power across feedback votes |
| `median_voting_power` | Median voting power |
| `top_feedback_tags` | Most common feedback labels/comments |
| `first_seen_at` | Earliest feedback timestamp |
| `last_seen_at` | Latest feedback timestamp |
| `split` | Dataset split |

## Recommended training setup

For a simple supervised setup:

- Input: `question` + `feedback`
- Label: `up_vote` or `vote`
- Sample weight: `voting_power`

For market-level ranking or triage:

- Input: `question`
- Targets/features: `up_vote_share`, `num_feedback_votes`, `total_voting_power`, `top_feedback_tags`

## Privacy changes

The source CSV contained `voter_id` and `private_subnet_id`. Those fields are intentionally excluded from this public release.