Define and measure article trending

Define what “trending” means for articles and how to measure it. Specify the raw event schema you require (e.g., article_id, user_id, timestamp, action_type, dwell_time, referrer, device, locale). Propose a scoring formula that combines signals (impressions, clicks, dwell time, reactions, reshares) with recency decay, de-duplication, and spam/bot controls. Explain time windows (e.g., 5m, 1h, 24h), baseline normalization (e.g., by author follower count or historical traffic), category/locale segmentation, and cold-start handling. Outline offline evaluation (precision/recall against editorial ground truth, calibration) and online A/B tests, success metrics (CTR, dwell time uplift, save/share rates), and guardrails (session depth, bounce rate, creator fairness).