Assess News Relevance

Requirements Clarification & Assessment

• Define Relevance: Understand how relevance is defined. Is it a binary classification (relevant/not relevant), or is there a continuous scale of relevance?
• User Context: Determine whether relevance is general or user-specific. Is the model personalized to individual user preferences or general for all users?
• Feedback Mechanism: Clarify if implicit signals (click-through rate, dwell time) or explicit feedback (likes, shares) will be used to determine relevance.
• Dataset Characteristics: Assess if the dataset is balanced or imbalanced in terms of relevant and non-relevant news articles.
• Business Objective: Understand the business goals. Is the model intended to increase user engagement, time spent on the platform, or another metric?
• Baseline Model: Determine if there is an existing model or baseline to compare the new model against.