引导式测验 · 中等 等级

人工智能在餐厅和菜单推荐测验中的应用

评估食物推荐、交互过滤、评论文本、检索、嵌入、可用性和用户上下文限制。

相关引导路径AI餐厅及菜单推荐
问题 1 的 8

协同过滤主要依靠什么来进行推荐？

预览每个问题8 问题

当您参加测验时，就会出现答案和解释。

  1. 协同过滤主要依靠什么来进行推荐？
    • 来自具有相似品味的用户的模式
    • 每道菜的营养成分
    • 餐厅的室内设计
    • 实时厨房温度
  2. 为什么评论挖掘 NLP 模型在食品应用程序中有用？
    • 他们把食物煮得更快
    • 他们总结情绪并从文本中提取菜肴级别的信号
    • 它们取代了对菜单的需求
    • 他们制定餐厅价格
  3. In a proposed two-stage food recommender, what would candidate retrieval do before ranking?
    • Confirm the restaurant reservation
    • Process the customer’s payment
    • Select a candidate subset for a later scoring stage
    • Guarantee the final recommendation matches every preference
  4. 嵌入在菜单推荐中有何用途？
    • 存储客户密码
    • 加密送货地址
    • 测量烤箱温度
    • 将菜肴表示为向量，以便相似的物品紧密结合在一起
  5. Why can a new diner with no recorded preferences pose a cold-start problem?
    • Interaction-based personalization has little user-history evidence to work from
    • The restaurant must change its menu
    • The model cannot represent food text
    • Opening hours become unavailable to everyone
  6. What can the MenuAI research system do with a menu image and a nutritional search key?
    • Book a table without user confirmation
    • Rank menu dishes according to a query such as calories or protein
    • Diagnose a food allergy from a photograph
    • Guarantee a dish is available today
  7. Why include restaurant availability or opening hours in recommendations?
    • It improves a model’s parameter count
    • It guarantees the restaurant is highest rated
    • A highly relevant option is not useful if it is unavailable at the requested time
    • It replaces user preferences
  8. What is one limitation of review-mining sentiment for food recommendations?
    • Sentiment scores are always objective
    • Reviews contain exact nutrition facts
    • Review mining removes popularity bias
    • A review’s language may be sparse, subjective, or unrepresentative of the current menu