Oliver Theobald

“Perhaps the most visionary of emerging recommender systems is Google’s patented environment-based recommender system. The tech behemoth has patented "advertising based on environmental conditions," which draws on environmental factors such as temperature and humidity collected through device sensors. In addition to climatic factors, the technology is said to gather light, sound, and air composition and translates this information into criteria for what ads to serve users.”
― Oliver Theobald, Machine Learning: Make Your Own Recommender System

“The meaning of a 3-star review, for example, can be interpreted differently among two users based on their average rating history. Standards also vary among countries and types of users (i.e. e-book readers versus physical book readers). Readers of physical books, for example, rate negatively when then are delivery delays/complications or production quality (i.e. paper quality) which doesn’t affect e-book readers who receive a digital copy on-demand.”
― Oliver Theobald, Machine Learning: Make Your Own Recommender System