| Summary: | 碩士 === 國立臺北科技大學 === 資訊與運籌管理研究所 === 101 === Nowadays, everyone uses smartphones to watch films on the Youtube, to login Facebook to Check-in, etc. We also use applications to facilitate our life. Every time we discuss what to eat with our friends or families, we will have troubles to decide which restaurants to eat in. There are hundreds of applications about restaurant recommendation in the real world, but no one provides personalized recommendation. This study constructs a mobile application using persons and their friends’ Facebook Check-in data to recommend more satisfied restaurants.
We first collect users’ data on the Facebook then calculate the similarity of users’ friends by using Cosine similarity measure. After we find K similar friends of the user, the K friends’ Check-in restaurants will be in the recommendation list. Second, users can type in their conditions about price, period as situation parameters. We use the positioning system to find a list of restaurants around users. Our combined recommendation lists include similar recommendation lists and location-based recommendation lists. Then we use our weight formula to order and recommend the combined list to users.
|
|---|
