結合情境資訊與適地性服務之餐廳推薦

透過智慧型手機可以隨時隨地上網、收發電子郵件或購物，使生活更加便利，同時，情境感知應用藉由智慧型手機輔助之下，發展更無可限量；手機內建許多感測器，可以用來取得使用者周遭情境資料，加以過濾之後，可以更準確掌握使用者情境資訊，視需求應用在不同作用上，情境資訊也可以改善推薦系統資料稀疏性問題。利用手機擷取使用者地理座標，演算Place of Interests，也就是使用者經常出現之地理區域，給予新使用者餐廳推薦之參考，可解決推薦的冷啟動問題。本研究結合情境感知、適地性服務及推薦系統等概念，藉由手機端取得使用者餐廳評分及情境資訊，包含地理座標、時間、天氣、速度以及方向等；伺服器端負責使用者之Place of Interests運算：找出潛在地緣關係、推薦模組運算：結合內容導向與協同過濾推薦方法以及情境分類，使餐廳推薦結果符合使用者需求時之情境。最終，在本研究推薦機制下，根據時間、天氣、速度、方向等情境修正，選擇Cosine 60度以下之相似使用者，預測誤差值約為0.5，在100位使用者實際使用後，給予本餐廳推薦App正面評價與建議，推薦結果符合使用者需求產生時之情境。

關鍵字

情境感知；推薦系統；適地性服務； Place of Interests

並列摘要

Due to the increasing of smartphone users, it becomes more convenient for people to surf on the net. In the meanwhile, more conveniences are found by combining the function of context awareness and the sensors which are equipped in the mobile phones. The sensors can catch users’ context data. After filtering and computing, it shows the users’ their contextual information. Context awareness can not only extract the context data, but also adapt to users’ using environment. In addition, since recommendation systems must face the increasing amounts of data, through the recommendation mechanism, we can find out the needed information easier. The study stands on three concepts, context awareness, location-based service, and recommendation system. A mobile App of restaurant recommendation first collects users’ recommending requests, users’ context (users’ coordinates, time, weather, speed and directions) and their ratings of restaurants. Then, the server, therefore, contains the collection of users’ Place-of-Interests, recommending items and the contextual classification. As result, this system can assist users by providing their personalized Place-of-Interests and current context. Also, it finally enhances the result of the restaurant recommendation.

並列關鍵字

Context-Awareness ； Recommendation Systems ； Location-based Services ； Place of Interests

參考文獻

1.李佳馨，整合內容導向式方法與混合式協同過濾之電影推薦，碩士論文，國立臺北科技大學資訊與運籌管理研究所，臺北，2013。

2.Abbas, M. B., Prabuwono, A. S., Abdullah, S. N. H. S., & Hussain, R. I. “Automated Visual Inspection for Surgical Instruments Based on Spatial Feature and K-Nearest Neighbor Algorithm,” Advanced Science Letters, Volume 20, Issue 1, 2014, pp.153-157.

3.Adomavicius, G., & Tuzhilin, E. “Toward the next generation of recommender systems：A survey of the state-of-the-art and possible extensions,” IEEE Transactions on Knowledge and Data Engineering, Volume 17, Issue 6, 2005, pp.734-749.

4.Adomavicius, G., & Tuzhilin, A. “Context-aware recommender systems,” In Recommender systems handbook: Springer, 2011, pp.217-253.

6.Balabanovic, M., & Shoham, Y. “Fab: content-based, collaborative recommendation,” Communications of the ACM, Volume 40, Issue 3, 1997, pp.66-72.

國際替代計量

結合情境資訊與適地性服務之餐廳推薦

全文下載

主題瀏覽