基於行動社交網路建立機率模型做時間相關之聯絡人推薦

Abstract

本論文提出了一種新穎的框架對使用者在一個特定的時間點從過去的歷史通信紀錄中做時間相關之聯絡人推薦。我們找出一些基本因素來處理互動資訊以及自動形成時間相關之聯絡人群組以配合一些場景，例如說清晨的時候我應該撥打電話給誰？在午夜的時候誰的Email 我應該最先回？我們建立了一個機率模型，不僅可以捕捉到使用者與候選聯絡人之間的時間相依傾向，還揉合了聯絡頻率及新近程度到聯絡人群組中。我們還利用此機率模型來支持兩種型態的時間相關之聯絡人推薦： Seedset Generation：單一互動推薦 以及 Friends Suggestion：多人互動推薦。我們在三種實際資料集上進行實驗，結果顯示，使用我們提出的機率模型可以有效地做時間相關之聯絡人推薦。

This paper presents a novel framework for time-dependent contacts recommendation for a query user at a given time point from historical communication logs. We identify the fundamental factors that govern interactions and aim to automatically form time-dependent contact groups for scenarios, such as, who should I dial to in the early morning? whose mail would I reply first at midnight? We develop a probabilistic model that not only captures temporal tendencies between the query user and each contacts candidate but also blends frequency and recency into group formation. We also utilize the model to support two types of time-dependent contacts recommendation: Seedset Generation: singleinteraction suggestion and Friends Suggestion: multiple interactions suggestion. Experimental results on Enron dataset, Call Detail Records and Reality Mining Data from MIT prove the effectiveness of time-dependent contacts recommendation with proposed probabilistic model.