基於臉書行為來預測臉書使用者空間類型

近年來，越來越多的社群網站如春筍般的出現，社群網站開始深入了我們現實生活的每部分，很多人把跟朋友的互動從現實逐漸轉向了這個虛擬世界，社群網站也提供了一個管道讓我們能去認識新朋友，甚至我們還可以在地方或國際發生大事的時候，第一時間從社群網站的討論上得知相關資訊，而不再需要靠電視新聞或者是報紙。而在所有社群網站中，無論從普及率或者是使用者個數來看，臉書都是目前最大的社群網站，基於這麼多的使用者使用臉書，臉書上充滿了很多使用者的行為資訊，這吸引了很多專家學長開始研究使用者在臉書上的行為。而在環境心理學的角度中，大多都是去研究一個實體空間跟人類之間的交互關係，而鮮少從虛擬空間的角度去探討，但隨著社群網路的發展，人類把很多在現實生活中的行為轉到了這個虛擬空間，那對於使用者來說，臉書算是個怎樣種類的空間呢？這個空間對使用者的意義是什麼呢？為了了解這個問題的答案，我們提出了一個模型去分析使用者在臉書上的行為，根據這些行為去預測使用者覺得臉書是個怎樣的空間類型，在把預測出來的結果去跟問卷做比較，最後結果顯示我們擁有還不錯的準確率。

關鍵字

社群網站；臉書；空間類型；分群

並列摘要

As the rapid development of social networks services (SNSs), SNSs like Face-book are getting more popular and have played a critical role in our daily lives. This is because we are able to interact with our friends or acquire latest information (or news) on Facebook in anytime at anyplace. That is, such new SNSs have changed the way of how to communicate or connect our friends and how to learn new knowledge or information. Among all social network sites, Facebook is currently the biggest social networking service based on global reach and total active users. For the large number of users, Facebook contains a lot of user behavior data. It attracts many researchers to start studying user behavior in Facebook. Hence, the focus of this work will study how people behavior within Facebook from the perspective of environmental psychology. We believe that learning user’s behavior within Facebook and what they feel like will be a key to successfully design better functionalities for Facebook. To deal with this issue, we design a user behavior model by crawling user data in Facebook to predict user’s space type. Results show that our approach is able to do well on prediction accuracy of user’s space type.

並列關鍵字

Social networks ； Facebook ； Space pattern ； Clustering

參考文獻

[1] Facebook Reports First Quarter 2015 Results [online]: http://newsroom.fb.com/content/default.aspx?NewsAreaId=22

[5] Bazarova, Natalya N., et al. "Social sharing of emotions on Facebook: Channel differences, satisfaction, and replies." Proceedings of the 18th ACM Conference on Computer Supported Cooperative Work & Social Computing. ACM, 2015.

[7] Po-Yu Chen (2014) Virtual geography of Facebook: analyzing pattern of Facebook usage based on space syntax, National Taiwan University.

[10] Frey, Brendan J., and Delbert Dueck. "Clustering by passing messages between data points." science 315.5814 (2007): 972-976.

[2] The Top 20 Valuable Facebook Statistics [online]: https://zephoria.com/top-15-valuable-facebook-statistics/

Google Scholar

國際替代計量

基於臉書行為來預測臉書使用者空間類型

全文下載

主題瀏覽