Classifying and Extracting Data from Facebook Posts for Online Persona Identification

Classifying and Extracting Data from Facebook Posts for Online Persona Identification Hazel Anne Brosas De La Salle University Manila, Philippines hazel brosas@dlsu.edu.ph Eugene Lim De La Salle University Manila, Philippines eugene lim@dlsu.edu.ph Danica Sevilla De La Salle University Manila, Philippines danica sevilla@dlsu.edu.ph Denise Anne Silva De La Salle University Manila, Philippines denise silva@dlsu.edu.ph Ethel Ong De La Salle University Manila, Philippines ethel.ong@delasalle.ph Abstract Large amount of user-generated data are posted online in social media platforms, in- cluding user preferences, dining and leisure activities, events, news and personal blogs. This resulted in varying efforts to process so- cial media data using NLP and ML algorithms for topic classification, sentiment analysis and detection, and events classification. Such in- formation are problematic to process, as they tend to be short, informal, inconsistent, and are highly contextualized. A series of tasks is involved from collecting, pre-processing, clas- sification and extraction before social media data can be used. In this study, we built a multi-class classifier model to process Face- book posts in order to identify a user’s online persona based on his/her preferences. Infor- mation extraction is then applied to find rele- vant data from the classified posts that can be used to generate a description of the user’s on- line persona. The classifier currently achieves an accuracy of 76.02% and an F1 score of 73.10% using 10-fold cross validation from a dataset containing 16,682 posts. 1 Introduction Facebook is the primary online social networking platform with an average of 1.45 billion daily active users worldwide. It has catered to the communica- tion and socialization of its users to achieve varying purposes, including personal, academics (Kramer, 2015; Prescott et al., 2015), and business transac- tions (Hutchings, 2012; Culnan et al., 2010). By building a public profile, users combine text, im- ages and video to share numerous kinds of con- tent ranging from personal stories and activities to events, news, memes, blog posts, and business- oriented posts, to reach a large audience base. People’s online activities, reflected through their participation, motivation and practices, can be used to construct their online persona (Queensberry, 2015). An online persona is defined to be the so- cial identity that an Internet user establishes in on- line communities and websites (Carminati et al., 2013). The study of Zhao, Grasmuck and Jason Martin (2008) revealed three (3) modes of online identity construction. These are visual with the use of display pictures, photos, and wall posts; enumer- ating hobbies, interests, and favorites; and describ- ing oneself through narratives. While the study of Soraj (2011) argues that the persona people exhibit in social media platforms may not necessarily re- flect their true identity, it is still worth exploring how automatic Facebook data classification accord- ing to user preferences can be used in online persona identification. This is essential in various fields, e.g. business, literature, and computing technology, wherein an understanding of the characteristics of users and/or consumers can aid in personalizing ser- vices and presentation of relevant information. PACLIC 32 63 32nd Pacific …

Content truncated to 500 words