Malaysian Lifelogging over Twitter Mohammad Arshi Saloot, Norisma Idris, Rohana Mahmud Faculty of Computer Science and Information Technology University of Malaya Kuala Lumpur, Malaysia Abstract—Twitter can be used to exchange a variety of data such as sensitive data, political news, commercial messages, real-time events, and lifelogging. Lifelogging is the process of recording aspects of one’s everyday life in digital form. This paper presents an empirical study to investigate whether Twittter is used for lifelogging by Malaysian users and to what extent. In this study, ten million Malaysian Twitter messages were gathered via human agents and Twitter’s API. These messages were analyzed to assess the extent of their lifelogging activity according to two main aspects: 1) geolocation of users; and 2) subject of messages. Our results indicate that Malaysians have a high tendency to capture and share their day to day life via Twitter. Keywords-lifelogging; twitter; data gathering; data analyzing I. INTRODUCTION Keeping diary and photo album is an enduring type of lifelogging. Since time immemorial, people keep a personal diary to record their daily activities or feelings. In 1990s, while Word Wide Web (WWW) was gaining popularity, conventional diary transmuted to an online diary. Though, online diaries were categorized into personal and journal. Today personal and journal online diaries exclusively called blog. Each blog writer, namely blogger, has its own purpose of writing blogs. The content of each blog varies from blogger to blogger. However the nuts and bolts of the blog’s content can be day-to-day activities, feelings and emotions, thoughts, social and political considerations, and any content that can be written in conventional paper-based diaries or journals. Microblog is an advanced type of blog. Microblog sites restrict authors to write short texts consisting of a limited number of characters only which is still one of its main features [1], [7]. Nowadays, microblogging sites are equipped with a variety of features such as multimedia content, searching, and geolocation functionality [2]. Moreover, microblogs greatly enriched via utilizing portable consumer electronics such as smart phones or mobile phones. In addition, microblogging sites have borrowed some functionality of Social Network Service (SNS) such as instant messaging, and building social relationship. Web 2.0 is an umbrella term used to lump together all microblog and SNS services, while each of them has their own specific design, functionality and goal. Recent figures show that Twitter, the most popular microblogging website in the world, had over 500 million registered users in 2012 [3]. The leading edge of lifelogging is exploiting wearable computers instead of desktop computers and portable consumer electronic. Steve Man, the first person who used wearable camera for lifelogging, has started to record his life round-the-clock and broadcast it to the Internet [9]. Since then, many researches on lifelogging via wearable computers has been done such as Ubiquitous Memories [10] but still there are divers insoluble challenges that frustrating popularization of passive lifelogging [11]. Therefore, in this study we disregard wearable computer lifelogging and consequently Twitter microblog has been subjected to close scrutiny. A semantic and quantitative study on Twitter’s messages has shown that 50% of Twitter messages are not in the English language. It is also discovered that the Malay language is the fourth most popular language among Twitter users that encompasses 6% of daily Tweets [4]. Thus, enormous usage of Twitter by Malaysian motivates us to take into consideration only Malaysian Twitter messages. Twitter can be used to exchange a variety of data such as sensitive data, political news, commercial messages, real- time events, and lifelogging. Therefore, in this paper we present an empirical study to investigate the purpose of Malaysian to use Twitter. Specifically, the paper answers these research questions: Is Twitter used for lifelogging in Malaysia and to what extent? We answer this question by designing a model, as shown in Fig. 1, to analyze the large amount of Twitter message’s metadata. Figure 1. Observing features of Twitter messages to assess lifelogging Quantitative and qualitative investigation of employed application and devices: Observing the channels of accessing Twitter reveals the lifelogging capacity of Twitter. Twitter is not only a Website but it is a Web service that provides microblogging facilities to its users. This means, users can Hyperlink Application & Device Time Topic Geolocation Subject of Message Lifelogging 2013 First International Conference on Artificial Intelligence, Modelling & Simulation 978-1-4799-3251-1/13 $31.00 © 2013 IEEE DOI 10.1109/AIMS.2013.27 104 2013 First International Conference on Artificial Intelligence, Modelling & Simulation 978-1-4799-3251-1/13 $31.00 © 2013 IEEE DOI 10.1109/AIMS.2013.27 110 2013 First International Conference on Artificial Intelligence, Modelling & Simulation 978-1-4799-3251-1/13 $31.00 © 2013 IEEE DOI 10.1109/AIMS.2013.27 110