Visualizing People's Emotions in Facebook N.AZMINA M. ZAMANI 1 , SITI Z. Z. ABIDIN 2,3 , NASIROH OMAR 3 , SHARIFAH ALIMAN 3 1 Faculty of Computer and Mathematical Sciences, UniversitiTeknologi MARA, Perak Branch, Tapah Campus, 35400 Tapah Road, Perak, MALAYSIA 2 Advanced Analytics Engineering Centre (AAEC), 3 Faculty of Computer and Mathematical Sciences,UniversitiTeknologi MARA, 40450 Shah Alam, Selangor, MALAYSIA namz.ina@gmail.com, {zaleha, nasiroh, sharifahali}@tmsk.uitm.edu.my Abstract Facebook is the most common platform where people can comment, give opinions and convey their feelings by using texts that normally are embedded with emoticons. The contents of every post and comment shared by the users contain various types of issues.It becomes complicated to observe these large amounts of data manually. Hence, this paper aims to visualize the results obtained which are the people's emotions based on the texts used by Facebook users in the posts and comments. The objectives are to quantify texts in English and Malay to obtain meaningful words, visualize data based on the obtained words and evaluate the results. Two methods are applied via a combination of both point-based and region- based visualization techniques; and pixel-oriented technique. Two case studies are performed on mobile phones Facebook pages(Nokia, Samsung and Sony) and UniversitiTeknologi MARA (UiTM) bursary Facebook page. The first case study uses Murvis tool which produce 2D and 3D views involving the point-based and region- based visualization techniques, whereas the pixel-oriented visualization technique is applied for the second case study. Murvis is an improvised tool that allows user to view analyzed results in both 2D and 3D views which gives an advantage to view the results from different perspectives. The pixel-based approach is helpful to view the opinion of each user by each sentences used by them. The findings obtained from the first case study is that Nokia phones has the most issues being mentioned compared to the other two phones based on point-based technique. This method gives an overall view of all three mobile phones issues whereas from the region-based method, each mobile phone has their own feature issues. The result has shown that Samsung battery feature has been mentioned the most compared to its other features. This is beneficial for the businessmen to analyze the issues faced by their customers regarding their products. Thus, they can improve their product and services from the findings. The second case study shows that most of the users are not quite happy with certain issues from the UiTM bursary page. The administrator should focus more on those users' sentence containing red-coloured pixels as it represents unhappy words.More combination of techniques can be used to visualize data for future work. Keywords:Visualization, Social Network, Quantifying sentiments, Facebook, Emotion analysis. 1 Introduction International Journal of Pure and Applied Mathematics Volume 118 No. 9 2018, 183-193 ISSN: 1311-8080 (printed version); ISSN: 1314-3395 (on-line version) url: http://www.ijpam.eu Special Issue ijpam.eu 183