Identifying the factors affecting internet memes to become viral on social media / Elly Atiqah Md Nawawi, Muhammad Faiz Izzuddin Hamdan and Nur Atiqah Mohd Zulkafli

Memes is a post that social network user usually post in social network to convey messages, feeling or opinion in sarcastic ways. Unfortunately, most of the social network users failed to use memes in a beneficial way such as for marketing and advertising. The tendency of people using memes nowadays should be exploited by the people in good and more useful ways. The aim of the study is to analyze the factors contributing to the largest number of internet memes shares or its virality. Besides that, the study also wants to find the best method that is suitable to predict number of shares of a meme. The data of the study is obtained from a famous Malaysian Facebook named 'Gags Malaysia'. Negative Binomial Regression (NBR) and Non-Linear Regression (NLR) method was used to find the factors that affects the Number of S.hares. NLR is used because the data was found not linear during the model adequacy checking process. While NBR was applied because there was over¬ dispersion issue exist when using Poisson Regression. R-square value was used to decide the best method to predict number of shares. Next, to find whether there is a relationship between the categorical variables, Chi-Square Test of Independence is applied. The result shown that NBR is chosen as the best method because the R¬ square value of NBR (1.000) is the perfect fit compared to NLR (0.1683). In the NBR analysis indicates that all variables are significant except for the main effect of Types of Memes and Self-enhancing, the main effect of Day of Posting and Weekdays and the Number of Comments. Lastly, The Chi-Square Test of Independence indicates that there was an association between Types of Memes and Day of Posting.