Social media is an integral part of our lives, yet it posses some problems. Hence, today we discuss detecting a social media profile as real or fake. Here we first, provide the introduction to social media then list the models used. Further, we discuss the results, and finally, we state the conclusion and future work.

Social media platforms are becoming popular day by day among people of all age groups, mostly teenagers. With time memes have gained a significant proportion of content shared on social media platforms. Memes usually have hilarious content but can be offensive some times, containing some hateful message or character image. Such memes may have harmful social impact on the society or maybe on an individual. Thus an automated system for evaluation of offensiveness in the meme content is required.This paper presents an approach to detect offense in memes using Natural Language Processing (NLP) and deep learning. The dataset used for this work consists of 6,992 memes which were labeled as not offensive, slightly offensive, very offensive, and hateful offensive. The model uses very simple architecture with a multi-layer dense network structure involving NLP with RNN and LSTM. We achieved an accuracy of 98% with log loss value 0.130 using the fasttest word embedding and an accuracy of 80% using the Glove word embedding. Therefore this paper suggests using fasttext word embeddings for offense detection work.