As the internet users in the Philippines have become increasingly rampant, Tagalog Sentiment Analysis is essential when determining general sentiments correlated to a topic. Among Tagalog tweets, backward slang has gained increased popularity on social media. Deep learning algorithms such as Long-Short Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM) have been proven more effective than traditional machine learning algorithms. This study explored and evaluated the effects of backward slang on two deep learning models using performance metrics for classification models. Results show that backward slang helps in the expressivity of textual data, given the improvement in overall sentiment score