We present a comparative study on toxicity detection, focusing on the problem of identifying toxicity types of low prevalence and possibly even unobserved at training time. For this purpose, we train our models on a dataset that contains only a weak type of toxicity, and test whether they are able to generalize to more severe toxicity types. We find that representation learning and ensembling exceed the classification performance of simple classifiers on toxicity detection, while also providing significantly better generalization and robustness. All models benefit from a larger training set size, which even extends to the toxicity types unseen during training.

No abstract

In essence, Sentiment Analysis (SA) is detection and determination of the response of the targeted consumers to a certain brand or product or maybe even a situation. But there are much more potential in SA and emerging research in this area has attracted and involved many brilliant academic minds till date. Human mind is biased with preferences and judgements. So, automated machine to identify and clarify opinions presented in electronic unstructured text has come into picture and became the main focus of present research. There are also some challenges present in this field, such as, accuracy. Further ongoing researches are focusing on solving these problems and creating more efficient tool for SA.This paper tried to cover the study and research on SA that has been done so far. It also talks about the challenges and future aspects of SA.