In this paper, we propose LexSub, a novel approach towards unifying lexical and distributional semantics. We inject knowledge about lexical-semantic relations into distributional word embeddings by defining subspaces of the distributional vector space in which a lexical relation should hold. Our framework can handle symmetric attract and repel relations (e.g., synonymy and antonymy, respectively), as well as asymmetric relations (e.g., hypernymy and meronomy). In a suite of intrinsic benchmarks, we show that our model outperforms previous approaches on relatedness tasks and on hypernymy classification and detection, while being competitive on word similarity tasks. It also outperforms previous systems on extrinsic classification tasks that benefit from exploiting lexical relational cues. We perform a series of analyses to understand the behaviors of our model. 1 Code available at https://github.com/aishikchakraborty/LexSub .

The word ordering in a Sanskrit verse is often not aligned with its corresponding prose order.Conversion of the verse to its corresponding prose helps in better comprehension of the construction.Owing to the resource constraints, we formulate this task as a word ordering (linearisation) task.In doing so, we completely ignore the word arrangement at the verse side.kāvya guru, the approach we propose, essentially consists of a pipeline of two pretraining steps followed by a seq2seq model.The first pretraining step learns task specific token embeddings from pretrained embeddings.In the next step, we generate multiple hypotheses for possible word arrangements of the input (Wang et al., 2018).We then use them as inputs to a neural seq2seq model for the final prediction.We empirically show that the hypotheses generated by our pretraining step result in predictions that consistently outperform predictions based on the original order in the verse.Overall, kāvya guru outperforms current state of the art models in linearisation for the poetry to prose conversion task in Sanskrit.

In Twitter, there is a rising trend in abusive behavior which often leads to incivility. This trend is affecting users mentally and as a result they tend to leave Twitter and other such social networking sites thus depleting the active user base. In this paper, we study factors associated with incivility. We observe that the act of incivility is highly correlated with the opinion differences between the account holder (i.e., the user writing the incivil tweet) and the target (i.e., the user for whom the incivil tweet is meant for or targeted), toward a named entity. We introduce a character level CNN model and incorporate the entity-specific sentiment information for efficient incivility detection which significantly outperforms multiple baseline methods achieving an impressive accuracy of 93.3% (4.9% improvement over the best baseline). In a post-hoc analysis, we also study the behavioral aspects of the targets and account holders and try to understand the reasons behind the incivility incidents. Interestingly, we observe that there are strong signals of repetitions in incivil behavior. In particular, we find that there are a significant fraction of account holders who act as repeat offenders - attacking the targets even more than 10 times. Similarly, there are also targets who get targeted multiple times. In general, the targets are found to have higher reputation scores than the account holders.