Update README.md

README.md

@@ -21,7 +21,7 @@ task_ids:
 ---
 # NewsMTSC dataset
 
-NewsMTSC is a high-quality dataset consisting of more than 11k manually labeled sentences sampled from English news articles. Each sentence was labeled by at least five human coders. The dataset is published as our [EACL 2021 paper *NewsMTSC: (Multi-)Target-dependent Sentiment Classification in News Articles*](https://aclanthology.org/2021.eacl-main.142.pdf).
+NewsMTSC is a high-quality dataset consisting of more than 11k manually labeled sentences sampled from English news articles. Each sentence was labeled by five human coders (the dataset contains only examples where the five coders assessed same or similar sentiment). The dataset is published at [EACL 2021 and titled: *NewsMTSC: (Multi-)Target-dependent Sentiment Classification in News Articles*](https://aclanthology.org/2021.eacl-main.142.pdf).
 
 ## Subsets and splits
 The dataset consists of two subsets (`rw` and `mt`), each consisting of three splits (train, validation, and test). We recommend to use the `rw` subset, which is also the default subset. Both subsets share the same train set, in which the three sentiment classes have similar frequency since we applied class boosting. The two subsets differ in their validation and test sets: `rw` contains validation and test sets that resemble real-world distribution of sentiment in news articles. In contrast, `mt`'s validation and test sets contain only sentences that each have two or more (different) targets, where each target's sentiment was labeled individually.