Dataset for Conflicting Statements Detection in Text

The files are from three different. One of the three datasets (SemEval) is downloaded from SemEval-2014 which was an international workshop on semantic evaluation conducted in Dublin (Ireland). Another dataset is same dataset (Stanford) as used by Marneffe et al. for their work on finding contradictions in text. Another dataset that we use is the PHEME RTE (Recognizing Textual Entailment). The attached dataset consists of annotated dataset into four different types of contradictions. It consists of intermediate results and feature values on our work on conflicting statements detection in text.本次所用文件来自三个不同的数据集。其中一个为SemEval（SemEval）数据集，该数据集下载自SemEval-2014，该项国际语义评测研讨会于爱尔兰都柏林举办。另一个数据集与马尔内夫（Marneffe）等人开展文本矛盾检测研究时所使用的斯坦福（Stanford）数据集完全一致。本次使用的第三个数据集为PHEME RTE（Recognizing Textual Entailment，文本蕴含识别）。随附的标注数据集被划分为四种不同的矛盾类别，涵盖了我们在文本冲突语句检测研究中得到的中间结果与特征值。