A corpus of Chinese abbreviation

我们建立了一个由短语和术语组成的数据集。此数据集中有10,786全形式，包括8,015正全形式和2,661负全形式。短语包含名词短语，动词短语，组织名称，位置名称等。分布如表2所示。对于实验，我们随机抽取7,551个样本作为训练集，1078个样本作为开发集，2,157个样本作为测试集。我们计算数据中的单词和字符 (包括重复项) 的数量。

We constructed a dataset composed of phrases and terms. It contains 10,786 full forms in total, including 8,015 positive full forms and 2,661 negative full forms. The phrases cover noun phrases, verb phrases, organization names, location names, and so on. Its distribution is presented in Table 2. For the experiment, we randomly sampled 7,551 samples as the training set, 1,078 samples as the development set, and 2,157 samples as the test set. We calculated the counts of words and characters (including duplicates) within the dataset.

搜集汇总

背景与挑战

背景概述

该数据集是一个中文缩略语语料库，包含10,786个全形式短语，涵盖名词短语、动词短语及组织名称等多种类型，并划分为训练集、开发集和测试集用于实验。它由北京大学于2017年发布，旨在支持中文缩略语的相关研究。