@inproceedings{ed326230ed1148d9802c6c6f1f374390,
title = "A Chinese new word detection approach based on independence testing",
abstract = "New word detection is of great significance for Chinese text information processing, which directly affects the capabilities of word segmentation, information retrieval and automatic translation. Focusing on the problem of Chinese new word detection, this paper proposes an independence-testing-based detection approach with no need of prior information. The paper analyzes statistical characteristics of new words in Chinese texts, uses statistical hypothesis testing to infer the correlations between adjacent semantic units, and proposes an iterative algorithm to detect new words gradually. Our algorithm is evaluated on both large-scale corpus and short news texts. Experimental results show that this approach can effectively detect new words from all kinds of news.",
keywords = "Hypothesis testing, New word detection, Semantic unit, Test of independence",
author = "Dongchen Jiang and Xiaoyu Chen and Xin Yang",
note = "Publisher Copyright: {\textcopyright} 2018, Springer Nature Switzerland AG.; 13th International Conference on Artificial Intelligence and Symbolic Computation, AISC 2018 ; Conference date: 16-09-2018 Through 19-09-2018",
year = "2018",
doi = "10.1007/978-3-319-99957-9\_17",
language = "英语",
isbn = "9783319999562",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "227--236",
editor = "Dongming Wang and Jacques Fleuriot and Jacques Calmet",
booktitle = "Artificial Intelligence and Symbolic Computation - 13th International Conference, AISC 2018, Proceedings",
address = "德国",
}