Keyword extraction is a process of identifying a set of meaningful and representative phrases or word sequences relevant to the examined document. A multilingual keyword extraction algorithm is presented with some distinct features: it requires no extra resources; the time and space complexity are linear in average case; the threshold, the only parameter in this algorithm, is easily tuned; keywords of any length can be identified; and narrower terms (long keywords) can be extracted as well as broader terms. In addition to applying the algorithm to English or other occidental languages in word level, it can be applied in character level.
|頁（從 - 到）||377-378|
|出版狀態||已發佈 - 1998|
|事件||Proceedings of the 1998 21st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR'98) - Melbourne, Vic., Aust|
持續時間: 1998 八月 24 → 1998 八月 28
ASJC Scopus subject areas