本文選題：大數(shù)據(jù)技術(shù)　切入點(diǎn)：分詞技術(shù)　出處：《中國礦業(yè)》2017年09期 　論文類型：期刊論文

【摘要】：地質(zhì)圖書館書籍多,數(shù)據(jù)資料龐大,然而卻存在數(shù)據(jù)資料增長(zhǎng)過快和難以發(fā)現(xiàn)讀者興趣點(diǎn)的問題。實(shí)現(xiàn)高效的圖書館借閱數(shù)據(jù)挖掘分析與推薦,是提高效率的重要手段。為此本文提出了基于大數(shù)據(jù)地質(zhì)文獻(xiàn)分析挖掘平臺(tái),包括聚類分析,中文分詞,推薦系統(tǒng),關(guān)聯(lián)分析功能,再通過Hadoop集群多節(jié)點(diǎn)進(jìn)行推薦,從而提高了工作的效率。
[Abstract]:There are a lot of books and huge data in geological library, but there are some problems such as too fast growth of data and difficulty to find readers' interesting points. In order to realize the efficient analysis and recommendation of library borrowing data mining, It is an important means to improve efficiency. Therefore, this paper puts forward a platform based on big data geological literature analysis and mining platform, including cluster analysis, Chinese word segmentation, recommendation system, association analysis function, and then recommend it through Hadoop cluster multi-node. Thus, the efficiency of the work is improved.
【作者單位】： 中國礦業(yè)大學(xué)(北京);國土資源部地質(zhì)信息技術(shù)重點(diǎn)實(shí)驗(yàn)室;中國地質(zhì)調(diào)查局發(fā)展研究中心;中國地質(zhì)大學(xué)(北京);中國地質(zhì)圖書館;中國科學(xué)院大學(xué);
【基金】：國土資源部公益性行業(yè)科研專項(xiàng)項(xiàng)目資助(編號(hào):201511079)
【分類號(hào)】：G250.7;P5
，

本文編號(hào)：1596390