植基於網路爬蟲和問與答的企業知識樹

文化大學機構典藏 CCUR > 商學院 > 資訊管理學系暨資訊管理研究所 > 博碩士論文 > Item 987654321/51126

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://irlib.pccu.edu.tw/handle/987654321/51126

题名:	植基於網路爬蟲和問與答的企業知識樹 Enterprise knowledge tree based on web crawler and FAQs
作者:	陳杰
贡献者:	資訊管理學系
关键词:	知識樹問答推薦知識管理關鍵詞提取
日期:	2021
上传时间:	2023-02-22 13:51:00 (UTC+8)
摘要:	在各種各樣電子資訊爆發的今日，用戶在接受各種各樣新產品的同時會產生大量的問題，因為無法第一時間獲得解答，各大問答社區成為了這些問題以及回復的載體，為了讓越來越多的問答資料能夠更好的發揮價值，讓問答與問答之間的關聯能力更強，有更好延展性和更高的性能。本研究提出了將問答知識庫建制成問答知識樹的方法，先爬取華為 5G 手機的問與答作為範例背景，再進行數據清理留下純文本數據，中文分詞，詞頻統計，關鍵字提取等中文文本處理，最後分別通過由下往上建樹和由上往下建樹兩種方式進行建造專門領域的知識樹，研究出的知識樹用更少的計算量讓節點與節點之間有著較為緊密的聯繫，可以很好地運用在問答的有序管理以及後續的相關問答推薦領域。 With the explosion of all kinds of electronic information today, users are faced with a variety of new products as well as a lot of questions about them. Because they often fail to get answers in the first time, the various Q&A communities have become the carriers of these questions and answers. In order to make more Q&A information be valuable and have better extensibility and higher performance, and to make the questions and answers more relevant, this study proposes a method to build a Q&A knowledge base into a Q&A knowledge tree. Firstly, the Q&A about Huawei 5G mobile phones were crawled as an example background. Then, the data was cleaned to leave the plain text data, and Chinese text processing such Chinese word segmentation, word frequency statistics, and keyword extraction were conducted. Finally, domain-specific knowledge trees were built respectively from bottom to up and from top to down. The designed knowledge tree can make the relationship between nodes and nodes closer with less computation amount, so it can be widely applied in fields such as management of the questions and answers and subsequent relevant question and answer recommendation field.
显示于类别:	[資訊管理學系暨資訊管理研究所 ] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	221	检视/开启

检视Licence

在CCUR中所有的数据项都受到原著作权保护.

数据加载中.....