陳漢明臺灣大學：機械工程學研究所汪修斌Wang, Siou-BinSiou-BinWang2007-11-282018-06-282007-11-282018-06-282007http://ntur.lib.ntu.edu.tw//handle/246246/61549這篇論文是在描述一個位於使用者與專利搜尋引擎之間的中介系統。當線上專利資料庫將數個網頁作為搜尋結果送出之後，本系統從那數個網頁中摘出這次搜尋結果的特徵值並將之視覺化。一篇專利文章是由標題、專利所有權人名稱或是摘要等等許多字段(field)組成的。而多個專利的相同字段位置出現一樣的描述時，比如說專利所有權人名稱相同時，這許多個專利可以歸在同一類。專利引證分析是許多專利分析軟體的關鍵技術，本文中也使用專利引證資料來計算不同專利之間的關聯度。本系統提供了一個樹狀架構的介面，來視覺化單篇專利文件及其專利引證資料。另外針對索引式搜尋結果裡的多個專利，使用力導向定位演算法(force-directed placement algorithm)來視覺化，而依照使用者選擇字段位置的不同，呈現出不同的動態介面。隨著使用者的操作，介面中所有代表專利的點會移動到其適當的位置，並且形成不同的專利引證資料網狀圖。使用者可以利用移動跟放大縮小視野，來觀察感興趣的區域。此外，使用者也可以選擇是否顯示字段標籤或專利引證資料間的連線。在文章最後將討論有哪些可行的改進，以求更好的顯示搜尋結果。This paper presents an intermediate system between users and a patent search engine. The online patent database provides Web pages as search results, and the system extracts parameters from those pages to visualize the search results. Patents are divided into many fields, such as title and assignee name, and the patents with the same value of a field can be classified as a group. Citation analysis is a key technique in many applications of patent analysis, and we use citation to identified relationships between the patents. The system provides a tree view to visualize a patent and citations of the patent, and then force-directed placement algorithm visualizes the indexing search results of a collection of patents with a dynamic view for the different fields. Depending on interaction of the users and the interface, the visualization moves patent dots to shape different field citation network. The users can navigate the view by zooming into or out of regions of interest. Optionally, field labels and citation links maybe shown by the system. Finally, potential improvements identified during the study are discussed, as are future directions for this approach to collection browsing.Table of Contents 口試委員會審定書………………………………………….……………. i 摘要………………………………………….………………….………… ii Abstract……………………………….………………….…………….… iii Chapter 1 Introduction……………………………………………………. 1 Chapter 2 Background………………………………………………….…. 4 2.1 Related Work………………………………………………….…. 4 2.1.1 Patent Analyses……………………………………………. 4 2.1.2 Literature Maps……………………………………….…… 4 2.1.3 Visualization Tools………………………………………… 5 2.2 Design Guides………………………………………………….… 6 2.2.1 Basic Steps………………………………………………… 6 2.2.2 Data Acquisition…………………………………………… 7 2.2.3 Patent Citation Analysis……………………………….…... 8 2.2.4 Visualization Construction………………………………... 10 Chapter 3 Implementation…………………………………………………………... 14 3.1 Our Approach………………………………………………....…. 14 3.1.1 Overview…………………………………………..…..….. 14 3.1.2 Web Browser…………………………………………..…. 16 3.1.3 Citation Tree View……………………………………...… 17 3.1.4 Scatter Plot…………………………………………….…. 18 3.2 Details of the Scatter Plot………………………………………. 19 3.2.1 MDS and Spring Model…………………………………. 19 3.2.2 Measure of Citation Similarity…………………………... 22 3.2.3 Converting Similarities to Distances…………………….. 25 3.2.4 Chalmers’ 1996 Algorithm………………………………. 26 Chapter 4 Demonstration……………………………………………..…. 29 4.1 Instance for Citation Tree View………………………………… 29 4.2 Instance for Scatter Plot………………………………………… 32 Chapter 5 Discussion……………………………………………….…… 42 5.1 Meanings of Scatter Plot……………………………………….. 42 5.2 Data Transfer and Storage………………………………………. 43 Chapter 6 Conclusion…………………………………………………… 44 References………………………………………………………………. 451308688 bytesapplication/pdfen-US資訊視覺化力導向定位演算法引證資料分析搜尋引擎專利分析information visualizationforce-directed placement algorithmcitation analysissearch enginepatent analysisthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/61549/1/ntu-96-R92522615-1.pdf