運用ELK Stack於電子商務賣方價格分析與資料視覺化

Tunghai University Institutional Repository > 工學院 > 資訊工程學系所 > 碩士論文 > Item 310901/31107

Please use this identifier to cite or link to this item: http://140.128.103.80:8080/handle/310901/31107

Title:	運用ELK Stack於電子商務賣方價格分析與資料視覺化
Other Titles:	Price Analysis and Data Visualization for E-Commerce Seller Using ELK Stack
Authors:	羅意智 LO, I-CHIH
Contributors:	楊朝棟 YANG, CHAO-TUNG 資訊工程學系
Keywords:	爬蟲 ELK
Date:	2018
Issue Date:	2019-01-10T09:05:50Z (UTC)
Abstract:	雲端運算是近年來很熱門的領域，借由硬體效能的提升與軟體應用的密切結合，在產官學各界的帶動下，使得雲端服務越完整也越多元，已經到無所不在的境界了。電腦系統的運作主要為程式邏輯與資料儲存，雲端運算的架構改變了程式設計與執行的模式，也改變了資料存取的檔案系統架構，在Amazon.com 平台上在網絡星期一收到的訂單超過 2800 萬件商品，Ebay.com 平台每年則有 660 萬戶的賣家在使用，如果沒有使用分散式運算與分散式儲存系統，將無法滿足使用者的需求。本研究是利用 ELK Stack 巨量資料分析軟體，透過 Logstash 讀取清洗後的資料並對資料類別進行過濾，輸出至 Elasticsearch 資料庫並建立索引，最終於 Kibana 顯示、分析結果。以 Ebay.com 購物網站為目標，蒐集使用者國籍資料並進行分析，以視覺化的方式呈現結果，讓管理人員能夠輕易解讀大量數據並對未來商務方向進行規劃。另外在簡易的查詢測試當中也發現 Elasticsearch 資料庫比一般的傳統資料庫快上非常的多。 Cloud computing is a very popular area in recent years, by enhancing the performance of the hardware and software applications closely, in industry, gov- ernment and academic communities, making cloud services more complete and the more diverse, has come to the omnipresent realm. Mainly for the operation of the computer system program logic and data storage, cloud computing architecture has changed the programming and execution model, also changed the data access file system architecture. For example, on the Amazon.com more than 28 million items were received on the Cyber Monday, and there is about 6.6 million sellers in the Ebay.com, examples of these cloud services. Without the use of distributed computing and distributed storage system, unable to meet the needs of users. This study uses ELK Stack, a big data analysis software, read and basic filter the data by Logstash, than output to Elasticsearch to made index for the data, finial dis- play on Kibana, let user can read data by visualize. We use Ebay.com as target, to analysis costumer’s nationality information to set managers easily interpret large of data and plan the future business direction.
Appears in Collections:	[資訊工程學系所] 碩士論文

Files in This Item:

File	Description	Size	Format
106THU00394017-001.pdf		5231Kb	Adobe PDF	1615	View/Open

Loading...