R & big data analysis 20120531
Download as PPTX, PDF13 likes2,076 views
This document discusses big data analysis processes and tools. It outlines different approaches for analyzing big data including using Rhadoop/Rhipe, Oracle R Enterprise, and RHive to perform analysis with R on Hadoop. It also provides examples of using these tools for cloud log analysis systems and analyzing telecommunications data like call detail records to perform customer segmentation, location-based call quality analysis, social network analysis, and anomaly detection.
![CDR 분석
Data
raw data summary size
(raw 1 yr +sum2 yrs)1 Month 1 Month
Wireless
Unrated CDR
(VOICE, Data, SMS, MMS)
3.7 2.5 104
Rated CDR 1.5 0.2 22
Wi-Fi 0.4 0.3 12
Wibro 1.5 1.0 42
Wireline Rated CDR 1.5 1.5 55
IPDR IP-TV 1.5 0.1 19
Total 10 5.6 254
Unit : TB[ KT CDR(Call Detail Record) ]
[분석 항목]
• 고객 Segmentation
• 위치 기반 통화 품질 분석
• SNA 분석
• Anomaly Detection](https://image.slidesharecdn.com/rbigdataanalysis20120531-150206080154-conversion-gate01/85/R-big-data-analysis-20120531-28-320.jpg)
R & big data analysis 20120531
- 1. Big Data Analysis Process 권정민 (cojette@gmail.com) Open Cloud & Big Data 2012
- 2. Data Science
- 4. There is nothing new under the sun
- 6. (출처: http://embedded.eecs.berkeley.edu/Alumni/wray/data-eng.html) Last updated Mon Jan 27 09:26:36 PST 1997
- 10. Why ?
- 11. muggle wizard
- 12. muggle wizard……
- 13. …… Limits - Environment - Functions - Money Free - Environment - Functions - Money
- 15. Rhadoop / Rhipe Oracle R Enterprise Rhive
- 16. Rhadoop / Rhipe
- 22. RHive
- 25. Big Data Analysis with R
- 26. 클라우드 로그 분석 시스템
- 27. 클라우드 로그 분석 시스템
- 28. CDR 분석 Data raw data summary size (raw 1 yr +sum2 yrs)1 Month 1 Month Wireless Unrated CDR (VOICE, Data, SMS, MMS) 3.7 2.5 104 Rated CDR 1.5 0.2 22 Wi-Fi 0.4 0.3 12 Wibro 1.5 1.0 42 Wireline Rated CDR 1.5 1.5 55 IPDR IP-TV 1.5 0.1 19 Total 10 5.6 254 Unit : TB[ KT CDR(Call Detail Record) ] [분석 항목] • 고객 Segmentation • 위치 기반 통화 품질 분석 • SNA 분석 • Anomaly Detection