Study
with the Nikkei 225 volatility index (VI), which
indicates the views of investors on the likely
size of future uctuations in the Nikkei 225
index. This information could, for example,
become part of the decision-making factors
in the investment strategies of securities
companies, and more ecient investment
may be feasible by capturing how the stock
market is moving.
Value of Big Data Lies in the
Ability to Transcend Barriers
Imposed by Time
To achieve high-speed data analysis, NTT
DATA congured a test system on a server
tted with four processors from the Intel Xeon
processor E7 v2 family, 256 GB memory,
and 64-bit CentOS* 6.4. As Koichi Nabetani,
senior researcher at NTT DATA Mathematical
Systems, explains, “For full-scale testing,
we looked to the high parallel processing
performance of the Intel Xeon processor
E7 v2 family.” Masatoshi Yokogawa, section
manager at NTT DATA Corporation, says, “We
decided to use the Intel Xeon processor E7 v2
family with its large number of cores and high
degree of integration with a view to the future
provision of the index in real time.”
The testing succeeded in performing the
positivity/negativity assessment and
sentiment analysis of 35 months of data in
approximately one day.
Nabetani comments,
“Along with the
improvement in quality
that results from being
able to complete each
analysis quickly, which
allowed for more work
through the plan–
do–check–act (PDCA)
continuous improvement
cycle, the work also
made it clear that our
ability to complete our
planned schedule of
testing several dozen
analysis patterns in
just one month was in
a large part due to the
performance of the Intel Xeon processor
E7 v2 family.” Yokogawa adds, “Our unit
performance results showed that a server
tted with a single processor from the Intel
Xeon processor E7 v2 family can analyze
3,700 tweets per second. This result shows
the potential for real-time analysis. As the
real-time processing capabilities improved,
it was clear that the server could be used in
actual business operations.”
Masahiro Kawagishi of NTT DATA Corporation
says, “We adopted measures to minimize
the total time for parallel processing of
large quantities of data while also seeking to
balance the load on each CPU core. Making
further improvements in parallel processing
performance is a challenge for the future.”
Referring to the business aspects of the
testing, Nakagawa emphasizes, “In business,
speed is everything. The value of big data
lies not just in the large quantities of data but
also in transcending the barriers imposed
by time—for example, being ahead of other
people in learning something that would have
taken several days to nd out in the past.”
Potential Applications Include
Assessing Product Reputations
and Detecting Complaints
NTT DATA has received many inquiries
about the Twitter sentiment index since
Fast parallel processing capabilities of the Intel® Xeon®
processor E7 v2 family help achieve high-speed
analysis of a complete Twitter* data set
Technique used in sentiment analysis to identify positive and negative tweets
Intel® Xeon® Processor E7 v2 Family
All tweets
(several tens of
billions)
Market-related tweets
(several hundred
million)
Market sentiment
(about 100 million)
Extract Sentiment analysis
~NTT DATA~ concerning graduate recruitment
~NT T DATA~ concerning graduate recruitment
KDDI is up, but it seems too good to be true
KDDI is up, but it seems too good to be true
KDDI is up, but it seems too good to be true
Abenomics seems to be pushing the Nikkei higher recently
Abenomics seems to be pushing the Nikkei higher recently
Abenomics seems to be pushing the Nikkei higher recently
Toyosu NowBALS!
Twitter index
its release was announced in March 2014,
particularly from securities companies.
Nabetani comments, “We plan to develop
additional Twitter sentiment indices for
industries such as retailers or other B2C
businesses. We have already started looking
into applications outside the nancial
industry such as reputation analysis and
detecting complaints.”
As it goes about creating new services and
applying advanced analytical techniques to
large amounts of data, NTT DATA has high
hopes for the Intel Xeon processor E7 v2
family. Nakagawa describes the outlook for
the future by saying, “Hardware and analysis
algorithms are inseparable technologies for
business analytics. As hardware advances,
analyses that were unthinkable 10 years
ago come into reach. We intend to continue
working with Intel to supply services with high
added value to our customers.”
Find the solution that’s right for your
organization. Contact your Intel representative,
visit Intel’s Business Success Stories for IT
Managers, or explore the Intel.com IT Center.
Keiichirō Nakagawa (front left)
Director (Doctor of Engineering)
NTT DATA Mathematical Systems Inc.
Masatoshi Yokogawa (front right)
Section Manager
NTT DATA Corporation
Koichi Nabetani (rear left)
Senior Researcher
NTT DATA Mathematical Systems Inc.
Masahiro Kawagishi (rear right)
NTT DATA Corporation
All performance tests were performed and are being reported by NTT Data Corporation. Please contact NTT Data for more information on any performance test reported here.
This paper is for informational purposes only. THIS DOCUMENT IS PROVIDED “AS IS” WITH NO WARRANTIES WHATSOEVER, INCLUDING ANY WARRANTY OF MERCHANTABILITY, NONINFRINGEMENT, FITNESS FOR
ANY PARTICULAR PURPOSE, OR ANY WARRANTY OTHERWISE ARISING OUT OF ANY PROPOSAL, SPECIFICATION, OR SAMPLE. Intel disclaims all liability, including liability for infringement of any proprietary rights,
relating to use of information in this specication. No license, express or implied, by estoppel or otherwise, to any intellectual property rights is granted herein.
Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using speci c
computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully
evaluating your contemplated purchases, including the performance of that product when combined with other products.
Copyright © 2014 Intel Corporation. All rights reserved. Intel, the Intel logo, Look Inside., the Look Inside. logo, and Xeon are trademarks of Intel Corporation in the U.S. and other countries.
* Other names and brands may be claimed as the property of others. JPN/1407/PDF/SE/ESS/SO 330536-001US