SlideShare a Scribd company logo
1 of 26
SAP InfiniteInsight:SAPが提供する機械学習エンジン
SAPジャパン株式会社
プラットフォーム事業本部
岩渕 聖
2015/3/29
© 2015 SAP SE. All rights reserved. 1
2© 2015 SAP SE or an SAP affiliate company. All rights reserved.
機械学習で何ができるの?(代表例)
Classification:分類
Who will (buy | fraud | churn …) next (week | month | year…) ?
Prediction:発生確率
How will the (revenue | # churners) be next (week | month..)?
Segmentation:セグメンテーション
What are the groups of customers with similar (behavior | profile …)?
Social Network Analysis: インフルエンサー分析
Analyze interactions to identify (communities | influencers…)
Association Rules:連関性分析
Analyze transactions to identify events likely to occur together
Forecasting:予測値計算
How will the (revenue | # churners…) be over next year on a monthly basis ?
3© 2013 SAP AG or an SAP affiliate company.All rights reserved.
本日の目的①
アルゴリズム云々のアカデミックな内容は
極力少なくし、人(データサイエンティストの方)の
苦労を如何に自動化するか。
その壁を越えた際に見えてくる世界。
そこにフォーカスして話をさせて頂きます。
4© 2013 SAP AG or an SAP affiliate company.All rights reserved.
本日の目的②
実際にこの時間の中で
“解析“~”システムへの適用”
まで実演します。(できます。)
5© 2015 SAP SE or an SAP affiliate company. All rights reserved.
ビジネス
バリュー
ビジネスチャンス
期間
試行錯誤を行うデータ解析の課題
通常の解析モデル
N日
時間 Nヶ月
データ加工
モデル構築
アクション
実行
90点
意思決定のニーズ
/依頼
6© 2015 SAP SE or an SAP affiliate company. All rights reserved.
ビジネス
バリュー
こんな世界を目指します。
n日 Nか月
時間
意思決定の
ニーズ
たとえばこのケースの
場合どちらを選択しますか?
アクション
実行
88点
ビジネスチャンス
期間 アクション
実行
90点
7© 2015 SAP SE or an SAP affiliate company. All rights reserved.
さらにBigdata化(縦の空間、横の空間)
コア事業
クレジット
カード
多ブランド
運営
プライベート
ブランド
アフター
サービス
ポイント
プログラム
テナント
運営
自社チャネル利用
明細(POS)
Web広告への反応
コールセンターなどの
接点ログ
カード利用履歴
契約情報
自社サイト利用履歴
Social Media
事業の多様化
顧客行動データの
収集
8© 2015 SAP SE or an SAP affiliate company. All rights reserved.
やれることはたくさんある
自社チャネル利用
明細(POS)
Web広告への反応
コールセンターなどの
接点ログ
カード利用履歴
契約情報
自社サイト利用履歴
Social Media
商品への
興味
家族構成
就労環境
ライフ
イベント
生活習慣
データ 情報
カード会員向け施策
自社小売チャネルへの
フィードバック
オムニチャネルの実現
加盟店向け
情報提供
施策
“モノ”から“コト”へのシフト
9© 2015 SAP SE or an SAP affiliate company. All rights reserved.
2000年代前半までのデータマイニングブーム
 モデルの精度がツールではなく、分析者の力量に依存する
 分析者によって嗜好が異なり、担当者が変わると再モデル作成
 モデルの開発に非常に長い時間を費やす
 データサイエンティストだけでなく、プログラミングスキル、ビジネスノウハウを持った“チーム“で取り
掛かる必要がある
データマイニングに必要なリソース
データマイニングアプローチ
サンプリング データ加工 変数選択
回帰分析
データ分割
決定木
ニューラル
MBR
アセスメント
・・・more
モデル合成
サンプリング
Data
対象となる分析シナリオを
すべて実装できず一部のみが対象と
なっている
データマイニング
採用企業における課題
モデルのメンテナンスが俗人化し、導入チー
ムが解散後はモデルが放置されている
新しいデータ、市場の変化に追随できず、
即時対応/精度の向上が望めない
量
継続性
Bigdata対応
© 2015 SAP SE. All rights reserved. 9
データマイニングツールでは越えられない壁・・・
Enterprise
Business Intelligence
Agile
Visualization
Advanced
Analytics
 SAP InfiniteInsight
 13年10月にKXEN社買収
 機械学習に基づくデータマイニング
自動化のユニークなアプローチ
11© 2015 SAP SE or an SAP affiliate company. All rights reserved.
Product Innovation 500+ Leading
Brands
Industry
Recognition
Telecom
Financial Services
Retail & e-Business
“Disruptor”
Leader
3x Faster
Predictive Analytics
SAP InfiniteInsightがもたらす新しいデータ解析
 世界初の商用の機械学習エンジン
(特許取得済み)
 機械学習を駆使した分析の自動化を通じ、労
働集約的な分析作業のシステム化を可能にし、
大量の施策への適用を実現
 分析人材への依存が下がることで、社内データ
の活用のボトルネック解消と、分析人材離職
時のリスクを低減
© 2015 SAP SE. All rights reserved. 11
12© 2015 SAP SE or an SAP affiliate company. All rights reserved.
SAP InfiniteInsight
13© 2015 SAP SE or an SAP affiliate company. All rights reserved.
モデル開発
自動化&
他システム連携
データ準備
モデルの展開
(スコアリング)
データ探索、傾向分析
テキスト分析
リンク(“つながり”)分析
インフルエンサーの発見
回帰分析
クラスタリング
アソシエーション
フォーキャスティング
・製品ターゲティング
ロジック開発
・ウェブサイトへの組込
スコアリングの
In-Database化
(Optimized SQL for
RDBMS:SAP Hana,
Sybase IQ,etc.)
他アプリへの組み込み
(C++, PMML, Java,
SAS, etc.)
モデルのリフレッシュ
再学習
スコアリング
精度低下の通知
SAP InfiniteInsightの特徴
目的別に用意されたメニューからのモデル開発
SAP InfiniteInsight
データマイニング自動化により生産性や意思決定精度が向上
サンプリ
ング
データ
分割
データ
加工
変数
選択
回帰分析
決定木
ニューラル
MBR
・・・
モデル生成
アセスメント
リサンプリング
アウトプット
データ準備作業(手作業)
約6週間伝統的なデータマイニング・アプローチ :各ノードを手動で設定・調整しマイニングを実行
SAP InfiniteInsightのアプローチ:データマイングプロセスの手動設定は殆ど必要無し
アウトプット
数時間
~1日
Explorerによる
自動処理
モデル構築・検証(手作業)
Modelerによる
自動処理
データ
データ
15© 2013 SAP AG or an SAP affiliate company.All rights reserved.
Modeler Demo
16© 2015 SAP SE or an SAP affiliate company. All rights reserved.
Estimation
(モデル作成用)
Analytical
Data Set
Model 1
1. データ分割
Cutting
Strategy
Validation
(評価用)
Test
(テスト用)
Model 2
Model N
2. 複数のモデル作成
3. 評価データに適用し、最適な
モデルを選択
(Additional)
最適モデルのテスト
SRM原理にて評価
二つのモデル評価指標
予測力 : KI
予測信頼度 : KR
各種
統計値/レポート
4.モデルの確認
•連続値をグループ化
•外れ値/欠損値の加工
•すべての変数から相関性の低い
項目を自動除外
•変数の組み合わせに応じて複数
モデル作成
【モデル精査の確認】
•分割データの概要
•モデル評価結果
【精査済みモデルの確認】
•Validation用データ適用結果
•変数別の基本統計値
•変数別貢献度とその内訳
•度数、累計とターゲット変数に対する
構成比
•自動作成変数(グループ化)詳細
モデルの自動作成機能(機械学習)
SAP InfiniteInsightの特徴①
データ加工プロセスの効率化
GUIデータ加工ツールにより、プログラミングスキルに依存せずにデータが作成可能
例)利用履歴から合計金額を算出、商品マスターの結合や生年月日
から年齢を算出、
顧客番号 性別 生年月日 年齢 金額合計 金額平均
1001 男 19730421 36 663 332
1002 女 19691231 39 450 450
・・・ ・・・ ・・・ ・・・ ・・・ ・・・
1010 女 19520708 56 0 0
顧客番号 性別 生年月日
1001 男 19730421
1002 女 19691231
・・・ ・・ ・・・
1004 女 19520708
顧客番号 購入日 金額
1001 20090301 540
1001 20090401 123
・・・ ・・・ ・・・
1002 20090131 450
例)縦持ちデータを横持ちデータに変換
顧客ID 期 数量
A 前期 3
A 後期 2
B 前期 0
B 後期 2
C 前期 5
C 後期 6
顧客ID 前期 後期
A 3 2
B 0 2
C 5 6
18© 2013 SAP AG or an SAP affiliate company.All rights reserved.
Explore Demo(動画)
19© 2015 SAP SE or an SAP affiliate company. All rights reserved.
SAP InfiniteInsightの特徴②
モデルの自動作成機能(機械学習)
モデル頑健性
リスク
ベストモデル トータルリスク
モデルの精度
モデルの複雑性(変数の多さ)
 連続値のグループ化、外れ値・欠損値などの自動化により、精度を向上
 モデルの質と頑健性の最適な点を見つけるSRM(Structural Risk Minimization:構造的
リスク最小化 )原理をベースにしたモデリングにより、品質と信頼性を保ちながら
プロセスを自動化
SAP InfiniteInsightの特徴③
システム組み込みコードの生成
C
SQL
Java
作成したモデル
 作成したモデルからC, Java, SQLなどのコードを生成
 モデルをシステムに組み込み、業務でのリアルタイムなアクションを支援
商品オファー、
解約防止
不正、リスクの
検知
ソースコード生成 システムへの組込 業務への適用、活用
需給予測
21© 2013 SAP AG or an SAP affiliate company.All rights reserved.
外部書き出し Demo
22© 2015 SAP SE or an SAP affiliate company. All rights reserved.
CUSTOMER
CUSTOMER ID
CUSTOMER NAME
CUSTOMER CITY
CUSTOMER POST
CUSTOMER ST
CUSTOMER ADDR
CUSTOMER PHONE
CUSTOMER FAX
ACCOUNT
ACCOUNT NUM
OPEN DATE
STATUS
HANDSET
MANUFACTURER
CALL TYPE
TYPE ID
NETWORK
DESCRIPTION
CDR
CALL TYPE
DURATION
モデル開発
2
2 KXENがADSを読み込んでモデルを構築。
Models Warehouse
4 SQLをデータウェアハウスの中に展開/もしくは業務アプリに展開
4
SQL/Scriptの生成
3
3 KXENがSQLモデルを変換
1 ADS Generatorを使い Analytical DataSetを構築
Analytical Data Set (ADS)
ADS
Generator
1
コールセンターアプリケーション
リスト生成アプリケーション5
5 コールセンターやリスト生成アプリケーションがデータウェアハウスに格納押されているSQLを実行する。
処理プロセス全体
Java, C,
Java script
23© 2015 SAP SE or an SAP affiliate company. All rights reserved.
 新規顧客の獲得
 顧客離反の食い止め
 クロス/アップセリング
 キャンペーン反応の向上
 顧客セグメンテーション
 レコメンデーション(Web)
 インフルエンサーの特定
Operations
CRM Risk
Fraud
Predictive
Analysis
 KPI forecasting
 生産、稼働機器の異常
検知
 利用予測
 価格・在庫最適化
 クレジットスコアリング
 コンプライアンス対応
(Basel II etc.)
 防犯、テロ対策
 マネーロンダリング
 カード不正利
 オンラインゲーム不正利用
 保険金請求の不正請求
SAPが実現するPredictive Analysis
24© 2015 SAP SE or an SAP affiliate company. All rights reserved.
機械学習エンジン活用事例
※時間の許す限り
25© 2013 SAP AG or an SAP affiliate company.All rights reserved.
Run simple.
26© 2015 SAP SE or an SAP affiliate company. All rights reserved.
Copyright 2015 SAP SE
All rights reserved
No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP SE. The information contained herein may be changed without prior notice.
Some software products marketed by SAP SE and its distributors contain proprietary software components of other software vendors.
SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, Duet, Business ByDesign, ByDesign, PartnerEdge and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP
SE in Germany and in several other countries all over the world. All other product and service names mentioned and associated logos displayed are the trademarks of their respective companies. Data contained in this document serves informational
purposes only. National product specifications may vary.
The information in this document is proprietary to SAP. This document is a preliminary version and not subject to your license agreement or any other agreement with SAP. This document contains only intended strategies, developments, and functionalities
of the SAP® product and is not intended to be binding upon SAP to any particular course of business, product strategy, and/or development. SAP assumes no responsibility for errors or omissions in this document. SAP does not warrant the accuracy or
completeness of the information, text, graphics, links, or other items contained within this material. This document is provided without a warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability,
fitness for a particular purpose, or non-infringement.
SAP shall have no liability for damages of any kind including without limitation direct, special, indirect, or consequential damages that may result from the use of these materials. This limitation shall not apply in cases of intent or gross negligence.
The statutory liability for personal injury and defective products is not affected. SAP has no control over the information that you may access through the use of hot links contained in these materials and does not endorse your use of third-party Web pages
nor provide any warranty whatsoever relating to third-party Web pages
Weitergabe und Vervielfältigung dieser Publikation oder von Teilen daraus sind, zu welchem Zweck und in welcher Form auch immer, ohne die ausdrückliche schriftliche Genehmigung durch SAP SE nicht gestattet. In dieser Publikation enthaltene
Informationen können ohne vorherige Ankündigung geändert werden.
Einige von der SAP SE und deren Vertriebspartnern vertriebene Softwareprodukte können Softwarekomponenten umfassen, die Eigentum anderer Softwarehersteller sind.
SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, Duet, Business ByDesign, ByDesign, PartnerEdge und andere in diesem Dokument erwähnte SAP-Produkte und Services sowie die dazugehörigen Logos sind Marken oder eingetragene
Marken der SAP SE in Deutschland und in mehreren anderen Ländern weltweit. Alle anderen in diesem Dokument erwähnten Namen von Produkten und Services sowie die damit verbundenen Firmenlogos sind Marken der jeweiligen Unternehmen. Die
Angaben im Text sind unverbindlich und dienen lediglich zu Informationszwecken. Produkte können länderspezifische Unterschiede aufweisen.
Die in diesem Dokument enthaltenen Informationen sind Eigentum von SAP. Dieses Dokument ist eine Vorabversion und unterliegt nicht Ihrer Lizenzvereinbarung oder einer anderen Vereinbarung mit SAP. Dieses Dokument enthält nur vorgesehene
Strategien, Entwicklungen und Funktionen des SAP®-Produkts und ist für SAP nicht bindend, einen bestimmten Geschäftsweg, eine Produktstrategie bzw. -entwicklung einzuschlagen. SAP übernimmt keine Verantwortung für Fehler oder Auslassungen in
diesen Materialien. SAP garantiert nicht die Richtigkeit oder Vollständigkeit der Informationen, Texte, Grafiken, Links oder anderer in diesen Materialien enthaltenen Elemente. Diese Publikation wird ohne jegliche Gewähr, weder ausdrücklich noch
stillschweigend, bereitgestellt. Dies gilt u. a., aber nicht ausschließlich, hinsichtlich der Gewährleistung der Marktgängigkeit und der Eignung für einen bestimmten Zweck sowie für die Gewährleistung der Nichtverletzung geltenden Rechts.
SAP übernimmt keine Haftung für Schäden jeglicher Art, einschließlich und ohne Einschränkung für direkte, spezielle, indirekte oder Folgeschäden im Zusammenhang mit der Verwendung dieser Unterlagen. Diese Einschränkung gilt nicht bei Vorsatz oder
grober Fahrlässigkeit.
Die gesetzliche Haftung bei Personenschäden oder die Produkthaftung bleibt unberührt. Die Informationen, auf die Sie möglicherweise über die in diesem Material enthaltenen Hotlinks zugreifen, unterliegen nicht dem Einfluss von SAP, und SAP
unterstützt nicht die Nutzung von Internetseiten Dritter durch Sie und gibt keinerlei Gewährleistungen oder Zusagen über Internetseiten Dritter ab.
Alle Rechte vorbehalten.

More Related Content

What's hot

SAP HANA 2 SPS03 highlights and SAP HANA express edition
SAP HANA 2 SPS03 highlights and SAP HANA express editionSAP HANA 2 SPS03 highlights and SAP HANA express edition
SAP HANA 2 SPS03 highlights and SAP HANA express editionKoji Shinkubo
 
B30 hanaってどうなの?sap hana概要 by 村田聡一郎
B30 hanaってどうなの?sap hana概要 by 村田聡一郎B30 hanaってどうなの?sap hana概要 by 村田聡一郎
B30 hanaってどうなの?sap hana概要 by 村田聡一郎Insight Technology, Inc.
 
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa HanakiInsight Technology, Inc.
 
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa HanakiInsight Technology, Inc.
 
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二Insight Technology, Inc.
 
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...Insight Technology, Inc.
 
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...Insight Technology, Inc.
 
[D35] インメモリーデータベース徹底比較 by Komori
[D35] インメモリーデータベース徹底比較 by Komori[D35] インメモリーデータベース徹底比較 by Komori
[D35] インメモリーデータベース徹底比較 by KomoriInsight Technology, Inc.
 
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド Sap business suite on hana適合シナリオ診断 bsr 申込ガイド
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド Shiroh Kinoshita
 
Sap inside track2019tokyo_d3-in2_processvisibility_public
Sap inside track2019tokyo_d3-in2_processvisibility_publicSap inside track2019tokyo_d3-in2_processvisibility_public
Sap inside track2019tokyo_d3-in2_processvisibility_publicMasashi Yamazawa
 
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...Insight Technology, Inc.
 
ABAPも進化が止まらない! ABAP RESTful Programming Model
ABAPも進化が止まらない! ABAP RESTful Programming ModelABAPも進化が止まらない! ABAP RESTful Programming Model
ABAPも進化が止まらない! ABAP RESTful Programming ModelSusumuHonna
 
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーション
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーションIntelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーション
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーションYasuko Sekiguchi
 
Oracle Cloud Platform - クラクドにおける 新たなデータベース開発
Oracle Cloud Platform - クラクドにおける新たなデータベース開発Oracle Cloud Platform - クラクドにおける新たなデータベース開発
Oracle Cloud Platform - クラクドにおける 新たなデータベース開発オラクルエンジニア通信
 
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料【2016年3月時点】Data Visualization Cloud Service ハンズオン資料
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料オラクルエンジニア通信
 
20160916 ビッグデータシンポジウム オラクル公開資料
20160916 ビッグデータシンポジウム オラクル公開資料20160916 ビッグデータシンポジウム オラクル公開資料
20160916 ビッグデータシンポジウム オラクル公開資料オラクルエンジニア通信
 
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニングオラクルエンジニア通信
 
Oracle Application Expressと Oracle BI Publisherの連携例
Oracle Application Expressと Oracle BI Publisherの連携例Oracle Application Expressと Oracle BI Publisherの連携例
Oracle Application Expressと Oracle BI Publisherの連携例utatu
 
Intalio|BPP as SaaS Platform
Intalio|BPP as SaaS PlatformIntalio|BPP as SaaS Platform
Intalio|BPP as SaaS PlatformTomoaki Sawada
 

What's hot (20)

SAP HANA 2 SPS03 highlights and SAP HANA express edition
SAP HANA 2 SPS03 highlights and SAP HANA express editionSAP HANA 2 SPS03 highlights and SAP HANA express edition
SAP HANA 2 SPS03 highlights and SAP HANA express edition
 
B30 hanaってどうなの?sap hana概要 by 村田聡一郎
B30 hanaってどうなの?sap hana概要 by 村田聡一郎B30 hanaってどうなの?sap hana概要 by 村田聡一郎
B30 hanaってどうなの?sap hana概要 by 村田聡一郎
 
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki
[D13] 次世代型インメモリデータベース SAP HANA その最新技術を理解する by Toshihisa Hanaki
 
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki
[B25] 【詳解】インメモリーデータベース SAP HANA:永続化の仕組みと障害/災害対応を踏まえた運用を理解しよう!by Toshihisa Hanaki
 
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二
20170518_今さら聞けないHANAのハナシの基本のき by SAPジャパン株式会社 新久保浩二
 
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...
B23,B31 sap sybase iq ~全部話します。IQのカラムストア方式、ビットワイズインデックス、DQP、カラム圧縮、等々 by Toshih...
 
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...
[db tech showcase Tokyo 2014] B11: [In-Memory DB: SAP HANA] OLTP/OLAPをシングルデータ...
 
[D35] インメモリーデータベース徹底比較 by Komori
[D35] インメモリーデータベース徹底比較 by Komori[D35] インメモリーデータベース徹底比較 by Komori
[D35] インメモリーデータベース徹底比較 by Komori
 
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド Sap business suite on hana適合シナリオ診断 bsr 申込ガイド
Sap business suite on hana適合シナリオ診断 bsr 申込ガイド
 
Sapporo tech bar 21
Sapporo tech bar 21Sapporo tech bar 21
Sapporo tech bar 21
 
Sap inside track2019tokyo_d3-in2_processvisibility_public
Sap inside track2019tokyo_d3-in2_processvisibility_publicSap inside track2019tokyo_d3-in2_processvisibility_public
Sap inside track2019tokyo_d3-in2_processvisibility_public
 
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...
[db tech showcase Tokyo 2015] D22:インメモリープラットホームSAP HANAのご紹介と最新情報 by SAPジャパン株式...
 
ABAPも進化が止まらない! ABAP RESTful Programming Model
ABAPも進化が止まらない! ABAP RESTful Programming ModelABAPも進化が止まらない! ABAP RESTful Programming Model
ABAPも進化が止まらない! ABAP RESTful Programming Model
 
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーション
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーションIntelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーション
Intelligent Enterprise Update S/4 HANA 導入と並行してすすめるイノベーション
 
Oracle Cloud Platform - クラクドにおける 新たなデータベース開発
Oracle Cloud Platform - クラクドにおける新たなデータベース開発Oracle Cloud Platform - クラクドにおける新たなデータベース開発
Oracle Cloud Platform - クラクドにおける 新たなデータベース開発
 
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料【2016年3月時点】Data Visualization Cloud Service ハンズオン資料
【2016年3月時点】Data Visualization Cloud Service ハンズオン資料
 
20160916 ビッグデータシンポジウム オラクル公開資料
20160916 ビッグデータシンポジウム オラクル公開資料20160916 ビッグデータシンポジウム オラクル公開資料
20160916 ビッグデータシンポジウム オラクル公開資料
 
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング
20151209 Oracle DDD オラクルで実現するクラウド・マシン・ラーニング
 
Oracle Application Expressと Oracle BI Publisherの連携例
Oracle Application Expressと Oracle BI Publisherの連携例Oracle Application Expressと Oracle BI Publisherの連携例
Oracle Application Expressと Oracle BI Publisherの連携例
 
Intalio|BPP as SaaS Platform
Intalio|BPP as SaaS PlatformIntalio|BPP as SaaS Platform
Intalio|BPP as SaaS Platform
 

Similar to 20150329 SAP InfiniteInsight(SAPが提供する機械学習エンジン)

Sap fiori apps recommendation service v3
Sap fiori apps recommendation service v3Sap fiori apps recommendation service v3
Sap fiori apps recommendation service v3Shiroh Kinoshita
 
SAP HANA SP10最新情報詳細版
SAP HANA SP10最新情報詳細版SAP HANA SP10最新情報詳細版
SAP HANA SP10最新情報詳細版Mana Matsudate
 
SAP TechED の歩き方2018
SAP TechED の歩き方2018SAP TechED の歩き方2018
SAP TechED の歩き方2018Mana Matsudate
 
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例オラクルエンジニア通信
 
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16Ryusuke Ashiya
 
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要オラクルエンジニア通信
 
AWSではじめるSAP HANA入門
AWSではじめるSAP HANA入門AWSではじめるSAP HANA入門
AWSではじめるSAP HANA入門Mana Matsudate
 
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ -
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ - Salesforce Einstein - SaaS企業のAI戦略とテクノロジ -
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ - Mitch Okamoto
 
SAP Cloud Platform から Microsoft Azure サービス利用は便利だった
SAP Cloud Platform から Microsoft Azureサービス利用は便利だったSAP Cloud Platform から Microsoft Azureサービス利用は便利だった
SAP Cloud Platform から Microsoft Azure サービス利用は便利だったMasayuki Sekihara
 
機械学習 - MNIST の次のステップ
機械学習 - MNIST の次のステップ機械学習 - MNIST の次のステップ
機械学習 - MNIST の次のステップDaiyu Hatakeyama
 
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」Masahiro Furusawa
 
食わず嫌いの為のSalesforce1 Platform入門
食わず嫌いの為のSalesforce1 Platform入門食わず嫌いの為のSalesforce1 Platform入門
食わず嫌いの為のSalesforce1 Platform入門Mitch Okamoto
 
AppExchangeパートナー&デベロッパー第1部:20071205版
AppExchangeパートナー&デベロッパー第1部:20071205版AppExchangeパートナー&デベロッパー第1部:20071205版
AppExchangeパートナー&デベロッパー第1部:20071205版Junichiro Tasaki
 
ERPのデータをフロントシステムでどう活かすか
ERPのデータをフロントシステムでどう活かすかERPのデータをフロントシステムでどう活かすか
ERPのデータをフロントシステムでどう活かすかRyuji Enoki
 
Heroku Getting Started
Heroku Getting StartedHeroku Getting Started
Heroku Getting StartedAyumu Aizawa
 
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!Accenture Japan
 
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...Insight Technology, Inc.
 
[CTO Night & Day 2019] ML services: MLOps #ctonight
[CTO Night & Day 2019] ML services: MLOps #ctonight[CTO Night & Day 2019] ML services: MLOps #ctonight
[CTO Night & Day 2019] ML services: MLOps #ctonightAmazon Web Services Japan
 
非SAPの人に贈るSAP on AWS
非SAPの人に贈るSAP on AWS非SAPの人に贈るSAP on AWS
非SAPの人に贈るSAP on AWSRyusaburo Tanaka
 

Similar to 20150329 SAP InfiniteInsight(SAPが提供する機械学習エンジン) (20)

Sap fiori apps recommendation service v3
Sap fiori apps recommendation service v3Sap fiori apps recommendation service v3
Sap fiori apps recommendation service v3
 
SAP HANA SP10最新情報詳細版
SAP HANA SP10最新情報詳細版SAP HANA SP10最新情報詳細版
SAP HANA SP10最新情報詳細版
 
SAP TechED の歩き方2018
SAP TechED の歩き方2018SAP TechED の歩き方2018
SAP TechED の歩き方2018
 
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例
Oracle Data Minerハンズオンセミナー170927:③Business Analytics概要と事例
 
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16
Tableau Blueprintの概要 for JTUG/RETAIL 2019/10/16
 
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要
Oracle Data Minerハンズオンセミナー170927:①Oracle 機械学習概要
 
AWSではじめるSAP HANA入門
AWSではじめるSAP HANA入門AWSではじめるSAP HANA入門
AWSではじめるSAP HANA入門
 
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ -
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ - Salesforce Einstein - SaaS企業のAI戦略とテクノロジ -
Salesforce Einstein - SaaS企業のAI戦略とテクノロジ -
 
SAP Cloud Platform から Microsoft Azure サービス利用は便利だった
SAP Cloud Platform から Microsoft Azureサービス利用は便利だったSAP Cloud Platform から Microsoft Azureサービス利用は便利だった
SAP Cloud Platform から Microsoft Azure サービス利用は便利だった
 
Oracle Advanced Analytics 概要
Oracle Advanced Analytics 概要Oracle Advanced Analytics 概要
Oracle Advanced Analytics 概要
 
機械学習 - MNIST の次のステップ
機械学習 - MNIST の次のステップ機械学習 - MNIST の次のステップ
機械学習 - MNIST の次のステップ
 
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」
2016 #meijisap - 明治大学理工学部情報科学科 情報システム論1講義「デジタルによるビジネスモデルの変革」
 
食わず嫌いの為のSalesforce1 Platform入門
食わず嫌いの為のSalesforce1 Platform入門食わず嫌いの為のSalesforce1 Platform入門
食わず嫌いの為のSalesforce1 Platform入門
 
AppExchangeパートナー&デベロッパー第1部:20071205版
AppExchangeパートナー&デベロッパー第1部:20071205版AppExchangeパートナー&デベロッパー第1部:20071205版
AppExchangeパートナー&デベロッパー第1部:20071205版
 
ERPのデータをフロントシステムでどう活かすか
ERPのデータをフロントシステムでどう活かすかERPのデータをフロントシステムでどう活かすか
ERPのデータをフロントシステムでどう活かすか
 
Heroku Getting Started
Heroku Getting StartedHeroku Getting Started
Heroku Getting Started
 
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!
Screen Personas 3.0を活用し、シンプルで楽しいFiori UXを!
 
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...
[db tech showcase Tokyo 2015] B27:インメモリーDBとスケールアップマシンによりBig Dataの課題を解決する by S...
 
[CTO Night & Day 2019] ML services: MLOps #ctonight
[CTO Night & Day 2019] ML services: MLOps #ctonight[CTO Night & Day 2019] ML services: MLOps #ctonight
[CTO Night & Day 2019] ML services: MLOps #ctonight
 
非SAPの人に贈るSAP on AWS
非SAPの人に贈るSAP on AWS非SAPの人に贈るSAP on AWS
非SAPの人に贈るSAP on AWS
 

20150329 SAP InfiniteInsight(SAPが提供する機械学習エンジン)

  • 2. 2© 2015 SAP SE or an SAP affiliate company. All rights reserved. 機械学習で何ができるの?(代表例) Classification:分類 Who will (buy | fraud | churn …) next (week | month | year…) ? Prediction:発生確率 How will the (revenue | # churners) be next (week | month..)? Segmentation:セグメンテーション What are the groups of customers with similar (behavior | profile …)? Social Network Analysis: インフルエンサー分析 Analyze interactions to identify (communities | influencers…) Association Rules:連関性分析 Analyze transactions to identify events likely to occur together Forecasting:予測値計算 How will the (revenue | # churners…) be over next year on a monthly basis ?
  • 3. 3© 2013 SAP AG or an SAP affiliate company.All rights reserved. 本日の目的① アルゴリズム云々のアカデミックな内容は 極力少なくし、人(データサイエンティストの方)の 苦労を如何に自動化するか。 その壁を越えた際に見えてくる世界。 そこにフォーカスして話をさせて頂きます。
  • 4. 4© 2013 SAP AG or an SAP affiliate company.All rights reserved. 本日の目的② 実際にこの時間の中で “解析“~”システムへの適用” まで実演します。(できます。)
  • 5. 5© 2015 SAP SE or an SAP affiliate company. All rights reserved. ビジネス バリュー ビジネスチャンス 期間 試行錯誤を行うデータ解析の課題 通常の解析モデル N日 時間 Nヶ月 データ加工 モデル構築 アクション 実行 90点 意思決定のニーズ /依頼
  • 6. 6© 2015 SAP SE or an SAP affiliate company. All rights reserved. ビジネス バリュー こんな世界を目指します。 n日 Nか月 時間 意思決定の ニーズ たとえばこのケースの 場合どちらを選択しますか? アクション 実行 88点 ビジネスチャンス 期間 アクション 実行 90点
  • 7. 7© 2015 SAP SE or an SAP affiliate company. All rights reserved. さらにBigdata化(縦の空間、横の空間) コア事業 クレジット カード 多ブランド 運営 プライベート ブランド アフター サービス ポイント プログラム テナント 運営 自社チャネル利用 明細(POS) Web広告への反応 コールセンターなどの 接点ログ カード利用履歴 契約情報 自社サイト利用履歴 Social Media 事業の多様化 顧客行動データの 収集
  • 8. 8© 2015 SAP SE or an SAP affiliate company. All rights reserved. やれることはたくさんある 自社チャネル利用 明細(POS) Web広告への反応 コールセンターなどの 接点ログ カード利用履歴 契約情報 自社サイト利用履歴 Social Media 商品への 興味 家族構成 就労環境 ライフ イベント 生活習慣 データ 情報 カード会員向け施策 自社小売チャネルへの フィードバック オムニチャネルの実現 加盟店向け 情報提供 施策 “モノ”から“コト”へのシフト
  • 9. 9© 2015 SAP SE or an SAP affiliate company. All rights reserved. 2000年代前半までのデータマイニングブーム  モデルの精度がツールではなく、分析者の力量に依存する  分析者によって嗜好が異なり、担当者が変わると再モデル作成  モデルの開発に非常に長い時間を費やす  データサイエンティストだけでなく、プログラミングスキル、ビジネスノウハウを持った“チーム“で取り 掛かる必要がある データマイニングに必要なリソース データマイニングアプローチ サンプリング データ加工 変数選択 回帰分析 データ分割 決定木 ニューラル MBR アセスメント ・・・more モデル合成 サンプリング Data 対象となる分析シナリオを すべて実装できず一部のみが対象と なっている データマイニング 採用企業における課題 モデルのメンテナンスが俗人化し、導入チー ムが解散後はモデルが放置されている 新しいデータ、市場の変化に追随できず、 即時対応/精度の向上が望めない 量 継続性 Bigdata対応 © 2015 SAP SE. All rights reserved. 9 データマイニングツールでは越えられない壁・・・
  • 10. Enterprise Business Intelligence Agile Visualization Advanced Analytics  SAP InfiniteInsight  13年10月にKXEN社買収  機械学習に基づくデータマイニング 自動化のユニークなアプローチ
  • 11. 11© 2015 SAP SE or an SAP affiliate company. All rights reserved. Product Innovation 500+ Leading Brands Industry Recognition Telecom Financial Services Retail & e-Business “Disruptor” Leader 3x Faster Predictive Analytics SAP InfiniteInsightがもたらす新しいデータ解析  世界初の商用の機械学習エンジン (特許取得済み)  機械学習を駆使した分析の自動化を通じ、労 働集約的な分析作業のシステム化を可能にし、 大量の施策への適用を実現  分析人材への依存が下がることで、社内データ の活用のボトルネック解消と、分析人材離職 時のリスクを低減 © 2015 SAP SE. All rights reserved. 11
  • 12. 12© 2015 SAP SE or an SAP affiliate company. All rights reserved. SAP InfiniteInsight
  • 13. 13© 2015 SAP SE or an SAP affiliate company. All rights reserved. モデル開発 自動化& 他システム連携 データ準備 モデルの展開 (スコアリング) データ探索、傾向分析 テキスト分析 リンク(“つながり”)分析 インフルエンサーの発見 回帰分析 クラスタリング アソシエーション フォーキャスティング ・製品ターゲティング ロジック開発 ・ウェブサイトへの組込 スコアリングの In-Database化 (Optimized SQL for RDBMS:SAP Hana, Sybase IQ,etc.) 他アプリへの組み込み (C++, PMML, Java, SAS, etc.) モデルのリフレッシュ 再学習 スコアリング 精度低下の通知 SAP InfiniteInsightの特徴 目的別に用意されたメニューからのモデル開発
  • 15. 15© 2013 SAP AG or an SAP affiliate company.All rights reserved. Modeler Demo
  • 16. 16© 2015 SAP SE or an SAP affiliate company. All rights reserved. Estimation (モデル作成用) Analytical Data Set Model 1 1. データ分割 Cutting Strategy Validation (評価用) Test (テスト用) Model 2 Model N 2. 複数のモデル作成 3. 評価データに適用し、最適な モデルを選択 (Additional) 最適モデルのテスト SRM原理にて評価 二つのモデル評価指標 予測力 : KI 予測信頼度 : KR 各種 統計値/レポート 4.モデルの確認 •連続値をグループ化 •外れ値/欠損値の加工 •すべての変数から相関性の低い 項目を自動除外 •変数の組み合わせに応じて複数 モデル作成 【モデル精査の確認】 •分割データの概要 •モデル評価結果 【精査済みモデルの確認】 •Validation用データ適用結果 •変数別の基本統計値 •変数別貢献度とその内訳 •度数、累計とターゲット変数に対する 構成比 •自動作成変数(グループ化)詳細 モデルの自動作成機能(機械学習)
  • 17. SAP InfiniteInsightの特徴① データ加工プロセスの効率化 GUIデータ加工ツールにより、プログラミングスキルに依存せずにデータが作成可能 例)利用履歴から合計金額を算出、商品マスターの結合や生年月日 から年齢を算出、 顧客番号 性別 生年月日 年齢 金額合計 金額平均 1001 男 19730421 36 663 332 1002 女 19691231 39 450 450 ・・・ ・・・ ・・・ ・・・ ・・・ ・・・ 1010 女 19520708 56 0 0 顧客番号 性別 生年月日 1001 男 19730421 1002 女 19691231 ・・・ ・・ ・・・ 1004 女 19520708 顧客番号 購入日 金額 1001 20090301 540 1001 20090401 123 ・・・ ・・・ ・・・ 1002 20090131 450 例)縦持ちデータを横持ちデータに変換 顧客ID 期 数量 A 前期 3 A 後期 2 B 前期 0 B 後期 2 C 前期 5 C 後期 6 顧客ID 前期 後期 A 3 2 B 0 2 C 5 6
  • 18. 18© 2013 SAP AG or an SAP affiliate company.All rights reserved. Explore Demo(動画)
  • 19. 19© 2015 SAP SE or an SAP affiliate company. All rights reserved. SAP InfiniteInsightの特徴② モデルの自動作成機能(機械学習) モデル頑健性 リスク ベストモデル トータルリスク モデルの精度 モデルの複雑性(変数の多さ)  連続値のグループ化、外れ値・欠損値などの自動化により、精度を向上  モデルの質と頑健性の最適な点を見つけるSRM(Structural Risk Minimization:構造的 リスク最小化 )原理をベースにしたモデリングにより、品質と信頼性を保ちながら プロセスを自動化
  • 20. SAP InfiniteInsightの特徴③ システム組み込みコードの生成 C SQL Java 作成したモデル  作成したモデルからC, Java, SQLなどのコードを生成  モデルをシステムに組み込み、業務でのリアルタイムなアクションを支援 商品オファー、 解約防止 不正、リスクの 検知 ソースコード生成 システムへの組込 業務への適用、活用 需給予測
  • 21. 21© 2013 SAP AG or an SAP affiliate company.All rights reserved. 外部書き出し Demo
  • 22. 22© 2015 SAP SE or an SAP affiliate company. All rights reserved. CUSTOMER CUSTOMER ID CUSTOMER NAME CUSTOMER CITY CUSTOMER POST CUSTOMER ST CUSTOMER ADDR CUSTOMER PHONE CUSTOMER FAX ACCOUNT ACCOUNT NUM OPEN DATE STATUS HANDSET MANUFACTURER CALL TYPE TYPE ID NETWORK DESCRIPTION CDR CALL TYPE DURATION モデル開発 2 2 KXENがADSを読み込んでモデルを構築。 Models Warehouse 4 SQLをデータウェアハウスの中に展開/もしくは業務アプリに展開 4 SQL/Scriptの生成 3 3 KXENがSQLモデルを変換 1 ADS Generatorを使い Analytical DataSetを構築 Analytical Data Set (ADS) ADS Generator 1 コールセンターアプリケーション リスト生成アプリケーション5 5 コールセンターやリスト生成アプリケーションがデータウェアハウスに格納押されているSQLを実行する。 処理プロセス全体 Java, C, Java script
  • 23. 23© 2015 SAP SE or an SAP affiliate company. All rights reserved.  新規顧客の獲得  顧客離反の食い止め  クロス/アップセリング  キャンペーン反応の向上  顧客セグメンテーション  レコメンデーション(Web)  インフルエンサーの特定 Operations CRM Risk Fraud Predictive Analysis  KPI forecasting  生産、稼働機器の異常 検知  利用予測  価格・在庫最適化  クレジットスコアリング  コンプライアンス対応 (Basel II etc.)  防犯、テロ対策  マネーロンダリング  カード不正利  オンラインゲーム不正利用  保険金請求の不正請求 SAPが実現するPredictive Analysis
  • 24. 24© 2015 SAP SE or an SAP affiliate company. All rights reserved. 機械学習エンジン活用事例 ※時間の許す限り
  • 25. 25© 2013 SAP AG or an SAP affiliate company.All rights reserved. Run simple.
  • 26. 26© 2015 SAP SE or an SAP affiliate company. All rights reserved. Copyright 2015 SAP SE All rights reserved No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP SE. The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its distributors contain proprietary software components of other software vendors. SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, Duet, Business ByDesign, ByDesign, PartnerEdge and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP SE in Germany and in several other countries all over the world. All other product and service names mentioned and associated logos displayed are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary. The information in this document is proprietary to SAP. This document is a preliminary version and not subject to your license agreement or any other agreement with SAP. This document contains only intended strategies, developments, and functionalities of the SAP® product and is not intended to be binding upon SAP to any particular course of business, product strategy, and/or development. SAP assumes no responsibility for errors or omissions in this document. SAP does not warrant the accuracy or completeness of the information, text, graphics, links, or other items contained within this material. This document is provided without a warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. SAP shall have no liability for damages of any kind including without limitation direct, special, indirect, or consequential damages that may result from the use of these materials. This limitation shall not apply in cases of intent or gross negligence. The statutory liability for personal injury and defective products is not affected. SAP has no control over the information that you may access through the use of hot links contained in these materials and does not endorse your use of third-party Web pages nor provide any warranty whatsoever relating to third-party Web pages Weitergabe und Vervielfältigung dieser Publikation oder von Teilen daraus sind, zu welchem Zweck und in welcher Form auch immer, ohne die ausdrückliche schriftliche Genehmigung durch SAP SE nicht gestattet. In dieser Publikation enthaltene Informationen können ohne vorherige Ankündigung geändert werden. Einige von der SAP SE und deren Vertriebspartnern vertriebene Softwareprodukte können Softwarekomponenten umfassen, die Eigentum anderer Softwarehersteller sind. SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, Duet, Business ByDesign, ByDesign, PartnerEdge und andere in diesem Dokument erwähnte SAP-Produkte und Services sowie die dazugehörigen Logos sind Marken oder eingetragene Marken der SAP SE in Deutschland und in mehreren anderen Ländern weltweit. Alle anderen in diesem Dokument erwähnten Namen von Produkten und Services sowie die damit verbundenen Firmenlogos sind Marken der jeweiligen Unternehmen. Die Angaben im Text sind unverbindlich und dienen lediglich zu Informationszwecken. Produkte können länderspezifische Unterschiede aufweisen. Die in diesem Dokument enthaltenen Informationen sind Eigentum von SAP. Dieses Dokument ist eine Vorabversion und unterliegt nicht Ihrer Lizenzvereinbarung oder einer anderen Vereinbarung mit SAP. Dieses Dokument enthält nur vorgesehene Strategien, Entwicklungen und Funktionen des SAP®-Produkts und ist für SAP nicht bindend, einen bestimmten Geschäftsweg, eine Produktstrategie bzw. -entwicklung einzuschlagen. SAP übernimmt keine Verantwortung für Fehler oder Auslassungen in diesen Materialien. SAP garantiert nicht die Richtigkeit oder Vollständigkeit der Informationen, Texte, Grafiken, Links oder anderer in diesen Materialien enthaltenen Elemente. Diese Publikation wird ohne jegliche Gewähr, weder ausdrücklich noch stillschweigend, bereitgestellt. Dies gilt u. a., aber nicht ausschließlich, hinsichtlich der Gewährleistung der Marktgängigkeit und der Eignung für einen bestimmten Zweck sowie für die Gewährleistung der Nichtverletzung geltenden Rechts. SAP übernimmt keine Haftung für Schäden jeglicher Art, einschließlich und ohne Einschränkung für direkte, spezielle, indirekte oder Folgeschäden im Zusammenhang mit der Verwendung dieser Unterlagen. Diese Einschränkung gilt nicht bei Vorsatz oder grober Fahrlässigkeit. Die gesetzliche Haftung bei Personenschäden oder die Produkthaftung bleibt unberührt. Die Informationen, auf die Sie möglicherweise über die in diesem Material enthaltenen Hotlinks zugreifen, unterliegen nicht dem Einfluss von SAP, und SAP unterstützt nicht die Nutzung von Internetseiten Dritter durch Sie und gibt keinerlei Gewährleistungen oder Zusagen über Internetseiten Dritter ab. Alle Rechte vorbehalten.