Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
機械学習・データサイエンスを活用したサービスのユーザー体験(UX)向上の取り組み / Eff...
Search
Sponsored
·
Ship Features Fearlessly
Turn features on and off without deploys. Used by thousands of Ruby developers.
→
Kiyota Yoji, Ph.D.
March 05, 2019
Technology
0
140
機械学習・データサイエンスを活用したサービスのユーザー体験(UX)向上の取り組み / Efforts to Improve UX of Service Utilizing Machine Learning and Data Science
DEIM 2019 Industrial Session @ Nagasaki
Kiyota Yoji, Ph.D.
March 05, 2019
Tweet
Share
More Decks by Kiyota Yoji, Ph.D.
See All by Kiyota Yoji, Ph.D.
不動産データを対象としたビジネス応⽤事例 (IDRユーザフォーラム2025) / Business Applications of Real Estate Data: Case Studies (IDR User Forum 2025)
ykiyota
0
160
紙の答案の採点・集計・ 返却を少しでも楽にしたい! / How to Simplify Grading, Tallying, and Returning Paper Exams
ykiyota
0
43
LIFULL HOME'Sデータセットを活用した研究の傾向分析と今後の展望 / Trend Analysis and Future Prospects for Research Using the LIFULL HOME'S Dataset
ykiyota
0
68
AIの視点からみた不動産のフロンティア / Frontiers of Real Estate from the Perspective of Artificial Intelligence
ykiyota
0
160
INFOSTA AI利活用研究会 (AISG-INFOSTA)のご紹介 / Introduction to AISG-INFOSTA
ykiyota
0
160
不動産情報サービスの研究開発における共有データ資源 / Shared data resources in research and development of real estate information services
ykiyota
0
550
JSAI 2023企画セッション「AI哲学マップ」企画の振り返り / JSAI 2023 Reflections on the AI Philosophy Map lecture series project
ykiyota
1
220
住まい探しの利便性向上にデータベース・情報アクセス技術が 果たした役割 / The Role of Database and Information Access Technology in Improving the Convenience of Housing Search
ykiyota
0
150
デジタル社会の行き着く先にライブラリアンが果たしうる役割を考える / Thinking about the role librarians can play in the destination of the digital society
ykiyota
0
2.1k
Other Decks in Technology
See All in Technology
今こそ学びたいKubernetesネットワーク ~CNIが繋ぐNWとプラットフォームの「フラッと」な対話
logica0419
5
510
マーケットプレイス版Oracle WebCenter Content For OCI
oracle4engineer
PRO
5
1.6k
22nd ACRi Webinar - NTT Kawahara-san's slide
nao_sumikawa
0
110
[CV勉強会@関東 World Model 読み会] Orbis: Overcoming Challenges of Long-Horizon Prediction in Driving World Models (Mousakhan+, NeurIPS 2025)
abemii
0
150
OWASP Top 10:2025 リリースと 少しの日本語化にまつわる裏話
okdt
PRO
3
850
Claude Code for NOT Programming
kawaguti
PRO
1
110
茨城の思い出を振り返る ~CDKのセキュリティを添えて~ / 20260201 Mitsutoshi Matsuo
shift_evolve
PRO
1
420
Oracle AI Database移行・アップグレード勉強会 - RAT活用編
oracle4engineer
PRO
0
110
登壇駆動学習のすすめ — CfPのネタの見つけ方と書くときに意識していること
bicstone
3
130
Bedrock PolicyでAmazon Bedrock Guardrails利用を強制してみた
yuu551
0
260
Context Engineeringが企業で不可欠になる理由
hirosatogamo
PRO
3
680
生成AIと余白 〜開発スピードが向上した今、何に向き合う?〜
kakehashi
PRO
0
170
Featured
See All Featured
Designing for humans not robots
tammielis
254
26k
Neural Spatial Audio Processing for Sound Field Analysis and Control
skoyamalab
0
180
Lessons Learnt from Crawling 1000+ Websites
charlesmeaden
PRO
1
1.1k
The Mindset for Success: Future Career Progression
greggifford
PRO
0
240
How People are Using Generative and Agentic AI to Supercharge Their Products, Projects, Services and Value Streams Today
helenjbeal
1
130
Building Applications with DynamoDB
mza
96
6.9k
The State of eCommerce SEO: How to Win in Today's Products SERPs - #SEOweek
aleyda
2
9.6k
The Art of Programming - Codeland 2020
erikaheidi
57
14k
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
16
1.8k
Chasing Engaging Ingredients in Design
codingconduct
0
120
GraphQLの誤解/rethinking-graphql
sonatard
74
11k
Google's AI Overviews - The New Search
badams
0
910
Transcript
" ! A I
D E I M 2 0 1 9 2 0 1 9 . 0 3 . 0 6 C o p y r i g h t © L I F U L L A l l R i g h t s R e s e r v e d .
- %4!+No.1 32&(.)6 5 7/ )6 ( , ,
, , etc.) ,850$# 1 (.0"'* : ( ) LIFULL
None
None
(NII-IDR) LIFULL HOME’S • 2015411!
1' • %350) #,* <;/ • 3: • • 201548!$& +0.7" • ) 56- • !(NEW! • 201547!201746! 24!8 • 2 269
() • .5-2J/ • TSV " (bz2$6)1#. 11 • 2015E7.2017E6.24#.K
• @<OM( • J/83> HN • ':FH • • *7A&=) • ! • 4,+?;BDP9%CDI0LBDP9ID G
UX "!20# UX$ •
• E • VR / AR • …
!" #
• •
etc. • • • • • • etc.
UX • Uber • Airbnb •
• …
UX • HW3FAX • GP@# • T. O*9& O*!
• 7U4-;RCS6 • MJ1:QUX'?)% … • > 3LB,/5DUX" • IKN+2(FAV<$08= O*E
• UX & 8*#) &
!39 • 50#'7UX - "+6 ,4$. & • 621 - ( /%:
(RSC 2015 )
› S › FI S 0 1 • S •
3 ʼ u U u u u 3 u u u u u u MO L u E u 1 U u u u u u IH u u (CNN)
› 4: 3 3 0 97.3% 152.0% E .91.0%
100.0% 1 40.1% 1 20.6% 12.1% 64.7% 26.2% : 3.1% 75.0% : 9.2% 3.7%
• #$
! # $ " ! • "% #
,' 1or% $* LIFULL HOME’S )-$ !.+ ↓ "/#&
0"/ (
E 24 * 1000 Cloud Machine
Learning Cloud Storage tensorflow
None
© LIFULL Co.,Ltd . © LIFULL Co.,Ltd .
None
3D Simon Fraser #"
https://www.cse.wustl.edu/~chenliu/floorplan-transformation.html → $ ICCV 2017!
None
LIFULL HOME’S • '&$$ • "48(% #!
UX • "
9( • 3# 3!#%' • '$42 • *0+57.-2 • • etc. • ,) 16&8 /
LIFULL AI • 2018:7#)56" AI(1<> /4 • * • =$
2 -0!+8? • ;9%&, =$3.7 • ;9% '9 • etc. • We are hiring!
None