Upgrade to Pro — share decks privately, control downloads, hide ads and more …

トピックモデルによる統計的潜在意味解析 3.2.3 ~ 3.2.4 / topic-model...

Avatar for ysekky ysekky
January 07, 2016
Research
0
540

トピックモデルによる統計的潜在意味解析 3.2.3 ~ 3.2.4 / topic-model-3-2-3

Avatar for ysekky

ysekky

January 07, 2016
Tweet

More Decks by ysekky

See All by ysekky
スタートアップの開発サイクルに学ぶ 研究活動の進め方 / research practices inspired by startup business strategy
Avatar for ysekky ysekky
0
2.3k
[論文紹介] A Method to Anonymize Business Metrics to Publishing Implicit Feedback Datasets (Recsys2020) / recsys20-reading-gunosy-datapub
Avatar for ysekky ysekky
3
2.7k
JSAI2020 OS-12 広告とAI オープニング / JSAI2020-OS-12-ads-and-ai-opening
Avatar for ysekky ysekky
0
2.1k
JSAI2020インダストリアルセッション - Gunosyにおける研究開発 / jsai2020-gunosy-rd-examples
Avatar for ysekky ysekky
1
780
ウェブサービス事業者における研究開発インターン[株式会社Gunosy] - テキストアナリティクスシンポジウム2019 / research-intern-case-study-at-gunosy
Avatar for ysekky ysekky
0
2.9k
Gunosyにおけるニュース記事推薦/ news-recommendation-in-gunosy-webdbf2019
Avatar for ysekky ysekky
1
1.5k
DEIM2019技術報告セッション - Gunosyの研究開発 / deim-2019-sponsor-session-gunosy-research
Avatar for ysekky ysekky
0
1.2k
Analysis of Bias in Gathering Information Between User Attributes in News Application (ABCCS 2018)
Avatar for ysekky ysekky
1
2.4k
世代による政治ニュース記事の閲覧傾向の違いの分析 - JSAI2018 / Analysis of differences in viewing behavior of politics news by age
Avatar for ysekky ysekky
0
4k

Other Decks in Research

See All in Research
Sat2City:3D City Generation from A Single Satellite Image with Cascaded Latent Diffusion
Avatar for SatAI.challenge satai
3
160
Learning to (Learn at Test Time): RNNs with Expressive Hidden States
Avatar for Hiroto Kurita kurita
1
280
Minimax and Bayes Optimal Best-arm Identification: Adaptive Experimental Design for Treatment Choice
Avatar for MasaKat0 masakat0
0
180
AIグラフィックデザインの進化：断片から統合（One Piece）へ / From Fragment to One Piece: A Survey on AI-Driven Graphic Design
Avatar for Shunsuke KITADA shunk031
0
520
AIスパコン「さくらONE」の
オブザーバビリティ / Observability for AI Supercomputer SAKURAONE
Avatar for Yuuki Tsubouchi (yuuk1) yuukit
2
440
Mamba-in-Mamba: Centralized Mamba-Cross-Scan in Tokenized Mamba Model for Hyperspectral Image Classification
Avatar for SatAI.challenge satai
3
130
AIスパコン「さくらONE」のLLM学習ベンチマークによる性能評価 / SAKURAONE LLM Training Benchmarking
Avatar for Yuuki Tsubouchi (yuuk1) yuukit
2
750
EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues
Avatar for SatAI.challenge satai
3
260
令和最新技術で伝統掲示板を再構築: HonoX で作る型安全なスレッドフロート型掲示板 / かろっく@calloc134 - Hono Conference 2025
Avatar for calloc134 calloc134
0
350
超高速データサイエンス
Avatar for Yusuke Matsui matsui_528
1
160
投資戦略202508
Avatar for Pragmaworks pw
0
570
アニメにおける宇宙猫ミームとその表現
Avatar for yttrium173340 yttrium173340
0
100

Featured

See All Featured
ReactJS: Keep Simple. Everything can be a component!
Avatar for Pedro Nauck pedronauck
667
130k
Optimising Largest Contentful Paint
Avatar for Harry Roberts csswizardry
37
3.5k
Fashionably flexible responsive web design (full day workshop)
Avatar for Andy Clarke malarkey
407
66k
Rails Girls Zürich Keynote
Avatar for Gregor Martynus gr2m
95
14k
Typedesign – Prime Four
Avatar for Hannes Fritz hannesfritz
42
2.8k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
Avatar for Chris Coyier chriscoyier
508
140k
RailsConf & Balkan Ruby 2019: The Past, Present, and Future of Rails at GitHub
Avatar for Eileen M. Uchitelle eileencodes
140
34k
Connecting the Dots Between Site Speed, User Experience & Your Business [WebExpo 2025]
Avatar for Tammy Everts tammyeverts
10
620
10 Git Anti Patterns You Should be Aware of
Avatar for Lemi Orhan Ergin lemiorhan
PRO
658
61k
Building Flexible Design Systems
Avatar for Yesenia Perez-Cruz yeseniaperezcruz
329
39k
CSS Pre-Processors: Stylus, Less & Sass
Avatar for Bermon Painter bermonpainter
359
30k
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
Avatar for Tammy Everts tammyeverts
55
3k

Transcript

  1. τϐοΫϞσϧʹΑΔ౷ܭతજࡏҙຯ ղੳ 3ষ ֶशΞϧΰϦζϜ(3.2.3 ~ 3.2.4) Yoshifumi Seki GunosyσʔλϚΠχϯάݚڀձ #98

    2016.01.06

  2. 3.2.3 LDAͷΪϒεαϯϓϦϯά ·ͣLDAͷશ֬཰ม਺ͷ݁߹෼෍Λߟ͑Δ ͔͜͜Βؔ܎ͷͳ͍߲Λফͯ͠αϯϓϧͷੜ੒Λߦ͏

  3. จॻ ʹ͓͚Δ ൪໨ͷજࡏม਺

  4. จॻ ʹ͓͚Δ ൪໨ͷજࡏม਺ ͱͳΔΑ͏ʹਖ਼نԽ͢Δͱ ϙΠϯτ • ෼฼΍৐਺ΛऔΓ෷ͬͯൺྫ஋ʹ͍ͯ͠ΔͷͰ, ૯࿨͕1ʹͳΔΑ͏ʹઃఆ͢Ε͹֬཰஋ʹͰ͖Δ

  5. ͷ৚݅෇͖ಠཱੑ άϥϑΟΧϧϞσϧΑΓ, ͸ ʹͷΈґଘؔ܎͕͋ Δɽ (3.29)͸͜ͷ৚݅Λຬ͍ͨͯ͠Δɽ

  6. ͷαϯϓϦϯά

  7. ͷαϯϓϦϯά Λ༻͍Δͱ • σΟϦΫϨ෼෍ͷࣄޙ෼෍ͳͷͰσΟϦΫϨ෼෍ͷܗʹͳΔΑ͏ʹ͍ͨ͠ɽ • ͳͷͰ૯৐Ͱද͢͜ͱ͕Ͱ͖ΔΑ͏ʹࣜมܗ͍ͯ͠Δ

  8. ͷαϯϓϦϯά (2.11)ΑΓ ͸Dirichelet෼෍ʹै͏ͷͰ

  9. ͷαϯϓϦϯά

  10. ͷαϯϓϦϯά Ҏ্͔Β

  11. ͷαϯϓϦϯά (2.11)ΑΓ ͸Dirichelet෼෍ʹै͏ͷͰ

  12. ΪϒεαϯϓϦϯάʹΑΔಋग़

  13. 3.2.4 LDAͷपลԽΪϒεαϯϓϦϯά ͸จॻ ͰજࡏτϐοΫ ͕ݱΕͨ୯ޠ਺ ͸શจॻʹରͯ͠τϐοΫ ͕୯ޠ ʹରͯ͠ਪఆ͞Εͨؔ਺

  14. ʹ͓͚ΔपลԽ

  15. ʹ͓͚ΔपลԽ ΑΓ

  16. ʹ͓͚ΔपลԽ σΟϦΫϨ෼෍ͳͷͰ, σΟϦΫϨ෼෍ͷظ଴஋ͱͯ͠ܭࢉ͢Δ

  17. None