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

結合トピックモデル

 結合トピックモデル

2016年3月29日に『トピックモデルによる統計的潜在意味解析』
読書会ファイナル ~佐藤一誠先生スペシャル~のLTで発表しました

2ab3dc02a9448f246bab64174b19dc1e?s=128

Kento Nozawa

March 29, 2016
Tweet

Transcript

  1. ݁߹τϐοΫϞσϧ ʰτϐοΫϞσϧʹΑΔ౷ܭతજࡏҙຯղੳʱ ಡॻձϑΝΠφϧ ~ࠤ౻Ұ੣ઌੜεϖγϟϧ~ ໺୔ ݈ਓ (@nzw0301) 2016-03-29

  2. ࣗݾ঺հ ͸͡Ί·ͯ͠ ໺୔ ݈ਓ (@nzw0301) य़͔Βஜ೾େͰM1 ڵຯ • ػցֶशɼNLPɼάϥϑɼDL

  3. ݁߹τϐοΫϞσϧ จॻσʔλͱରԠ͢Δ৘ใΛ߹Θֶͤͯश • ೔ຊޠͱӳޠ • Ϩγϐͱࡐྉ • ୯ޠͱͦͷ඼ࢺ ࢀߟɿ௨ৗͷLDAͷάϥϑΟΧϧϞσϧ 3

    D N2 N1 K 2 1 w2 i w1 i ✓ ↵ 1 z1 i z2 i 2 D N K wi ✓ ↵ zi
  4. z ͷαϯϓϦϯάࣜ • 3ষp55ʹैͬͯಋग़Մೳ • ৄ͘͠͸ http://nzw0301.github.io/2016/02/jointTopicModelsEquation • ҎԼͷ͔ࣜΒGibbs SamplingͷࣜΛٻΊΔ

    ࢀߟɿ௨ৗͷLDA 4 p(z1 d,i = k|w1 d,i = v, W1 \d,i , W2, Z1 \d,i , Z2, ↵, 1, 2) p(z1 d,i = k|w1 d,i = v, W1 \d,i , Z1 \d,i , ↵, 1)
  5. ࣜมܗͷ݁Ռ • ݁߹τϐοΫϞσϧͷαϯϓϦϯάࣜ • φ͸௨ৗͷLDAͱಉ͡ • θʹ͍ͭͯɼؚ·ΕΔ߲͕ิॿ৘ใͷ਺͚ͩ૿͑Δ ࢀߟɿ௨ৗͷLDAͷαϯϓϦϯάࣜ 5 n1

    k,v,\d,i + v P v0 (n1 k,v0,\d,i + v0 ) n1 d,k,\d,i + n2 d,k + ↵k P k0 (n1 d,k0,\d,i + n2 d,k0 + ↵k0 ) n1 k,v,\d,i + v P v0 (n1 k,v0,\d,i + v0 ) n1 d,k,\d,i + ↵k P k0 (n1 d,k0,\d,i + ↵k0 )
  6. ࣮ߦྫ • ର༁ίʔύεΛ૝ఆͨ͠؆୯ͳྫ • ݴޠ͕ҧͬͯ΋τϐοΫ͸ڞ௨ http://nzw0301.github.io/2016/02/jointTopicModelsEquation ࣮૷ɿLDAͷαϯϓϧࣜͰ࢖͏౷ܭྔΛ഑ྻʹ 6

  7. ࢀߟจݙ • ؠా ۩࣏. τϐοΫϞσϧ. ߨஊࣾ. 2015. (MLPػցֶशϓϩϑΣογϣφϧγϦʔζ). • ࠤ౻

    Ұ੣. τϐοΫϞσϧʹΑΔ౷ܭతજࡏҙຯղੳ. ίϩφࣾ. 2015. (ࣗવݴޠॲཧγ Ϧʔζ, 8). • David Mimno, Hanna M. Wallach, Jason Naradowsky, David A. Smith and Andrew McCallum. 2009. Polylingual Topic Models. in EMNLP. 7