|
|
|
- しまな きせんばる
- 5 years ago
- Views:
Transcription
1
2
3 i Google Yahoo! Amazon E
4 ii x = arg max f(x) x x X (arg max ) (f(x) ) (X ) 3
5 iii IoT (Internet of Things)
6 iv
7 v
8 vi
9 vii A Imagerous
10 viii BML Bricolage [Kumar 11] A A [Hohnhold 15]
11 ix Z CTA [ 12] A.1 x =(0, 0, 0, 0) A.2 x =(1, 0, 0, 0) A.3 x =(0, 1, 0, 0) A.4 x =(0, 0, 1, 0) A.5 x =(0, 0, 0, 1) A.6 x =(1, 2, 2, 1)
12 x A B A A A, B, C CTA (n = 300)
13 Amazon *1 *2 E *3 Google *4 Yahoo! *5 Twitter *6 Facebook *7 *8 *1 Amazon *2 * *4 Google *5 Yahoo! *6 Twitter *7 Facebook *8
14 2 1 Android Wear *9 Apple Watch * 10 Google Home * 11 Amazon Echo * 12 (Internet of Things: IoT) Amazon [Kohavi 09] Google Bing * 13 [Kohavi 14, Hohnhold 15] * 14 * 15 *9 Android Wear *10 Apple Watch *11 Google Home *12 Amazon Echo *13 Bing * html/nc html *15 27 BtoC consumer/consumer/pdf/sns_report.pdf
15 1.1 3 * 16 E * 17 *16 4 *17 (variation)
16 4 1 Bing 10 [Kohavi 14] * Family Image LEARN MORE [Crook 09] (variant) variant *18 How Obama Raised $60 Million by Running a Simple Experiment /11/29/how-obama-raised-60-million-by-running-a-simple-experiment/
17 1.1 研究の背景 5 Media variation Original Image Family Image Change Image Barack s Video Sam s Video Springfield Video Button variation 図 1.1: 2008 年アメリカ合衆国大統領選挙にて バラック オバマ氏公式ウェブサイトの実験 で用いられた写真とボタンのバリエーション ウェブサイト最適化の定義 広義 ウェブサイトを改善し より企業活動の上で望ましい状態にすること 狭義 ユーザをサンプルとした対照実験にもとづいて より望ましいユーザ行動を引き 出すウェブサイトのバリエーションを探索すること 広義のウェブサイト最適化には パフォーマンス最適化や検索エンジン最適化も含まれる し かし これらに対する有効な手法は検索エンジンの仕様やその時々で用いることができる技術 に大きく左右されるため 解法を一般化することは難しい 一方 狭義のウェブサイト最適化 は人間の感覚 知覚 認知の過程をブラックボックスとして扱い そこから生まれる反応を最 適化する試みである したがって 人間であるユーザと対峙するウェブサイトを作る限り有用 なものであると考えられる そこで 本研究では特に狭義のウェブサイト最適化を研究対象と して取り上げることにする
18 Google * 19 [Tang 10] [Wolpert 97] *19 SimilarWeb ( )
19
20 x = arg max f(x) x x X X f(x) x arg max f(x) X 2 3 arg max
21 : 4 f(x)
22 X 6 7
23 [Kohavi 11]
24 12 2 Fisher [ 11]
25 [ 93] [Womack 90] [Spear 04] ATM NATO [ 92]
26 : 2.1 [Sommerville 11]
27 : [ 88] [Leffingwell 07]
28 B2B *1 B2C *2 B2B *3 B2C *1 Business to Business *2 Business to Consumer *3
29 Eric Ries (Build) (Measure) (Learn) 3 BML [Ries 11] BML 2.3 BML BML 2.4 Eric Ries *4 *4 Eric Ries Lean Startup Presentation For Web 2.0 Expo April ADisciplined Approach To Imagining, Designing, And Building New Products. net/startuplessonslearned/eric-ries-lean-startup-presentation-for-web-20-expo-april-
30 : BML ATM a-disciplined-approach-to-imagining-designing-and-building-new-products
31 (a) (b) (c) 2.4:
32 [ 12]
33 (Interactive Evolutionary Computation) [Takagi 01] [Lewis 08] [ 00] [Takagi 01]
34 [ 15] x X f(x) x x = arg max f(x) x X 3
35 f f x f(x) [ 16]
36 :,, [ 95] [ 12] [ 15] 2.3.1
37 N x {0, 1} N w R w x *5 D = {(x n,y n ) n =1,...,N} x n f(w x n ) y n l(x n,y n, w) L(D, w) = 1 N N n=1 l(x n,y n, w) w w L(D, w) w t+1 = w t + η L(D, w t ) w *6 w t t w η *5 S more, some more *6 L(D, w t ) L(D, w t )
38 26 2 t =1,...,T x t wt x t x w Follow-The-Leader (FTL) Follow-the-Regularized-Leader (FTRL) [ 15] D L(D, w) w (x n,y n ) D (x n,y n ) l(x n,y n, w) w t+1 = w t + η l(x n,y n, w t ) n (x n,y n ) l(x n,y n, w t ) l(x n,y n, w t ) f(x) =w x f l(x n,y n, w t )
39 x x 1, x 2, x 3 3 x 3 V 3 = {,,, }
40 28 2 n x R n y p(y x) x maximize E[y x] (2.1) subject to 0 x i L i x i Z, i=1...n y p(y x) y R L i i y f(y) y
41 x y [Hoffman 14]
42 [ 00] [ 04] 2.3.1
43 τ
44 32 2 [ 03] (Bayesian Optimization Algorithm: BOA) [Bengoetxea 02] [ 16]
45 [ 16] T R(T ) E[R(T )] = E [ max i T r i (t) t=1 ] T r i (t) t=1 r i (t) t i ε-greedy ε-greedy 0 <ε< 1 ε 1 ε Softmax ε-greedy Softmax Softmax
46 34 2 X x X E[y x] x p(x) p(x) = exp(e[y x]/τ) x X exp(e[y x]/τ) τ τ x p(x) 1/ X 1 UCB UCB (Upper Confidence Bound) UCB E[y x] x X N x x UCB U x U x = E[y x]+ 2 log( x X N x) N x UCB x UCB (optimism in face of uncertainty)
47 [Birattari 02, Heidrich-Meisner 09] x [µ lower x,µ upper x ] x loser µ upper x loser < min x X\{x loser } µlower x x winner µ lower x winner < max x X\{x winner } µupper x x i R d d LinUCB Lin-UCB UCB x i µ i θ µ i = θ T x i x x a x b x a x b
48 36 2 Li Yahoo! LinUCB [Li 10a] Li u t u t x i,t = x i u t f(x i, u t ) f(x i, u t ) x i u t f(x i, u t ) ( x θx t = arg max i θ u ) T ( xi u t ) = arg max θx T x i i GLM-UCB [Filippi 10] f(x) c >0 s.t. f(x ) f(x) c x x x, x X
49 [Srinivas 10] (PI: Probability of Improvement) (EI: Expected Improvement) [Brochu 10] GP-UCB (Gaussian Process Upper Confidence Bound) GP-UCB GP-UCB t x t µ t (x) σ t (x) x t+1 = arg max µ t (x)+ βσ t (x) x X β β RBF (Radial Basis Function) k(x, x ) = exp( x x /σ 2 ) k(x, x )=x T x UCB LinUCB UCB
50 Aslliani [Asllani 07] [Park 07] [Penalver 98] [ 00]
51 Yahoo! Li [Li 10a] [Lu 10, Li 10b] [Lomas 16] [Kohavi 09] 2.3.4
52 Google Bing Yahoo! Deng 3 [Deng 13] 1% Google Bing t [Belle 98] = µ µ /σ µ µ σ
53 [Deng 13, Guo 15] Borodovsky [Borodovsky 11] WAU: Weekly Active Users MAU: Monthly Active Users
54 42 2 Google Bing Google Play *7 App Store *8 [Dmitriev 16] *7 Google Play *8 itunes Store
55 [Kohavi 09] [Guo 15] Hohnhold Google [Hohnhold 15] Bing [Dmitriev 16] Yahoo! Agarwal Yahoo! [Agarwal 11] Yahoo!
56 Webpage Segmentation (Information Retrieval) [Nie 09] Webpage Segmentation HTML *9 DOM * 10 DOM HTML DOM DOM HTML <table> <ul> *9 HyperText Markup Language: *10 Document Object Model: HTML XML
57 [Lin 02] [Kohlschütter 08] [Joshi 09] HTML HTML5 * 11 CSS * 12 DOM [Kumar 11] VIPS [Cai 03] VIPS DOM Zone Tree Model [Le 06] HTML CSS *11 HTML 5 HTML CSS *12 Cascading Style Sheets:
![[Kumar 11]](/docs-images/93/114409253/images/58-2.jpg "Web Page")
![Thompson 09]](/docs-images/93/114409253/images/58-5.jpg "HTML")
58 46 } 2 2.5: Bricolage [Kumar 11] Web Page Segmention [Bajwa 06, Thompson 09] HTML [Sengamedu 08] Bricolage [Kumar 11] 2.5
59 [Kohavi 14] Kohavi Microsoft A/B [Kohavi 12] [Kohavi 09] * HTML *13 [Kohavi 09] Traffic splitting, Page rewriting, Client-side assignment, Serverside assignment Traffic splitting Page rewriting Server-side assignment
60 request 2. assign & rewrite 3. response User Browser Web Server 2.6: HTML Javascript 2.7 HTML HTML Javascript Javascript Javascript HTML
61 assign & response 5. rewrite Optimization Server 3. request (AJAX) 1. request 2. response User Browser Web Server 2.7: GUI Apptimize * 14 Software Development Kit (SDK) *14
62 50 2 (GUI) Apptimize Conductrics * 15 GUI Google Analytics Content Experiments * Google URL [Russo 16] Google Optimize * Google Google Optimize 360 Google Optimize Google Analytics Content Experiments Google Optimize Conductrics Optimizely Javascript CSS URL A/B GUI The Grid * 18 *15 *16 *17 *18
63 Kaizen Platform * 19 Kaizen Platform Kaizen Platform Optimizely * 20 1 Dan Siroker GUI Taplytics * 21 GUI Visual Website Optimizer * 22 GUI AI * 23 Google Analytics WACUL * 24 *19 *20 *21 *22 *23 *24
64 [Wolpert 97]
65 [ 00] 2.1 a n na n a 3.1
66 :
67 A/B A/B (Control) (Treatment) A/B/n A/B A/B (Multivariate Testing) [Ash 12] (Full Factorial Test Design) (Fractional Factorial Test Design)
68 56 3 k O(c k ) O(k) A/B x 0 x x i V i X A/B x j A/B
69 : A/B f(x) =g 1 (x 1 )+ + g m (x m )+a g i (x i ) x i f(x) a E[y x] x A/B
70 58 3 A/B Algorithm 1 t α 1 β t N T TEST N 1 2 α 1 β [Cohen 88] x X 3 x x X 5 N t 6 N 1 7 5,6,7 N n 3.2.2
71 Algorithm 1 Require: α as the significance level. Require: 1 β as the power. Require: as the effect size. Require: N as the sample size bound. Require: X as the set of solutions. 1: Set n 0 as the number of consumed samples. 2: Set N 1 N T TEST (α, β, ) 3: x RandomChoice(X) 4: while n<n do 5: x Neighbor(x,X) 6: x Compare(x, x,n 1 ) 7: n n + N 1 8: return x as the optimal solution. LALS (Linear Assumption Local Search) 3.1 LALS
72 60 3 Algorithm 2 Algorithm 1 LALS t N 2 N 1 2, 3 N 1 N ANOV A k = m i=1 l i [Cohen 88] Algorithm 2 LALS Require: α as the significance level. Require: 1 β as the power. Require: as the effect size. Require: N as the sample size bound. Require: X as the set of solutions. 1: Set n 0 as the number of consumed samples. 2: Set N 1 N T TEST (α, β, ) 3: Set N 2 N ANOV A (α, β,,x) 4: x LinearAssumption(X, N 2 ) 5: n n + N 2 6: while n<n do 7: x Neighbor(x,X) 8: x Compare(x, x,n 1 ) 9: n n + N 1 10: return x as the optimal solution. LALS N 1 4, 5 X N 1 N 1
73 LALS LALS+ LALS+ Algorithm 3 LALS Algorithm 2 LALS+ LALS N 1,N 2 2, 3 LALS N 1 LALS 4 N 1 n <N 1 5 N 2
74 62 3 Algorithm 3 LALS+ Require: α as the significance level. Require: 1 β as the power. Require: as the effect size. Require: N as the sample size bound. Require: X as the set of solutions. 1: Set n 0 as the number of consumed samples. 2: Set N 1 N T TEST (α, β, ) 3: Set N 2 N ANOV A (α, β,,x) 4: x,n DynamicLinearAssumption(X, N 2 ) 5: n n + n n N 2 6: while n<n do 7: x Neighbor(x,X) 8: x,n DynamicCompare(x, x,n 1 ) 9: n n + n n N 1 10: return x as the optimal solution BF (Brute Force) LA
75 : BF N/A LA N/A LS LALS LALS+ + + LS A/B LS t N T TEST LALS, LALS f 1 (x) f 2 (x) f 1 (x) =x 1 + x 2 + x 3 x 4 x 5 x 6 + N(0, 1) f 2 (x) =x 1 + x 2 + x 3 x 4 x 5 x 6 x 1 x 2 + N(0, 1) m =6 x i V = {0, 1, 2} N(0, 1) µ =0 σ =1 f 1 (x) x i =0 x i =2 f 2 (x) x 1 x 2 f 1 (x)
76 : N 1 f 1 (x) 500 (N <N 1 ) 2 f 1 (x) 2000 (N >N 1 ) 3 f 2 (x) 500 (N <N 1 ) 4 f 2 (x) 2000 (N >N 1 ) (x 1,x 2 )=(2, 2) f 2 (x) g 1 (x 1 )g 2 (x 2 ) N N 1 N N 1 LALS N >N (= 729) N x α =0.05, 1 β =0.8, =0.2 N 1 = 550,N 2 = 400 [Auer 02] N BF LS LA LALS LALS
77 : BF LA LS LALS LALS N LALS *1 A A A A 1.17% B, C *1
78 66 第 3 章 線形性仮定と局所探索を組み合わせた探索手法の提案 図 3.2: ウェブサイト A の相関図ページのスクリーンショット 施策による効果を 3.5 にまとめる 施策 A は人物の相関図を コ の字のように取り囲むようにしてバナー広告を配置する施 策である バナー広告の配置を工夫した単純な施策であるが 大きくクリック率を向上するこ とが確かめられた 施策 B は相関図内に表示される人物をカテゴリ別にフィルタリングでき る機能を実装した施策である 表示される情報の整理ができるようになることで使いやすくな り よりユーザの利用度合いが増え 広告収益も向上することを狙った施策であったが 大き なクリック率の向上は見られなかった 施策 C は時間経過する毎に相関図内に表示される人 物の数が増えるというものである 提供する情報の量を段階的に増やすことで ユーザの飽き を防ぐ狙いであったが こちらの大きな改善は見られなかった オリジナルのバリエーション と各施策を施したバリエーションの間でクリック率のカイ二乗検定を行なった結果 有意水準 0.01 で有意差があることが認められた ここで 施策 i を行うか否かの二値 (0, 1) をとる変数 xi を用いると ある解 x を訪問者に
79 : A (95% ) A 9, % (7.04%, 8.09%) B 14, % (5.36%, 6.10%) C 5, % (4.50%, 5.62%) 71, % (6.22%, 6.58%) q(x) = x A x B x C f(x) x q(x) 1 1 q(x) 0 α =0.05, 1 β =0.8, =0.05 N 1 = 5500,N 2 = BF LA LS LA LALS BF LS LALS+ LALS
80 : A Imagerous *2 B *2 Imagerous
81 : B x 1 0px, 5px x 2 0px, 5px, 10px x 3 100px, 200px, 300px x 4, B x =(0, 2, 1, 0) { : 0px, : 10px, : 200px, : } A LS LALS 2
82 request Javascript 4. response 7. track events 8. send event log 1. load 5. modify 3. assign a variation Optimization Server User Browser 6. render HTML 3.4: LALS A/B LS LALS+ LALS α = β =0.8 =0.3 N 1 = 120,N 2 = B HTML Javascript Javascript DOM CSS Javascript HTML Javascript Javascript
83 : LS LALS 0 (1, 1, 2, 1) N/A N/A 288 (1, 2, 2, 1) (1, 2, 2, 1) N/A N/A N/A 122 (1, 0, 0, 0) Inf. 376 (0, 0, 0, 0) (0, 0, 0, 0) B 3.7 LALS LS t 0.01 LS LALS N X = m i=1 l i m N 2 = m i=1 l i
84 72 3 [Cohen 88] A B 2 A B
85 [ 13] 3.4.3
86 LinUCB RBF A/B A/B 2.6 Google Optimize A/B LinUCB GP-UCB
87 GP-UCB 3.5
88 Deng [Deng 13] Bing
89 Hohnhold 3 [Hohnhold 15] Google
90 78 4 y x x 1 3 x 2 z 2 z 1 p(y x), q(z x) Website Action x Reward y Feedback z User 4.1: 4.1 GP-UCB [Srinivas 10]
91 Deng [Deng 13] Bing
92 80 4 Y Y = E[Y ] Var[Y ] = Var[Y ]/n n K k Y k Ŷ = K k=1 w ky k w k k k n k w k = n k /n Var[Y ]= K k=1 w k n σ2 k + K k=1 w k n (µ k Y ) 2 K k=1 w k n σ2 k = Var[Ŷ ] E[Ŷ ] µ k,σ 2 k k [Deng 13] Y Z
93 y µ = E[y] y z E[z] y y = y + c(e[z] z) c y Var[y ] = Var[y]+c 2 Var[z] 2cCov[y, z] (4.1) Var[y] y Cov[y, z] y z (4.1) c Var[y ] c c =Cov[y, z]/var[z] Var[y ] Var[y ] = Var[y] (1 Cov[y, ) z]2 = Var[y](1 ρ 2 ) Var[y]Var[z] ρ y z y ρ z y y y z Deng [Dmitriev 16]
94 82 4 [Hohnhold 15] Hohnhold Google Google Hohnhold φ 1 φ 2 φ 3 3 φ 2 φ 3 Google 100
95 RPM Short-term Long-term Treatment duration in days 4.2: [Hohnhold 15] φ 2 φ RPM *1 Hohnhold *1 RPM Revenue Per Mille 1000 Mille 1000
96
97 x x ω y y z (a) (b) 4.3: u U x y p(y x) p(y x) E[y x] E a u y z x z ω x p(x ω) ω z p(z ω) x z y p(y x, z) 4.3b ω x
98 86 4 z y x z y x X z ω p(x ω), p(z ω) x z y p(y x, z) x x = arg max E[y x, z] s.t. x p(x ω), z p(z ω), y p(y x, z) x X z x z 2.3.2
99 Chosen 4.4: GP-UCB [Brochu 10] 4.4 Algorithm 4 x t w y t z t Z y µ(z) σ(z) z Z UCB UCB z z x = arg min(z E[z x]) 2 x t+1 x X w 3 X = {x A,x B,x C } E[y x] x
100 88 4 Algorithm 4 Require: X as the set of variations. Require: Z as the input space. Require: α as the GP prior parameter. Require: β as the confidence parameter. 1: Set µ 0 (z) 0, z Z. 2: Set σ 0 (z) N(µ 0 (z),α 1 I). 3: for t =1, 2,... do 4: Choose z = arg max µ t 1 (z)+ βσ t 1 (z) z Z 5: Choose x t = arg min(z E[z x]) 2 x X 6: Sample y t and z t by showing the variation x t to user u t. 7: Update µ t and σ t using Gaussian Process with parameter α. 8: return x t as the optimal solution. w u w y z w x t u t y t z t N (µ(z),σ(z)) z z x B = arg min(z E[z x]) 2 x X u t+1 x B X Z α β α 0 β UCB UCB β
101 α 0 β UCB UCB Algorithm 5 Algorithm 5 UCB Require: X as the set of variations. 1: Set N x 0, x X as the number of observations. 2: for t =1, 2,... do 3: Choose x t = arg max E[y x]+ x X 2log x X N x N x 4: Sample y t by showing the variation x t to user u t. 5: N xt N xt +1 6: return x t as the optimal solution.
102 A y A z z 1 z 2 A A 3 X = {x 1,x 2,x 3 } x 1 x 2 x 3 A 3.2
103 : A X y z 1 z 2 x 1 Bernouli(0.0534) Bernouli(0.403) Gamma(1.46, 21.6) x 2 Bernouli(0.0716) Bernouli(0.417) Gamma(1.24, 22.9) x 3 Bernouli(0.0397) Bernouli(0.412) Gamma(1.48, 21.0) 4.1 y z 1,z 1 y z 1 p Bernouli(p) z 2 0 k θ Gamma(k, θ) x 2 x 2 α = 10,β = UCB
104 baseline SroSosed 0.8 Average Accuracy amSle 4.5: Z y Profile z 1 Time on page z 2 Click-through rate E[y] z 1,z (Profile, Time on page) = (1, 0) 4.1 x 2 (0.417 >
105 baseline SrRSRsed 100 Cumulative 5ewards amSle 4.6: max(0.403, 0.412)) (1.24 < min(1.46, 1.48)) x p(z x) 4.4.2
106 : Z A
107 : A X y z 1 z 2 x 4 Bernouli(0.0551) Bernouli(0.4113) Gamma(1.511, ) x 5 Bernouli(0.0683) Bernouli(0.3764) Gamma(1.341, ) x 6 Bernouli(0.0594) Bernouli(0.3910) Gamma(1.189, ) x 7 Bernouli(0.0656) Bernouli(0.3858) Gamma(1.318, ) x 8 Bernouli(0.0846) Bernouli(0.3917) Gamma(1.345, ) X = {x 4,x 5,x 6,x 7,x 8 } y z 1,z (1, 2, 3 ) x 4 = (10, 2, 0),x 5 =(3, 3, 3),x 6 =(8, 3, 0),x 7 =(6, 4, 0) x A B UCB
108 96 4 Z GP-UCB x X y z 1,z 2 GP-UCB A B / X = B A B B A t 0.05 A 500 B
109 : 4.5
110 98 4 Google Analytics [ 08]
111 KPI 4.6
112 100 4
113
![1: [ 08] [](/docs-images/93/114409253/images/114-1.jpg "09] 5.")
![1 *1 [ 13]](/docs-images/93/114409253/images/114-2.jpg "[Diaz 95]")
114 102 5 (a) (b) (c) 5.1: [ 08] [ 09] 5.1 *1 [ 13] [Diaz 95] *1 3 (volume) (pixel) (voxel)
115 [ 15] 5.2 AI
116 : Call-To-Action (CTA) CTA E 5.5 CTA CTA *2 CTA *2
117 CTA Recurrent Neural Network (RNN) 3 [Farhadi 10] [ 16] [ 09, 14]
118 106 5 RNN [Sutskever 14] [Vinyals 15] CTA CTA 5.1 [Kohavi 14, Siroker 13] CTA CTA CTA
119 CTA [Siroker 13] CTA 5.3 CTA w W CTA Contents CTA Component Contents Bag-of-Word Component CTA x y D =(X, Y ) X Y X f k X k w Contents f x y CTA Contents CTA E
120 108 5 X = Contents(W) Y = Label(W) X Y f(x) W X k Y x = Contents(w ) x k = f(x ) x k y w 5.3: CTA CTA Contents Contents Contents w T w 0, 1 Bag-of-Word w m x m w W T w x n w W n m X X
121 X Latent Semantic Analysis (LSA)[Landauer 98] LSA n m M M = UΣV T M Σ M U V U V U, Σ,V k +1 U k, Σ k,v k M k = U k Σ k n k X U, Σ,V X k CTA w Contents Bag-of-Word x LSA V x k = V T k x k X k cos(x k, x k ) 5.4 K
122 110 5 A B C CTA (0.083) (0.090) (0.096) (0.099) (0.100) (0.103) (0.110) (0.114) (0.116) (0.125) (0.092) (0.118) (0.118) (0.126) (0.127) 5.1: A, B, C CTA AI Google Analytics
123 CTA CTA CTA W Mecab *3 T w CTA CTA CTA CTA k = 30 LSA A B E C 5.4 CTA A *3 Mecab
124 第 5 章 ウェブページの文脈を用いた有望な解候補の生成方法 112 (a) ウェブサイト A (b) ウェブサイト B (c) ウェブサイト C 図 5.4: 実験対象のウェブサイトのスクリーンショット きるウェブサービスであり アカウントの登録が主な目的である 提案手法を用いることに よって サインアップ や 無料会員登録はこちら 無料登録はこちら など 目的との親 和性が高いラベルが候補として生成されている ウェブサイト B は家事代行サービスは家事代行サービスであり お問い合わせや資料請求 が主な目的となっている 会員登録 や ログイン など サービスの申し込みとは関係のな いラベル候補も生成されてしまっているが それに続く候補として お問い合わせ 資料請 求 といった ウェブサイトの目的に適ったラベルを生成することができている また 無料 おためし体験 といった 訴求力が高いと考えられるラベルの生成にも成功している なお ここでは特定のサービス名が特定できるラベルは省略している ウェブサイト C はペットにまつわる商品に特化した E コマースサイトであり ログインや アカウントの作成 さらにカートの閲覧が主な目的として設定されている これまでのウェブ サイトとは異なり カート や オーダーフォーム といった 購買行動に関する候補が並ん でいる このことから 提案手法が E コマースサイトという特性を捉えてラベルを出し分け ていることが伺える 交差検証による定量的評価 交差検証による評価では 事例ウェブサイトを K = 100 個に分割して訓練データおよびテ ストデータを生成した 大きな割合をテストデータに分けると学習したモデルのパフォーマン スが落ちてしまうため ある程度 K が大きい分割方法を採用した
125 CTA CTA [Le 14] t α = CTA
126 *4 CTA % *4
127 CTA % 34.7% 17.7% 21.0% 46.7% 15.0% 86.3% 89.0% 62.3% 48.7% 56.8% 31.7% 85.0% 89.3% 67.0% 85.0% 90.3% 64.3% 63.7% 70.0% 36.0% 77.9% 83.2% 55.8% 5.2: (n = 300) CTA
128 116 5 AI CTA AI CTA HTML CTA CTA CTA E CTA CTA
129 CTA CTA Expedia *5 Agoda *6 HTML CSS Javascript *5 Expedia *6 Agoda
130 118 5 Feature Location [Dit 13] 5.6 CTA CTA CTA
131
132
133 : 6.1 x y x y x u c x =(x, u, c) Google u
134 122 6 *1 y 4 y [Feliot 17] 3 RBF 4 *1 Google Web Forum
135 c Bag-of-Word c x z x
136 [ 16] V 6.3.2
137 Twitter [Jansen 09] [Chamlertwat 12] E [Moore 02] E
138 126 6 Google Music *2 Apple Music *3 Google Cloud Platform *4 Amazon Web Services *5 *6 *2 Google Music *3 Apple Music *4 Google Cloud Platform *5 Amazon Web Services *6
139 Bing [Kohavi 12]
140 A/B
141 A B *7 4 *7 pr e totype
142 [ 12] IHIP *8 [Edvardsson 05] *8 inseparability heterogeneity intangibility perishability
![の考察 5 6.4 ウェブサイト以外への応用可能性 131 ᗈ (A)స ᥦ౪ (D)స せồ ㄆ ๓ ᚋ ᚋ ഛ䠄タィ䠅 䝣䜱䞊䝗䝞䝑䜽 ᣦ స (B)స ᐇ య ᰂ 䝅䝇䝔䝮 ᚋ ኚ ഛ ᅉᯝ㐃㙐 (C) స ᑐ ๓ ᚋ ኚ ቃ 図 6.2: 住田らによる作用モデル 図は [住田 12] より引用 図2 サービスと製品機能の比較のモデル (基本モデル ᗈ 図 6.](/docs-images/93/114409253/images/143-0.jpg "2 に 住田らが提案する作用モデルの概要を示す 作用モデルでは サービスおよび製 (D)㢳ᐈ㻌 㻌 㻌 㻌 㻌 (A)䝬䝑䝃䞊䝆ᗑ ㄆ 品にまつわるステークホルダーとして (A) 作用提供者 (B) 作用実行主体 (C) 作用対象 (D) のモデル例 ANY ഛ 作用要求者の 4 者を仮定する ここで (B) 作用実行主体は求められる作用を提供するサービ ഛ 䝣䜱䞊䝗䝞䝑䜽 (C)")
143 の考察 ウェブサイト以外への応用可能性 131 ᗈ (A)స ᥦ౪ (D)స せồ ㄆ ๓ ᚋ ᚋ ഛ䠄タィ䠅 䝣䜱䞊䝗䝞䝑䜽 ᣦ స (B)స ᐇ య ᰂ 䝅䝇䝔䝮 ᚋ ኚ ഛ ᅉᯝ㐃㙐 (C) స ᑐ ๓ ᚋ ኚ ቃ 図 6.2: 住田らによる作用モデル 図は [住田 12] より引用 図2 サービスと製品機能の比較のモデル (基本モデル ᗈ 図 6.2 に 住田らが提案する作用モデルの概要を示す 作用モデルでは サービスおよび製 (D)㢳ᐈ㻌 㻌 㻌 㻌 㻌 (A)䝬䝑䝃䞊䝆ᗑ ㄆ 品にまつわるステークホルダーとして (A) 作用提供者 (B) 作用実行主体 (C) 作用対象 (D) のモデル例 ANY ഛ 作用要求者の 4 者を仮定する ここで (B) 作用実行主体は求められる作用を提供するサービ ഛ 䝣䜱䞊䝗䝞䝑䜽 (C) 作用対象に対して作用を提供する 教育サービスのよ るモデルスもしくは製品であり 顧客である ᅉᯝ㐃㙐 うに 作用の提供を求める主体 たとえば母親 と作用を与える対象 たとえば子ども が異 䝸䝷䝑䜽䝇䛥䛫䜛 (C) 㢳ᐈ 念を捉えるなることがあるため 作用対象に加えて (D) 作用要求者の存在を仮定している 作用要求者と (B)䝬䝑䝃䞊䝆ᖌ ᗑ 比較するこ作用対象は同一の場合もある 一方 作用を実際に提供する従業員を雇用するサービス運営者 ᰂ (a)䝃䞊䝡䝇 や 製品を製造する製造者の存在を表現するため (A) 作用提供者の存在を仮定している ビスと製品 ᗈ このモデルではサービスと製品はともに作用を提供する主体であり 作用だけに着目すれば (A) 䝯䞊䜹䞊 (D) 䝴䞊䝄 ㄆ 下で両者の 本質的な違いはない しかし その二つの間を分かつ性質として住田らは 作用要求者と作用 ANY サービス実行主体の間の所有関係 を挙げている たとえば サービスであるマッサージ師によるマッ ഛ 述べる 䝣䜱䞊䝗䝞䝑䜽 ഛ ᅉᯝ㐃㙐 サージと 製品であるマッサージチェアを例に挙げて考えてみる サービスの場合は マッ 䝸䝷䝑䜽䝇䛥䛫䜛 サージ師が作用実行主体となって顧客の筋肉をほぐすという作用を提供するが 顧客である作 (C) 䝴䞊䝄 (B)䝬䝑䝃䞊䝆 用要求者とマッサージ師の間に所有関係は無い あくまでマッサージ師は経営者である作用提 䝏䜵䜰 ᰂ 㒊ᒇ 供者との雇用関係にある 一方で製品の場合は 作用実行主体であるマッサージチェアと作用 を特殊化 (b)〇ရᶵ 要求者である顧客の間に所有関係が発生する このとき 作用提供者と作用実行主体との直接 ぶ 図 1 (a) 察 [笹島 96, 図3 サービスと製品機能の全体システムのモデリング例 える状態変 これを目 次に この抽象作用モデルを含み 更にその抽象作用
144 : N/A 6.1
145 *9 CD DVD IoT ID *9
146 134 6 * 10 * IoT *10 Tesla *11 Tesla s Over-the-Air Fix: Best Example Yet of the Internet of Things? WIRED wired.com/insights/2014/02/teslas-air-fix-best-example-yet-internet-things/
147
148
149
150 138 A Imagerous A Imagerous Imagerous ( B) 3.6 x x =(1, 1, 2, 0) { : 5px, : 5px, : 300px, : }
151 139 図 A.1: x = (0, 0, 0, 0) 図 A.2: x = (1, 0, 0, 0) 図 A.3: x = (0, 1, 0, 0) 図 A.4: x = (0, 0, 1, 0) 図 A.5: x = (0, 0, 0, 1) 図 A.6: x = (1, 2, 2, 1)
152 140 Wacul
153 141 Google Hai Kyung Min 1 Nicholas Walsh
154 142 [Agarwal 11] Agarwal, D., Chen, B.-C., Elango, P., and Wang, X.: Click shaping to optimize multiple objectives, in Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2011) [Ash 12] Ash, T., Ginty, M., and Page, R.: Landing Page Optimization: The Definitive Guide to Testing and Tuning for Conversions, ITPro collection, Wiley (2012) [Asllani 07] Asllani, A. and Lari, A.: Using genetic algorithm for dynamic and multiple criteria web-site optimizations, European journal of operational research, Vol. 176, No. 3, pp (2007) [Auer 02] Auer, P., Cesa-Bianchi, N., and Fischer, P.: Finite-time analysis of the multiarmed bandit problem, Machine learning, Vol. 47, No. 2-3, pp (2002) [Bajwa 06] Bajwa, I. S., Siddique, I., and Choudhary, M.: Web layout mining (wlm): a new paradigm for intelligent web layout design, in ITI fourth International Conference on Information and Communications Technology (2006) [Belle 98] Belle, van G. and Millard, S. P.: STRUTS: Statistical rules of thumb, Departments of Environmental Health and Biostatistics, University of Washington (1998) [Bengoetxea 02] Bengoetxea, E.: Inexact graph matching using estimation of distribution algorithms, PhD thesis, Ecole Nationale Supérieure des Télécommunications, Paris (2002) [Birattari 02] Birattari, M., Stützle, T., Paquete, L., and Varrentrapp, K.: A racing algorithm for configuring metaheuristics, in Proceedings of the fourth Annual Conference on Genetic and Evolutionary Computation, pp , Morgan Kaufmann Publishers
155 143 Inc. (2002) [Borodovsky 11] Borodovsky, S. and Rosset, S.: A/B testing at SweetIM: The importance of proper statistical analysis, in Proceedings of the 2011 IEEE 11th International Conference on Data Mining Workshops, pp , IEEE (2011) [Brochu 10] Brochu, E., Cora, V. M., and De Freitas, N.: A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning, arxiv preprint arxiv: (2010) [Cai 03] Cai, D., Yu, S., Wen, J.-R., and Ma, W.-Y.: Extracting content structure for web pages based on visual representation, in Web Technologies and Applications: fifth Asia-Pacific Web Conference, pp , Springer (2003) [Chamlertwat 12] Chamlertwat, W., Bhattarakosol, P., Rungkasiri, T., and Haruechaiyasak, C.: Discovering consumer insight from twitter via sentiment analysis, J. UCS, Vol. 18, No. 8, pp (2012) [Cohen 88] Cohen, J.: Statistical Power Analysis for the Behavioral Sciences, Psychology Press (1988) [Crook 09] Crook, T., Frasca, B., Kohavi, R., and Longbotham, R.: Seven pitfalls to avoid when running controlled experiments on the web, in Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2009) [Deng 13] Deng, A., Xu, Y., Kohavi, R., and Walker, T.: Improving the sensitivity of online controlled experiments by utilizing pre-experiment data, in Proceedings of the sixth ACM International Conference on Web Search and Data Mining, pp , ACM (2013) [Diaz 95] Diaz, A. and Sigmund, O.: Checkerboard patterns in layout optimization, Structural and Multidisciplinary Optimization, Vol. 10, No. 1, pp (1995) [Dit 13] Dit, B., Revelle, M., Gethers, M., and Poshyvanyk, D.: Feature location in source code: a taxonomy and survey, Journal of Software: Evolution and Process, Vol. 25, No. 1, pp (2013)
156 144 [Dmitriev 16] Dmitriev, P., Frasca, B., Gupta, S., Kohavi, R., and Vaz, G.: Pitfalls of long-term online controlled experiments, in IEEE International Conference on Big Data, pp , IEEE (2016) [Edvardsson 05] Edvardsson, B., Gustafsson, A., and Roos, I.: Service portraits in service research: a critical review, International Journal of Service Industry Management, Vol. 16, No. 1, pp (2005) [Farhadi 10] Farhadi, A., Hejrati, M., Sadeghi, M. A., Young, P., Rashtchian, C., Hockenmaier, J., and Forsyth, D.: Every picture tells a story: Generating sentences from images, in Computer Vision ECCV 2010, pp , Springer (2010) [Feliot 17] Feliot, P., Bect, J., and Vazquez, E.: A Bayesian approach to constrained single-and multi-objective optimization, Journal of Global Optimization, Vol. 67, No. 1-2, pp (2017) [Filippi 10] Filippi, S., Cappe, O., Garivier, A., and Szepesvári, C.: Parametric bandits: The generalized linear case, in Advances in Neural Information Processing Systems, pp (2010) [Guo 15] Guo, Y. and Deng, A.: Flexible Online Repeated Measures Experiment, arxiv preprint arxiv: (2015) [Heidrich-Meisner 09] Heidrich-Meisner, V. and Igel, C.: Hoeffding and Bernstein races for selecting policies in evolutionary direct policy search, in Proceedings of the 26th Annual International Conference on Machine Learning, pp (2009) [Hoffman 14] Hoffman, M., Shahriari, B., and Freitas, N.: On correlation and budget constraints in model-based bandit optimization with application to automatic machine learning, in Artificial Intelligence and Statistics, pp (2014) [Hohnhold 15] Hohnhold, H., O Brien, D., and Tang, D.: Focusing on the long-term: It s good for users and business, in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2015) [Jansen 09] Jansen, B. J., Zhang, M., Sobel, K., and Chowdury, A.: Twitter power: Tweets as electronic word of mouth, Journal of the American society for information
157 145 science and technology, Vol. 60, No. 11, pp (2009) [Joshi 09] Joshi, P. M. and Liu, S.: Web document text and images extraction using DOM analysis and natural language processing, in Proceedings of the ninth ACM Symposium on Document Engineering, pp , ACM (2009) [Kohavi 09] Kohavi, R., Longbotham, R., Sommerfield, D., and Henne, R. M.: Controlled experiments on the web: survey and practical guide, Data Mining and Knowledge Discovery, Vol. 18, No. 1, pp (2009) [Kohavi 11] Kohavi, R. and Longbotham, R.: Unexpected results in online controlled experiments, ACM SIGKDD Explorations Newsletter, Vol. 12, No. 2, pp (2011) [Kohavi 12] Kohavi, R., Deng, A., Frasca, B., Longbotham, R., Walker, T., and Xu, Y.: Trustworthy online controlled experiments: Five puzzling outcomes explained, in Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2012) [Kohavi 14] Kohavi, R., Deng, A., Longbotham, R., and Xu, Y.: Seven rules of thumb for web site experimenters, in Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2014) [Kohlschütter 08] Kohlschütter, C. and Nejdl, W.: A densitometric approach to web page segmentation, in Proceedings of the 17th ACM Conference on Information and Knowledge Management, pp , ACM (2008) [Kumar 11] Kumar, R., Talton, J. O., Ahmad, S., and Klemmer, S. R.: Bricolage: example-based retargeting for web design, in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp , ACM (2011) [Landauer 98] Landauer, T. K., Foltz, P. W., and Laham, D.: An introduction to latent semantic analysis, Discourse processes, Vol. 25, No. 2-3, pp (1998) [Le 06] Le, D., Thoma, G. R., and Zou, J.: Combining DOM tree and geometric layout analysis for online medical journal article segmentation, in Proceedings of the sixth ACM/IEEE-CS Joint Conference on Digital Libraries, pp , IEEE (2006) [Le 14] Le, Q. V. and Mikolov, T.: Distributed Representations of Sentences and Doc-
158 146 uments., in Proceedings of the 31st International Conference on Machine Learning, Vol. 14, pp (2014) [Leffingwell 07] Leffingwell, D.: Scaling software agility: best practices for large enterprises, Pearson Education (2007) [Lewis 08] Lewis, M.: Evolutionary visual art and design, in The Art of Artificial Evolution, pp. 3 37, Springer (2008) [Li 10a] Li, L., Chu, W., Langford, J., and Schapire, R. E.: A contextual-bandit approach to personalized news article recommendation, in Proceedings of the 19th International Conference on World Wide Web, pp , ACM (2010) [Li 10b] Li, W., Wang, X., Zhang, R., Cui, Y., Mao, J., and Jin, R.: Exploitation and exploration in a performance based contextual advertising system, in Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2010) [Lin 02] Lin, S.-H. and Ho, J.-M.: Discovering informative content blocks from Web documents, in Proceedings of the eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , ACM (2002) [Lomas 16] Lomas, J. D., Forlizzi, J., Poonwala, N., Patel, N., Shodhan, S., Patel, K., Koedinger, K., and Brunskill, E.: Interface design optimization as a multi-armed bandit problem, in Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, pp , ACM (2016) [Lu 10] Lu, T., Pál, D., and Pál, M.: Contextual multi-armed bandits, in Proceedings of the 13th International Conference on Artificial Intelligence and Statistics, pp , PMLR (2010) [Moore 02] Moore, G. A.: Crossing the chasm, Capstone (2002) [Nie 09] Nie, Z., Wen, J.-R., and Ma, W.-Y.: Webpage understanding: beyond page-level search, SIGMOD Rec., Vol. 37, No. 4, pp (2009) [Park 07] Park, S.: Webpage design optimization using genetic algorithm driven CSS, PhD thesis, Iowa State University (2007)
159 147 [Penalver 98] Penalver, J. G. and Merelo, J.: Optimizing web page layout using an annealed genetic algorithm as client-side script, in Proceedings of the International Conference on Parallel Problem Solving from Nature, pp , Springer (1998) [Ries 11] Ries, E.: The Lean Startup: How Today s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses, Crown Books (2011) [Russo 16] Russo, D. and Van Roy, B.: An information-theoretic analysis of Thompson sampling, Journal of Machine Learning Research, Vol. 17, No. 68, pp (2016) [Sengamedu 08] Sengamedu, S. H. and Mehta, R. R.: Web page layout optimization using section importance, in Proceedings of the 17th International Conference on World Wide Web, ACM (2008) [Siroker 13] Siroker, D. and Koomen, P.: A/B Testing: The Most Powerful Way to Turn Clicks Into Customers, John Wiley & Sons (2013) [Sommerville 11] Sommerville, I.: Software Engineering, Pearson (2011) [Spear 04] Spear, S. J.: Learning to lead at Toyota, Harvard Business Review, Vol. 82, No. 5, pp (2004) [Srinivas 10] Srinivas, N., Krause, A., Seeger, M., and Kakade, S. M.: Gaussian process optimization in the bandit setting: no regret and experimental design, in Proceedings of the 27th International Conference on Machine Learning, pp , Omnipress (2010) [Sutskever 14] Sutskever, I., Vinyals, O., and Le, Q. V.: Sequence to sequence learning with neural networks, in Advances in neural information processing systems, pp (2014) [Takagi 01] Takagi, H.: Interactive evolutionary computation: Fusion of the capabilities of EC optimization and human evaluation, Proceedings of the IEEE, Vol. 89, No. 9, pp (2001) [Tang 10] Tang, D., Agarwal, A., O Brien, D., and Meyer, M.: Overlapping experiment infrastructure: More, better, faster experimentation, in Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 17
160 148 26, ACM (2010) [Thompson 09] Thompson, A.: Automated generation of website content and layout, Michigan Celebration of Women in Computing, p. 43 (2009) [Vinyals 15] Vinyals, O., Toshev, A., Bengio, S., and Erhan, D.: Show and tell: A neural image caption generator, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp (2015) [Wolpert 97] Wolpert, D. H. and Macready, W. G.: No free lunch theorems for optimization, IEEE Transactions on Evolutionary Computation, Vol. 1, No. 1, pp (1997) [Womack 90] Womack, J. P., Jones, D. T., and Roos, D.: Machine that changed the world, Simon and Schuster (1990) [ 95],,, (1995) [ 14],, Twitter,, Vol. 29, No. 1, pp (2014) [ 15], (2015) [ 08],,, Vol. 23, No. 6, pp (2008) [ 15],,,, (2015) [ 12], Pedroso, J. P.,, Rais, A. : Python Gurobi, (2012) [ 04], Proximate Optimality Principle Tabu Search,. C, Vol. 124, No. 3, pp (2004) [ 11], (2011) [ 15],!, (2015) [ 16],,,,,,,,, Vol. 57, No. 3, pp (2016)
161 149 [ 09],,,,, A, Vol. 75, No. 753, pp (2009) [ 09],, WWW,, Vol. 24, No. 6, pp (2009) [ 12],,,,,, Vol. 27, No. 3, pp (2012) [ 08],, (2008) [ 92], (1992) [ 13] : 2.,, Vol. 54, No. 2, pp (2013) [ 03],,,, Vol. 18, No. 5, pp (2003) [ 13] :, (2013) [ 93],, Vol. 1, No. 2, pp. 2 7 (1993) [ 16],, (2016) [ 00],,, Vol. 41, No. 3, pp (2000) [ 16],,,,,,,,,,,,,,,,,,,,,, (2016) [ 12],, :, (2012) [ 16],, (2016) [ 00],,
162 150 D, Vol. 83, No. 1, pp (2000) [ 88],, (1988) [ 16],,, 30, Vol. 30, (2016)
163 151,,,..,,,,,,.. D Vol.J99-D No.1, pp , 2016.,.. Vol.29 No.5, p , 2014.,,,,,,,. Web. Vol.29 No.5, p , Shuhei Iitsuka, Kazuya Kawakami, Seigen Hagiwara, Takayoshi Kawakami, Takayuki Hamada, and Yutaka Matsuo. Inferring Win Lose Product Network from User Behavior. In Proceedings of 2017 IEEE/WIC/ACM International Conference on Web Intelligence, Shuhei Iitsuka and Yutaka Matsuo. Website Optimization Problem and Its Solutions. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp , 2015.
164 152 Naoki Tani, Danushka Bollegala, Naiwala Chandrasiri, Keisuke Okamoto, Kazunari Nawa, Shuhei Iitsuka, and Yutaka Matsuo. Collaborative exploratory search in real-world context. In Proceedings of the 20th ACM international conference on Information and knowledge management, pp , 2011., , ,, ,. KPI , ,,,,,,..net ,,,,,,,. Web ,,
4. C i k = 2 k-means C 1 i, C 2 i 5. C i x i p [ f(θ i ; x) = (2π) p 2 Vi 1 2 exp (x µ ] i) t V 1 i (x µ i ) 2 BIC BIC = 2 log L( ˆθ i ; x i C i ) + q
![4. C i k = 2 k-means C 1 i, C 2 i 5. C i x i p [ f(θ i ; x) = (2π) p 2 Vi 1 2 exp (x µ ] i) t V 1 i (x µ i ) 2 BIC BIC = 2 log L( ˆθ i ; x i C i ) + q](/thumbs/92/110416683.jpg "4. C i k = 2 k-means C 1 i, C 2 i 5. C i x i p [ f(θ i ; x) = (2π) p 2 Vi 1 2 exp (x µ ] i) t V 1 i (x µ i ) 2 BIC BIC = 2 log L( ˆθ i ; x i C i ) + q")
x-means 1 2 2 x-means, x-means k-means Bayesian Information Criterion BIC Watershed x-means Moving Object Extraction Using the Number of Clusters Determined by X-means Clustering Naoki Kubo, 1 Kousuke
_314I01BM浅谷2.indd
587 ネットワークの表現学習 1 1 1 1 Deep Learning [1] Google [2] Deep Learning [3] [4] 2014 Deepwalk [5] 1 2 [6] [7] [8] 1 2 1 word2vec[9] word2vec 1 http://www.ai-gakkai.or.jp/my-bookmark_vol31-no4 588 31 4 2016
(a) (b) (c) Canny (d) 1 ( x α, y α ) 3 (x α, y α ) (a) A 2 + B 2 + C 2 + D 2 + E 2 + F 2 = 1 (3) u ξ α u (A, B, C, D, E, F ) (4) ξ α (x 2 α, 2x α y α,
[II] Optimization Computation for 3-D Understanding of Images [II]: Ellipse Fitting 1. (1) 2. (2) (edge detection) (edge) (zero-crossing) Canny (Canny operator) (3) 1(a) [I] [II] [III] [IV ] E-mail sugaya@iim.ics.tut.ac.jp
Convolutional Neural Network A Graduation Thesis of College of Engineering, Chubu University Investigation of feature extraction by Convolution
Convolutional Neural Network 2014 3 A Graduation Thesis of College of Engineering, Chubu University Investigation of feature extraction by Convolutional Neural Network Fukui Hiroshi 1940 1980 [1] 90 3
[1] SBS [2] SBS Random Forests[3] Random Forests ii
![[1] SBS [2] SBS Random Forests[3] Random Forests ii](/thumbs/74/70333192.jpg "[1] SBS [2] SBS Random Forests[3] Random Forests ii")
Random Forests 2013 3 A Graduation Thesis of College of Engineering, Chubu University Proposal of an efficient feature selection using the contribution rate of Random Forests Katsuya Shimazaki [1] SBS
i E
37-126847 i E ii i 1 1 1.1................................... 1 1.1.1........... 1 1.1.2..... 2 1.1.3...................... 4 1.2................................... 4 1.2.1.................... 5 1.2.2.............
Computational Semantics 1 category specificity Warrington (1975); Warrington & Shallice (1979, 1984) 2 basic level superiority 3 super-ordinate catego
Computational Semantics 1 category specificity Warrington (1975); Warrington & Shallice (1979, 1984) 2 basic level superiority 3 super-ordinate category preservation 1 / 13 analogy by vector space Figure
Haiku Generation Based on Motif Images Using Deep Learning Koki Yoneda 1 Soichiro Yokoyama 2 Tomohisa Yamashita 2 Hidenori Kawamura Scho
Haiku Generation Based on Motif Images Using Deep Learning 1 2 2 2 Koki Yoneda 1 Soichiro Yokoyama 2 Tomohisa Yamashita 2 Hidenori Kawamura 2 1 1 School of Engineering Hokkaido University 2 2 Graduate
2007/8 Vol. J90 D No. 8 Stauffer [7] 2 2 I 1 I 2 2 (I 1(x),I 2(x)) 2 [13] I 2 = CI 1 (C >0) (I 1,I 2) (I 1,I 2) Field Monitoring Server
![2007/8 Vol. J90 D No. 8 Stauffer [7] 2 2 I 1 I 2 2 (I 1(x),I 2(x)) 2 [13] I 2 = CI 1 (C >0) (I 1,I 2) (I 1,I 2) Field Monitoring Server](/thumbs/91/107223763.jpg "2007/8 Vol. J90 D No. 8 Stauffer [7] 2 2 I 1 I 2 2 (I 1(x),I 2(x)) 2 [13] I 2 = CI 1 (C >0) (I 1,I 2) (I 1,I 2) Field Monitoring Server")
a) Change Detection Using Joint Intensity Histogram Yasuyo KITA a) 2 (0 255) (I 1 (x),i 2 (x)) I 2 = CI 1 (C>0) (I 1,I 2 ) (I 1,I 2 ) 2 1. [1] 2 [2] [3] [5] [6] [8] Intelligent Systems Research Institute,
1 IDC Wo rldwide Business Analytics Technology and Services 2013-2017 Forecast 2 24 http://www.soumu.go.jp/johotsusintokei/whitepaper/ja/h24/pdf/n2010000.pdf 3 Manyika, J., Chui, M., Brown, B., Bughin,
IPSJ SIG Technical Report Vol.2017-MUS-116 No /8/24 MachineDancing: 1,a) 1,b) 3 MachineDancing MachineDancing MachineDancing 1 MachineDan
MachineDancing: 1,a) 1,b) 3 MachineDancing 2 1. 3 MachineDancing MachineDancing 1 MachineDancing MachineDancing [1] 1 305 0058 1-1-1 a) s.fukayama@aist.go.jp b) m.goto@aist.go.jp 1 MachineDancing 3 CG
TF-IDF TDF-IDF TDF-IDF Extracting Impression of Sightseeing Spots from Blogs for Supporting Selection of Spots to Visit in Travel Sat
1 1 2 1. TF-IDF TDF-IDF TDF-IDF. 3 18 6 Extracting Impression of Sightseeing Spots from Blogs for Supporting Selection of Spots to Visit in Travel Satoshi Date, 1 Teruaki Kitasuka, 1 Tsuyoshi Itokawa 2
(a) (b) 1 JavaScript Web Web Web CGI Web Web JavaScript Web mixi facebook SNS Web URL ID Web 1 JavaScript Web 1(a) 1(b) JavaScript & Web Web Web Webji
Webjig Web 1 1 1 1 Webjig / Web Web Web Web Web / Web Webjig Web DOM Web Webjig / Web Web Webjig: a visualization tool for analyzing user behaviors in dynamic web sites Mikio Kiura, 1 Masao Ohira, 1 Hidetake
ばらつき抑制のための確率最適制御
( ) http://wwwhayanuemnagoya-uacjp/ fujimoto/ 2011 3 9 11 ( ) 2011/03/09-11 1 / 46 Outline 1 2 3 4 5 ( ) 2011/03/09-11 2 / 46 Outline 1 2 3 4 5 ( ) 2011/03/09-11 3 / 46 (1/2) r + Controller - u Plant y
IPSJ SIG Technical Report 1,a) 1,b) 1,c) 1,d) 2,e) 2,f) 2,g) 1. [1] [2] 2 [3] Osaka Prefecture University 1 1, Gakuencho, Naka, Sakai,
![IPSJ SIG Technical Report 1,a) 1,b) 1,c) 1,d) 2,e) 2,f) 2,g) 1. [1] [2] 2 [3] Osaka Prefecture University 1 1, Gakuencho, Naka, Sakai,](/thumbs/94/119099134.jpg "IPSJ SIG Technical Report 1,a) 1,b) 1,c) 1,d) 2,e) 2,f) 2,g) 1. [1] [2] 2 [3] Osaka Prefecture University 1 1, Gakuencho, Naka, Sakai,")
1,a) 1,b) 1,c) 1,d) 2,e) 2,f) 2,g) 1. [1] [2] 2 [3] 1 599 8531 1 1 Osaka Prefecture University 1 1, Gakuencho, Naka, Sakai, Osaka 599 8531, Japan 2 565 0871 Osaka University 1 1, Yamadaoka, Suita, Osaka
No. 3 Oct The person to the left of the stool carried the traffic-cone towards the trash-can. α α β α α β α α β α Track2 Track3 Track1 Track0 1
ACL2013 TACL 1 ACL2013 Grounded Language Learning from Video Described with Sentences (Yu and Siskind 2013) TACL Transactions of the Association for Computational Linguistics What Makes Writing Great?
IPSJ SIG Technical Report Vol.2009-CVIM-167 No /6/10 Real AdaBoost HOG 1 1 1, 2 1 Real AdaBoost HOG HOG Real AdaBoost HOG A Method for Reducing
Real AdaBoost HOG 1 1 1, 2 1 Real AdaBoost HOG HOG Real AdaBoost HOG A Method for Reducing number of HOG Features based on Real AdaBoost Chika Matsushima, 1 Yuji Yamauchi, 1 Takayoshi Yamashita 1, 2 and
CVaR
CVaR 20 4 24 3 24 1 31 ,.,.,. Markowitz,., (Value-at-Risk, VaR) (Conditional Value-at-Risk, CVaR). VaR, CVaR VaR. CVaR, CVaR. CVaR,,.,.,,,.,,. 1 5 2 VaR CVaR 6 2.1................................................
1 1 CodeDrummer CodeMusician CodeDrummer Fig. 1 Overview of proposal system c
CodeDrummer: 1 2 3 1 CodeDrummer: Sonification Methods of Function Calls in Program Execution Kazuya Sato, 1 Shigeyuki Hirai, 2 Kazutaka Maruyama 3 and Minoru Terada 1 We propose a program sonification
1 AND TFIDF Web DFIWF Wikipedia Web Web 2. 3. 4. AND 5. Wikipedia AND 6. Wikipedia Web 7. 8. 2. Ma [4] Ma URL AND Tian [8] Tian Tian Web Cimiano [3] [
![1 AND TFIDF Web DFIWF Wikipedia Web Web 2. 3. 4. AND 5. Wikipedia AND 6. Wikipedia Web 7. 8. 2. Ma [4] Ma URL AND Tian [8] Tian Tian Web Cimiano [3] [](/thumbs/41/22779750.jpg "1 AND TFIDF Web DFIWF Wikipedia Web Web 2. 3. 4. AND 5. Wikipedia AND 6. Wikipedia Web 7. 8. 2. Ma [4] Ma URL AND Tian [8] Tian Tian Web Cimiano [3] [")
DEIM Forum 2015 B1-5 606 8501 606 8501 E-mail: komurasaki@dl.kuis.kyoto-u.ac.jp, tajima@i.kyoto-u.ac.jp Web Web AND AND Web 1. Twitter Facebook SNS Web Web Web Web [5] Bollegala [2] Web Web 1 Google Microsoft
Lyra 2 2 2 X Y X Y ivis Designer Lyra ivisdesigner Lyra ivisdesigner 2 ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) (1) (2) (3) (4) (5) Iv Studio [8] 3 (5) (4) (1) (
![Lyra 2 2 2 X Y X Y ivis Designer Lyra ivisdesigner Lyra ivisdesigner 2 ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) (1) (2) (3) (4) (5) Iv Studio [8] 3 (5) (4) (1) (](/thumbs/42/23098909.jpg "Lyra 2 2 2 X Y X Y ivis Designer Lyra ivisdesigner Lyra ivisdesigner 2 ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) (1) (2) (3) (4) (5) Iv Studio [8] 3 (5) (4) (1) (")
1,a) 2,b) 2,c) 1. Web [1][2][3][4] [5] 1 2 a) ito@iplab.cs.tsukuba.ac.jp b) misue@cs.tsukuba.ac.jp c) jiro@cs.tsukuba.ac.jp [6] Lyra[5] ivisdesigner[6] [7] 2 Lyra ivisdesigner c 2012 Information Processing
IPSJ SIG Technical Report Vol.2010-CVIM-170 No /1/ Visual Recognition of Wire Harnesses for Automated Wiring Masaki Yoneda, 1 Ta
1 1 1 1 2 1. Visual Recognition of Wire Harnesses for Automated Wiring Masaki Yoneda, 1 Takayuki Okatani 1 and Koichiro Deguchi 1 This paper presents a method for recognizing the pose of a wire harness
DEIM Forum 2012 E Web Extracting Modification of Objec
DEIM Forum 2012 E4-2 670 0092 1 1 12 E-mail: nd11g028@stshse.u-hyogo.ac.jp, {dkitayama,sumiya}@shse.u-hyogo.ac.jp Web Extracting Modification of Objects for Supporting Map Browsing Junki MATSUO, Daisuke
untitled
IT E- IT http://www.ipa.go.jp/security/ CERT/CC http://www.cert.org/stats/#alerts IPA IPA 2004 52,151 IT 2003 12 Yahoo 451 40 2002 4 18 IT 1/14 2.1 DoS(Denial of Access) IDS(Intrusion Detection System)
1: 2: 3: 4: 2. 1 Exploratory Search [4] Exploratory Search 2. 1 [7] [8] [9] [10] Exploratory Search
![1: 2: 3: 4: 2. 1 Exploratory Search [4] Exploratory Search 2. 1 [7] [8] [9] [10] Exploratory Search](/thumbs/49/25839215.jpg "1: 2: 3: 4: 2. 1 Exploratory Search [4] Exploratory Search 2. 1 [7] [8] [9] [10] Exploratory Search")
DEIM Forum 2013 D2-1 112 8610 2-1-1 E-mail: {aco,itot}@itolab.is.ocha.ac.jp, chiemi@is.ocha.ac.jp Exploratory Search A product Search System for women adjusting amount of browsed items Abstract Eriko KOIKE,
独立行政法人情報通信研究機構 Development of the Information Analysis System WISDOM KIDAWARA Yutaka NICT Knowledge Clustered Group researched and developed the infor
独立行政法人情報通信研究機構 KIDAWARA Yutaka NICT Knowledge Clustered Group researched and developed the information analysis system WISDOM as a research result of the second medium-term plan. WISDOM has functions that
2008 : 80725872 1 2 2 3 2.1.......................................... 3 2.2....................................... 3 2.3......................................... 4 2.4 ()..................................
The 19th Game Programming Workshop 2014 SHOT 1,a) 2 UCT SHOT UCT SHOT UCT UCT SHOT UCT An Empirical Evaluation of the Effectiveness of the SHOT algori
SHOT 1,a) 2 UCT SHOT UCT SHOT UCT UCT SHOT UCT An Empirical Evaluation of the Effectiveness of the SHOT algorithm in Go and Gobang Masahiro Honjo 1,a) Yoshimasa Tsuruoka 2 Abstract: Today, UCT is the most
1 Kinect for Windows M = [X Y Z] T M = [X Y Z ] T f (u,v) w 3.2 [11] [7] u = f X +u Z 0 δ u (X,Y,Z ) (5) v = f Y Z +v 0 δ v (X,Y,Z ) (6) w = Z +
![1 Kinect for Windows M = [X Y Z] T M = [X Y Z ] T f (u,v) w 3.2 [11] [7] u = f X +u Z 0 δ u (X,Y,Z ) (5) v = f Y Z +v 0 δ v (X,Y,Z ) (6) w = Z +](/thumbs/81/83701592.jpg "1 Kinect for Windows M = [X Y Z] T M = [X Y Z ] T f (u,v) w 3.2 [11] [7] u = f X +u Z 0 δ u (X,Y,Z ) (5) v = f Y Z +v 0 δ v (X,Y,Z ) (6) w = Z +")
3 3D 1,a) 1 1 Kinect (X, Y) 3D 3D 1. 2010 Microsoft Kinect for Windows SDK( (Kinect) SDK ) 3D [1], [2] [3] [4] [5] [10] 30fps [10] 3 Kinect 3 Kinect Kinect for Windows SDK 3 Microsoft 3 Kinect for Windows
DPA,, ShareLog 3) 4) 2.2 Strino Strino STRain-based user Interface with tacticle of elastic Natural ObjectsStrino 1 Strino ) PC Log-Log (2007 6)
1 2 1 3 Experimental Evaluation of Convenient Strain Measurement Using a Magnet for Digital Public Art Junghyun Kim, 1 Makoto Iida, 2 Takeshi Naemura 1 and Hiroyuki Ota 3 We present a basic technology
IS1-09 第 回画像センシングシンポジウム, 横浜,14 年 6 月 2 Hough Forest Hough Forest[6] Random Forest( [5]) Random Forest Hough Forest Hough Forest 2.1 Hough Forest 1 2.2
![IS1-09 第 回画像センシングシンポジウム, 横浜,14 年 6 月 2 Hough Forest Hough Forest[6] Random Forest( [5]) Random Forest Hough Forest Hough Forest 2.1 Hough Forest 1 2.2](/thumbs/91/106196100.jpg "IS1-09 第 回画像センシングシンポジウム, 横浜,14 年 6 月 2 Hough Forest Hough Forest[6] Random Forest( [5]) Random Forest Hough Forest Hough Forest 2.1 Hough Forest 1 2.2")
IS1-09 第 回画像センシングシンポジウム, 横浜,14 年 6 月 MI-Hough Forest () E-mail: ym@vision.cs.chubu.ac.jphf@cs.chubu.ac.jp Abstract Hough Forest Random Forest MI-Hough Forest Multiple Instance Learning Bag Hough Forest
[2][3][4][5] 4 ( 1 ) ( 2 ) ( 3 ) ( 4 ) 2. Shiratori [2] Shiratori [3] [4] GP [5] [6] [7] [8][9] Kinect Choi [10] 3. 1 c 2016 Information Processing So
![[2][3][4][5] 4 ( 1 ) ( 2 ) ( 3 ) ( 4 ) 2. Shiratori [2] Shiratori [3] [4] GP [5] [6] [7] [8][9] Kinect Choi [10] 3. 1 c 2016 Information Processing So](/thumbs/76/74386745.jpg "[2][3][4][5] 4 ( 1 ) ( 2 ) ( 3 ) ( 4 ) 2. Shiratori [2] Shiratori [3] [4] GP [5] [6] [7] [8][9] Kinect Choi [10] 3. 1 c 2016 Information Processing So")
1,a) 2 2 1 2,b) 3,c) A choreographic authoring system reflecting a user s preference Ryo Kakitsuka 1,a) Kosetsu Tsukuda 2 Satoru Fukayama 2 Naoya Iwamoto 1 Masataka Goto 2,b) Shigeo Morishima 3,c) Abstract:
27 YouTube YouTube UGC User Generated Content CDN Content Delivery Networks LRU Least Recently Used UGC YouTube CGM Consumer Generated Media CGM CGM U
YouTube 2016 2 16 27 YouTube YouTube UGC User Generated Content CDN Content Delivery Networks LRU Least Recently Used UGC YouTube CGM Consumer Generated Media CGM CGM UGC UGC YouTube k-means YouTube YouTube
IPSJ SIG Technical Report Vol.2010-MPS-77 No /3/5 VR SIFT Virtual View Generation in Hallway of Cybercity Buildings from Video Sequen
VR 1 1 1 1 1 SIFT Virtual View Generation in Hallway of Cybercity Buildings from Video Sequences Sachiyo Yoshida, 1 Masami Takata 1 and Joe Kaduki 1 Appearance of Three-dimensional (3D) building model
130 Oct Radial Basis Function RBF Efficient Market Hypothesis Fama ) 4) 1 Fig. 1 Utility function. 2 Fig. 2 Value function. (1) (2)
Vol. 47 No. SIG 14(TOM 15) Oct. 2006 RBF 2 Effect of Stock Investor Agent According to Framing Effect to Stock Exchange in Artificial Stock Market Zhai Fei, Shen Kan, Yusuke Namikawa and Eisuke Kita Several
3 2 2 (1) (2) (3) (4) 4 4 AdaBoost 2. [11] Onishi&Yoda [8] Iwashita&Stoica [5] 4 [3] 3. 3 (1) (2) (3)
![3 2 2 (1) (2) (3) (4) 4 4 AdaBoost 2. [11] Onishi&Yoda [8] Iwashita&Stoica [5] 4 [3] 3. 3 (1) (2) (3)](/thumbs/68/59204737.jpg "3 2 2 (1) (2) (3) (4) 4 4 AdaBoost 2. [11] Onishi&Yoda [8] Iwashita&Stoica [5] 4 [3] 3. 3 (1) (2) (3)")
(MIRU2012) 2012 8 820-8502 680-4 E-mail: {d kouno,shimada,endo}@pluto.ai.kyutech.ac.jp (1) (2) (3) (4) 4 AdaBoost 1. Kanade [6] CLAFIC [12] EigenFace [10] 1 1 2 1 [7] 3 2 2 (1) (2) (3) (4) 4 4 AdaBoost
Vol. 42 No. SIG 8(TOD 10) July HTML 100 Development of Authoring and Delivery System for Synchronized Contents and Experiment on High Spe
Vol. 42 No. SIG 8(TOD 10) July 2001 1 2 3 4 HTML 100 Development of Authoring and Delivery System for Synchronized Contents and Experiment on High Speed Networks Yutaka Kidawara, 1 Tomoaki Kawaguchi, 2
Fig. 2 28th Ryuou Tournament, Match 5, 59th move. The last move is Black s Rx5f. 1 Tic-Tac-Toe Fig. 1 AsearchtreeofTic-Tac-Toe. [2] [3], [4]
![Fig. 2 28th Ryuou Tournament, Match 5, 59th move. The last move is Black s Rx5f. 1 Tic-Tac-Toe Fig. 1 AsearchtreeofTic-Tac-Toe. [2] [3], [4]](/thumbs/91/107302309.jpg "Fig. 2 28th Ryuou Tournament, Match 5, 59th move. The last move is Black s Rx5f. 1 Tic-Tac-Toe Fig. 1 AsearchtreeofTic-Tac-Toe. [2] [3], [4]")
1,a) 2 3 2017 4 6, 2017 9 5 Predicting Moves in Comments for Shogi Commentary Generation Hirotaka Kameko 1,a) Shinsuke Mori 2 Yoshimasa Tsuruoka 3 Received: April 6, 2017, Accepted: September 5, 2017 Abstract:
DEIM Forum 2009 C8-4 QA NTT QA QA QA 2 QA Abstract Questions Recomme
DEIM Forum 2009 C8-4 QA NTT 239 0847 1 1 E-mail: {kabutoya.yutaka,kawashima.harumi,fujimura.ko}@lab.ntt.co.jp QA QA QA 2 QA Abstract Questions Recommendation Based on Evolution Patterns of a QA Community
A Study on Throw Simulation for Baseball Pitching Machine with Rollers and Its Optimization Shinobu SAKAI*5, Yuichiro KITAGAWA, Ryo KANAI and Juhachi
A Study on Throw Simulation for Baseball Pitching Machine with Rollers and Its Optimization Shinobu SAKAI*5, Yuichiro KITAGAWA, Ryo KANAI and Juhachi ODA Department of Human and Mechanical Systems Engineering,
IPSJ SIG Technical Report Vol.2013-HCI-152 No /3/13 1,a) 1,b) 2,c) / GPS Bluetooth(BT) WiFi BT WiFi 1. Bluetooth WiFi 1 / 1 2 a)
1,a) 1,b) 2,c) / GPS Bluetooth(BT) WiFi BT WiFi 1. Bluetooth WiFi 1 / 1 2 a) rtokuami@kwansei.ac.jp b) kono@kwansei.ac.jp c) nakamura@dl.kuis.kyoto-u.ac.jp / 2. Apple iphoto Google Picasa GPS GPS GPS [1][2]
18 2 20 W/C W/C W/C 4-4-1 0.05 1.0 1000 1. 1 1.1 1 1.2 3 2. 4 2.1 4 (1) 4 (2) 4 2.2 5 (1) 5 (2) 5 2.3 7 3. 8 3.1 8 3.2 ( ) 11 3.3 11 (1) 12 (2) 12 4. 14 4.1 14 4.2 14 (1) 15 (2) 16 (3) 17 4.3 17 5. 19
理工ジャーナル ２３‐１☆／１．外村
Yoshinobu TONOMURA Professor, Department of Media Informatics 1 10 YouTube 2 1900 100 1 3 2 3 3 3 1 2 3 4 90 1 90 MIT Project Athena 1983 1991 2 3 4 5 6 7 8 9 10 2 90 11 12 7 13 14 15 16 17 18 19 390 5
,.,. NP,., ,.,,.,.,,, (PCA)...,,. Tipping and Bishop (1999) PCA. (PPCA)., (Ilin and Raiko, 2010). PPCA EM., , tatsukaw
,.,. NP,.,. 1 1.1.,.,,.,.,,,. 2. 1.1.1 (PCA)...,,. Tipping and Bishop (1999) PCA. (PPCA)., (Ilin and Raiko, 2010). PPCA EM., 152-8552 2-12-1, tatsukawa.m.aa@m.titech.ac.jp, 190-8562 10-3, mirai@ism.ac.jp
SEJulyMs更新V7
1 2 ( ) Quantitative Characteristics of Software Process (Is There any Myth, Mystery or Anomaly? No Silver Bullet?) Zenya Koono and Hui Chen A process creates a product. This paper reviews various samples
HP cafe HP of A A B of C C Map on N th Floor coupon A cafe coupon B Poster A Poster A Poster B Poster B Case 1 Show HP of each company on a user scree
LAN 1 2 3 2 LAN WiFiTag WiFiTag LAN LAN 100% WiFi Tag An Improved Determination Method with Multiple Access Points for Relative Position Estimation Using Wireless LAN Abstract: We have proposed a WiFiTag
untitled
The Impact of Digitization on Music Production: From a Perspective of Modularity 51 2 pp. 87-108 2003 12 I 21 3 Information and Communication Technology, ICT 0 1 1 20 1 199820012000 1 MP3 CD 2 3 II CD
Microsoft Word - toyoshima-deim2011.doc
DEIM Forum 2011 E9-4 252-0882 5322 252-0882 5322 E-mail: t09651yt, sashiori, kiyoki @sfc.keio.ac.jp CBIR A Meaning Recognition System for Sign-Logo by Color-Shape-Based Similarity Computations for Images
(MIRU2008) HOG Histograms of Oriented Gradients (HOG)
(MIRU2008) 2008 7 HOG - - E-mail: katsu0920@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp Histograms of Oriented Gradients (HOG) HOG Shape Contexts HOG 5.5 Histograms of Oriented Gradients D Human
教師情報を必要としないWebページ群のコンテンツ自動抽出ツールの提案
DEIM Forum 2009 A8-4 Web 305-8573 1-1-1 305-8573 1-1-1 E-mail: m.yoshida@mibel.cs.tsukuba.ac.jp, myama@cs.tsukuba.ac.jp CMS Web Web Web Web Web Web Web Web,,, HTML, Web, Web, Primary Content Extraction
2009/9 Vol. J92 D No. 9 HTML [3] Microsoft PowerPoint Apple Keynote OpenOffice Impress XML 4 1 (A) (C) (F) 2. 2. 1 1484 Fig. 1 1 An example of slide i
![2009/9 Vol. J92 D No. 9 HTML [3] Microsoft PowerPoint Apple Keynote OpenOffice Impress XML 4 1 (A) (C) (F) 2. 2. 1 1484 Fig. 1 1 An example of slide i](/thumbs/39/20176621.jpg "2009/9 Vol. J92 D No. 9 HTML [3] Microsoft PowerPoint Apple Keynote OpenOffice Impress XML 4 1 (A) (C) (F) 2. 2. 1 1484 Fig. 1 1 An example of slide i")
a) Structure Extraction from Presentation Slide Information Tessai HAYAMA a), Hidetsugu NANBA, and Susumu KUNIFUJI Web 1. Web Graduate School of Knowledge Science, Japan Advanced Institute of Science and
Sobel Canny i
21 Edge Feature for Monochrome Image Retrieval 1100311 2010 3 1 3 3 2 2 7 200 Sobel Canny i Abstract Edge Feature for Monochrome Image Retrieval Naoto Suzue Content based image retrieval (CBIR) has been
三石貴志.indd
流通科学大学論集 - 経済 情報 政策編 - 第 21 巻第 1 号,23-33(2012) SIRMs SIRMs Fuzzy fuzzyapproximate approximatereasoning reasoningusing using Lukasiewicz Łukasiewicz logical Logical operations Operations Takashi Mitsuishi
1 Fogg Fogg Behavior Model [1] information cascade [2] TPO [3] Fig. 2 Target area of this paper. 1 Fig. 1 Fogg b
![1 Fogg Fogg Behavior Model [1] information cascade [2] TPO [3] Fig. 2 Target area of this paper. 1 Fig. 1 Fogg b](/thumbs/91/105674179.jpg "1 Fogg Fogg Behavior Model [1] information cascade [2] TPO [3] Fig. 2 Target area of this paper. 1 Fig. 1 Fogg b")
1,a) 1 1 1 2014 9 20, 2015 1 5 TPO Extracting Purpose-for-Action to Enhance Local Information Service Noriko Yokoyama 1,a) Kaname Funakoshi 1 Hiroyuki Toda 1 Yoshimasa Koike 1 Received: September 20, 2014,
IPSJ SIG Technical Report Vol.2009-DPS-141 No.20 Vol.2009-GN-73 No.20 Vol.2009-EIP-46 No /11/27 1. MIERUKEN 1 2 MIERUKEN MIERUKEN MIERUKEN: Spe
1. MIERUKEN 1 2 MIERUKEN MIERUKEN MIERUKEN: Speech Visualization System Based on Augmented Reality Yuichiro Nagano 1 and Takashi Yoshino 2 As the spread of the Augmented Reality(AR) technology and service,
03.Œk’ì
HRS KG NG-HRS NG-KG AIC Fama 1965 Mandelbrot Blattberg Gonedes t t Kariya, et. al. Nagahara ARCH EngleGARCH Bollerslev EGARCH Nelson GARCH Heynen, et. al. r n r n =σ n w n logσ n =α +βlogσ n 1 + v n w
2). 3) 4) 1.2 NICTNICT DCRA Dihedral Corner Reflector micro-arraysdcra DCRA DCRA DCRA 3D DCRA PC USB PC PC ON / OFF Velleman K8055 K8055 K8055
1 1 1 2 DCRA 1. 1.1 1) 1 Tactile Interface with Air Jets for Floating Images Aya Higuchi, 1 Nomin, 1 Sandor Markon 1 and Satoshi Maekawa 2 The new optical device DCRA can display floating images in free
3.1 Thalmic Lab Myo * Bluetooth PC Myo 8 RMS RMS t RMS(t) i (i = 1, 2,, 8) 8 SVM libsvm *2 ν-svm 1 Myo 2 8 RMS 3.2 Myo (Root
1,a) 2 2 1. 1 College of Information Science, School of Informatics, University of Tsukuba 2 Faculty of Engineering, Information and Systems, University of Tsukuba a) oharada@iplab.cs.tsukuba.ac.jp 2.
IT IBM Corporation
2009/9/25 ATC. 1 2009 IBM Corporation 1. 1. 2. 3. IT 2 2009 IBM Corporation 2006 8 9 (?) Google CEO, Eric Schmidt @ Search Engine Strategies Conference (*) emergent () 10 Network ComputerAjax LAMP (Linux
untitled
2 : n =1, 2,, 10000 0.5125 0.51 0.5075 0.505 0.5025 0.5 0.4975 0.495 0 2000 4000 6000 8000 10000 2 weak law of large numbers 1. X 1,X 2,,X n 2. µ = E(X i ),i=1, 2,,n 3. σi 2 = V (X i ) σ 2,i=1, 2,,n ɛ>0
IPSJ SIG Technical Report GPS LAN GPS LAN GPS LAN Location Identification by sphere image and hybrid sensing Takayuki Katahira, 1 Yoshio Iwai 1
1 1 1 GPS LAN GPS LAN GPS LAN Location Identification by sphere image and hybrid sensing Takayuki Katahira, 1 Yoshio Iwai 1 and Hiroshi Ishiguro 1 Self-location is very informative for wearable systems.
WISS 2018 [2 4] [5,6] Query-by-Dancing Query-by- Dancing Cao [1] OpenPose 2 Ghias [7] Query by humming Chen [8] Query by rhythm Jang [9] Query-by-tapp
![WISS 2018 [2 4] [5,6] Query-by-Dancing Query-by- Dancing Cao [1] OpenPose 2 Ghias [7] Query by humming Chen [8] Query by rhythm Jang [9] Query-by-tapp](/thumbs/91/107555259.jpg "WISS 2018 [2 4] [5,6] Query-by-Dancing Query-by- Dancing Cao [1] OpenPose 2 Ghias [7] Query by humming Chen [8] Query by rhythm Jang [9] Query-by-tapp")
Query-by-Dancing: WISS 2018. Query-by-Dancing Query-by-Dancing 1 OpenPose [1] Copyright is held by the author(s). DJ DJ DJ WISS 2018 [2 4] [5,6] Query-by-Dancing Query-by- Dancing Cao [1] OpenPose 2 Ghias
Input image Initialize variables Loop for period of oscillation Update height map Make shade image Change property of image Output image Change time L
1,a) 1,b) 1/f β Generation Method of Animation from Pictures with Natural Flicker Abstract: Some methods to create animation automatically from one picture have been proposed. There is a method that gives
IPSJ SIG Technical Report Vol.2013-CVIM-187 No /5/30 1,a) 1,b), 1,,,,,,, (DNN),,,, 2 (CNN),, 1.,,,,,,,,,,,,,,,,,, [1], [6], [7], [12], [13]., [
![IPSJ SIG Technical Report Vol.2013-CVIM-187 No /5/30 1,a) 1,b), 1,,,,,,, (DNN),,,, 2 (CNN),, 1.,,,,,,,,,,,,,,,,,, [1], [6], [7], [12], [13]., [](/thumbs/94/121682292.jpg "IPSJ SIG Technical Report Vol.2013-CVIM-187 No /5/30 1,a) 1,b), 1,,,,,,, (DNN),,,, 2 (CNN),, 1.,,,,,,,,,,,,,,,,,, [1], [6], [7], [12], [13]., [")
,a),b),,,,,,,, (DNN),,,, (CNN),,.,,,,,,,,,,,,,,,,,, [], [6], [7], [], [3]., [8], [0], [7],,,, Tohoku University a) omokawa@vision.is.tohoku.ac.jp b) okatani@vision.is.tohoku.ac.jp, [3],, (DNN), DNN, [3],
IPSJ SIG Technical Report Vol.2014-MBL-70 No.49 Vol.2014-UBI-41 No /3/15 2,a) 2,b) 2,c) 2,d),e) WiFi WiFi WiFi 1. SNS GPS Twitter Facebook Twit
2,a) 2,b) 2,c) 2,d),e) WiFi WiFi WiFi 1. SNS GPS Twitter Facebook Twitter Ustream 1 Graduate School of Information Science and Technology, Osaka University, Japan 2 Cybermedia Center, Osaka University,
Wikipedia YahooQA MAD 4)5) MAD Web 6) 3. YAMAHA 7) 8) 2 3 4 5 6 2. Vocaloid2 2006 1 PV 2009 1 1100 200 YouTube 1 minato minato ussy 3D MAD F EDis ussy
1, 2 3 1, 2 Web Fischer Social Creativity 1) Social Creativity CG Network Analysis of an Emergent Massively Collaborative Creation Community Masahiro Hamasaki, 1, 2 Hideaki Takeda 3 and Takuichi Nishimura
Trial for Value Quantification from Exceptional Utterances 37-066593 1 5 1.1.................................. 5 1.2................................ 8 2 9 2.1.............................. 9 2.1.1.........................
TA3-4 31st Fuzzy System Symposium (Chofu, September 2-4, 2015) Interactive Recommendation System LeonardoKen Orihara, 1 Tomonori Hashiyama, 1
Interactive Recommendation System 1 1 1 1 LeonardoKen Orihara, 1 Tomonori Hashiyama, 1 Shun ichi Tano 1 1 Graduate School of Information Systems, The University of Electro-Communications Abstract: The
IPSJ SIG Technical Report Vol.2014-HCI-157 No.26 Vol.2014-GN-91 No.26 Vol.2014-EC-31 No /3/15 1,a) 2 3 Web (SERP) ( ) Web (VP) SERP VP VP SERP
1,a) 2 3 Web (SERP) ( ) Web (VP) SERP VP VP SERP VP Web 1. Web Web Web Web Google SERP SERP 1 1 2-1-1, Hodokubo, Hino, Tokyo 191 8506, Japan 2 4-12-3, Higash-Shinagawa, Shinagawa, Tokyo 140 0002, Japan
専門力_総合力
Program for Leading Graduate Schools Program for Leading Graduate Schools INTRODUCTION 04 CONTENTS 10 15 23 24 30 32 33 34 36 38 39 DATA 40 42 44 Program for Leading Graduate Schools What is a DOCTOR??
FIT2014( 第 13 回情報科学技術フォーラム ) RD-002 Web SNS Yuanyuan Wang Gouki Yasui Yuji Hosokawa Yukiko Kawai Toyokazu Akiyama Kazutoshi Sumiya 1. Twitter 1 Facebo
RD-002 Web SNS Yuanyuan Wang Gouki Yasui Yuji Hosokawa Yukiko Kawai Toyokazu Akiyama Kazutoshi Sumiya 1. Twitter 1 Facebook 2 SNS SNS SNS Twitter SNS [1] SNS [2] Twitter Web Web Web Web SNS Web Web 2 Web
和文タイトル
Paper Browsing System with Structure Analysis and Displaying Annotation on Side-note Windows Takeshi Abekawa Akiko Aizawa National Institute of Informatics Abstract: In this paper, we introduce our on-going
: ( 1) () 1. ( 1) 2. ( 1) 3. ( 2)
Acquiring Organized Information from News by Incremental Theme Refinements 1 1 1 Yutaro Taniguchi 1 Tetsunori Kobayashi 1 Yoshihiko Hayashi 1 1 1 School of Science and Engineering, Waseda University Abstract:
B 20 Web
B 20 Web 0753018 21 1 29 1 1 6 2 8 3 UI 10 3.1........................ 10 3.2 Web............ 11 3.3......... 12 4 UI 14 4.1 Web....................... 15 4.2 Web........... 16 4.3 Web....................
[1] AI [2] Pac-Man Ms. Pac-Man Ms. Pac-Man Pac-Man Ms. Pac-Man IEEE AI Ms. Pac-Man AI [3] AI 2011 UCT[4] [5] 58,990 Ms. Pac-Man AI Ms. Pac-Man 921,360
![[1] AI [2] Pac-Man Ms. Pac-Man Ms. Pac-Man Pac-Man Ms. Pac-Man IEEE AI Ms. Pac-Man AI [3] AI 2011 UCT[4] [5] 58,990 Ms. Pac-Man AI Ms. Pac-Man 921,360](/thumbs/93/114217704.jpg "[1] AI [2] Pac-Man Ms. Pac-Man Ms. Pac-Man Pac-Man Ms. Pac-Man IEEE AI Ms. Pac-Man AI [3] AI 2011 UCT[4] [5] 58,990 Ms. Pac-Man AI Ms. Pac-Man 921,360")
TD(λ) Ms. Pac-Man AI 1,a) 2 3 3 Ms. Pac-Man AI Ms. Pac-Man UCT (Upper Confidence Bounds applied to Trees) TD(λ) UCT UCT Progressive bias Progressive bias UCT UCT Ms. Pac-Man UCT Progressive bias TD(λ)
WII-D 2017 (1) (2) (1) (2) [Tanaka 07] [ 04] [ 10] [ 13, 13], [ 08] [ 13] (1) (2) 2 2 e.g., Wikipedia [ 14] Wikipedia [ 14] Linked Open
![WII-D 2017 (1) (2) (1) (2) [Tanaka 07] [ 04] [ 10] [ 13, 13], [ 08] [ 13] (1) (2) 2 2 e.g., Wikipedia [ 14] Wikipedia [ 14] Linked Open](/thumbs/92/108593098.jpg "WII-D 2017 (1) (2) (1) (2) [Tanaka 07] [ 04] [ 10] [ 13, 13], [ 08] [ 13] (1) (2) 2 2 e.g., Wikipedia [ 14] Wikipedia [ 14] Linked Open")
Web 2017 Original Paper Supporting Exploratory Information Access Based on Comic Content Information 1 Ryo Yamashita Byeongseon Park Mitsunori Matsushita Nomura Research Institute, LTD. r-yamashita@nri.co.jp
1034 IME Web API Web API 1 IME Fig. 1 Suitable situations for context-aware IME. IME IME IME IME 1 GPS Web API Web API Web API Web )
Vol. 52 No. 3 1033 1044 (Mar. 2011) IME 1 2 1 1 IME Web PC Android Dynamic Dictionary Generation Method for Context-aware Input Method Editor Yutaka Arakawa, 1 Shinji Suematsu, 2 Shigeaki Tagashira 1 and
1 [1, 2, 3, 4, 5, 8, 9, 10, 12, 15] The Boston Public Schools system, BPS (Deferred Acceptance system, DA) (Top Trading Cycles system, TTC) cf. [13] [
![1 [1, 2, 3, 4, 5, 8, 9, 10, 12, 15] The Boston Public Schools system, BPS (Deferred Acceptance system, DA) (Top Trading Cycles system, TTC) cf. [13] [](/thumbs/91/106628466.jpg "1 [1, 2, 3, 4, 5, 8, 9, 10, 12, 15] The Boston Public Schools system, BPS (Deferred Acceptance system, DA) (Top Trading Cycles system, TTC) cf. [13] [")
Vol.2, No.x, April 2015, pp.xx-xx ISSN xxxx-xxxx 2015 4 30 2015 5 25 253-8550 1100 Tel 0467-53-2111( ) Fax 0467-54-3734 http://www.bunkyo.ac.jp/faculty/business/ 1 [1, 2, 3, 4, 5, 8, 9, 10, 12, 15] The
IPSJ SIG Technical Report Vol.2015-MUS-106 No.10 Vol.2015-EC-35 No /3/2 BGM 1,4,a) ,4 BGM. BGM. BGM BGM. BGM. BGM. BGM. 1.,. YouTube 201
BGM 1,4,a) 1 2 2 3,4 BGM. BGM. BGM BGM. BGM. BGM. BGM. 1.,. YouTube 2015 1 100.. Web.. BGM.BGM [1]. BGM BGM 1 Waseda University, Shinjuku, Tokyo 169-8555, Japan 2 3 4 JST CREST a) ha-ru-ki@asagi.waseda.jp.
(a) 1 (b) 3. Gilbert Pernicka[2] Treibitz Schechner[3] Narasimhan [4] Kim [5] Nayar [6] [7][8][9] 2. X X X [10] [11] L L t L s L = L t + L s
![(a) 1 (b) 3. Gilbert Pernicka[2] Treibitz Schechner[3] Narasimhan [4] Kim [5] Nayar [6] [7][8][9] 2. X X X [10] [11] L L t L s L = L t + L s](/thumbs/75/72066128.jpg "(a) 1 (b) 3. Gilbert Pernicka[2] Treibitz Schechner[3] Narasimhan [4] Kim [5] Nayar [6] [7][8][9] 2. X X X [10] [11] L L t L s L = L t + L s")
1 1 1, Extraction of Transmitted Light using Parallel High-frequency Illumination Kenichiro Tanaka 1 Yasuhiro Mukaigawa 1 Yasushi Yagi 1 Abstract: We propose a new sharpening method of transmitted scene
IPSJ SIG Technical Report Vol.2009-DBS-149 No /11/ Bow-tie SCC Inter Keyword Navigation based on Degree-constrained Co-Occurrence Graph
1 2 1 Bow-tie SCC Inter Keyword Navigation based on Degree-constrained Co-Occurrence Graph Satoshi Shimada, 1 Tomohiro Fukuhara 2 and Tetsuji Satoh 1 We had proposed a navigation method that generates
Microsoft PowerPoint - SSII_harada pptx
The state of the world The gathered data The processed data w d r I( W; D) I( W; R) The data processing theorem states that data processing can only destroy information. David J.C. MacKay. Information
情報処理学会研究報告 IPSJ SIG Technical Report Vol.2015-DBS-162 No /11/26 1,a) 1,b) EM Designing and developing an interactive data minig tool for rapid r
1,a) 1,b) EM Designing and developing an interactive data minig tool for rapid repeating trials Daishi Kato 1,a) Miki Kiyokazu 1,b) Abstract: Data mining has got attention for finding rules and knowledge
Optical Flow t t + δt 1 Motion Field 3 3 1) 2) 3) Lucas-Kanade 4) 1 t (x, y) I(x, y, t)
http://wwwieice-hbkborg/ 2 2 4 2 -- 2 4 2010 9 3 3 4-1 Lucas-Kanade 4-2 Mean Shift 3 4-3 2 c 2013 1/(18) http://wwwieice-hbkborg/ 2 2 4 2 -- 2 -- 4 4--1 2010 9 4--1--1 Optical Flow t t + δt 1 Motion Field
258 5) GPS 1 GPS 6) GPS DP 7) 8) 10) GPS GPS 2 3 4 5 2. 2.1 3 1) GPS Global Positioning System
Vol. 52 No. 1 257 268 (Jan. 2011) 1 2, 1 1 measurement. In this paper, a dynamic road map making system is proposed. The proposition system uses probe-cars which has an in-vehicle camera and a GPS receiver.
GUI(Graphical User Interface) GUI CLI(Command Line Interface) GUI
24 GUI(Graphical User Interface) GUI CLI(Command Line Interface) GUI 1 1 1.1 GUI................................... 1 1.2 GUI.................... 1 1.2.1.......................... 1 1.2.2...........................
鉄鋼協会プレゼン
NN :~:, 8 Nov., Adaptive H Control for Linear Slider with Friction Compensation positioning mechanism moving table stand manipulator Point to Point Control [G] Continuous Path Control ground Fig. Positoining
21 Pitman-Yor Pitman- Yor [7] n -gram W w n-gram G Pitman-Yor P Y (d, θ, G 0 ) (1) G P Y (d, θ, G 0 ) (1) Pitman-Yor d, θ, G 0 d 0 d 1 θ Pitman-Yor G
![21 Pitman-Yor Pitman- Yor [7] n -gram W w n-gram G Pitman-Yor P Y (d, θ, G 0 ) (1) G P Y (d, θ, G 0 ) (1) Pitman-Yor d, θ, G 0 d 0 d 1 θ Pitman-Yor G](/thumbs/89/99361107.jpg "21 Pitman-Yor Pitman- Yor [7] n -gram W w n-gram G Pitman-Yor P Y (d, θ, G 0 ) (1) G P Y (d, θ, G 0 ) (1) Pitman-Yor d, θ, G 0 d 0 d 1 θ Pitman-Yor G")
ol2013-nl-214 No6 1,a) 2,b) n-gram 1 M [1] (TG: Tree ubstitution Grammar) [2], [3] TG TG 1 2 a) ohno@ilabdoshishaacjp b) khatano@maildoshishaacjp [4], [5] [6] 2 Pitman-Yor 3 Pitman-Yor 1 21 Pitman-Yor
fiš„v8.dvi
(2001) 49 2 333 343 Java Jasp 1 2 3 4 2001 4 13 2001 9 17 Java Jasp (JAva based Statistical Processor) Jasp Jasp. Java. 1. Jasp CPU 1 106 8569 4 6 7; fuji@ism.ac.jp 2 106 8569 4 6 7; nakanoj@ism.ac.jp
2015 3
JAIST Reposi https://dspace.j Title ターン制ストラテジーゲームにおける候補手の抽象化 によるゲーム木探索の効率化 Author(s) 村山, 公志朗 Citation Issue Date 2015-03 Type Thesis or Dissertation Text version author URL http://hdl.handle.net/10119/12652
和文タイトル
Twitter A Proposal of a Topic Transition Analysis System for Tweets 1 1 1 Center for Information and Communication Technology, Hitotsubashi University Abstract: In this paper, we propose an interactive
PFI
PFI 23 3 3 PFI PFI 1 1 2 3 2.1................................. 3 2.2..................... 4 2.3.......................... 5 3 7 3.1................................ 7 3.2.................................
1 Web [2] Web [3] [4] [5], [6] [7] [8] S.W. [9] 3. MeetingShelf Web MeetingShelf MeetingShelf (1) (2) (3) (4) (5) Web MeetingShelf
![1 Web [2] Web [3] [4] [5], [6] [7] [8] S.W. [9] 3. MeetingShelf Web MeetingShelf MeetingShelf (1) (2) (3) (4) (5) Web MeetingShelf](/thumbs/79/80104027.jpg "1 Web [2] Web [3] [4] [5], [6] [7] [8] S.W. [9] 3. MeetingShelf Web MeetingShelf MeetingShelf (1) (2) (3) (4) (5) Web MeetingShelf")
1,a) 2,b) 4,c) 3,d) 4,e) Web A Review Supporting System for Whiteboard Logging Movies Based on Notes Timeline Taniguchi Yoshihide 1,a) Horiguchi Satoshi 2,b) Inoue Akifumi 4,c) Igaki Hiroshi 3,d) Hoshi
IPSJ SIG Technical Report Vol.2015-MUS-107 No /5/23 HARK-Binaural Raspberry Pi 2 1,a) ( ) HARK 2 HARK-Binaural A/D Raspberry Pi 2 1.
HARK-Binaural Raspberry Pi 2 1,a) 1 1 1 2 3 () HARK 2 HARK-Binaural A/D Raspberry Pi 2 1. [1,2] [2 5] () HARK (Honda Research Institute Japan audition for robots with Kyoto University) *1 GUI ( 1) Python
2015 9
JAIST Reposi https://dspace.j Title ウェブページからのサイト情報 作成者情報の抽出 Author(s) 堀, 達也 Citation Issue Date 2015-09 Type Thesis or Dissertation Text version author URL http://hdl.handle.net/10119/12932 Rights Description
IPSJ SIG Technical Report iphone iphone,,., OpenGl ES 2.0 GLSL(OpenGL Shading Language), iphone GPGPU(General-Purpose Computing on Graphics Proc
iphone 1 1 1 iphone,,., OpenGl ES 2.0 GLSL(OpenGL Shading Language), iphone GPGPU(General-Purpose Computing on Graphics Processing Unit)., AR Realtime Natural Feature Tracking Library for iphone Makoto
,,, 2 ( ), $[2, 4]$, $[21, 25]$, $V$,, 31, 2, $V$, $V$ $V$, 2, (b) $-$,,, (1) : (2) : (3) : $r$ $R$ $r/r$, (4) : 3
![,,, 2 ( ), $[2, 4]$, $[21, 25]$, $V$,, 31, 2, $V$, $V$ $V$, 2, (b) $-$,,, (1) : (2) : (3) : $r$ $R$ $r/r$, (4) : 3](/thumbs/91/104771717.jpg ",,, 2 ( ), $[2, 4]$, $[21, 25]$, $V$,, 31, 2, $V$, $V$ $V$, 2, (b) $-$,,, (1) : (2) : (3) : $r$ $R$ $r/r$, (4) : 3")
1084 1999 124-134 124 3 1 (SUGIHARA Kokichi),,,,, 1, [5, 11, 12, 13], (2, 3 ), -,,,, 2 [5], 3,, 3, 2 2, -, 3,, 1,, 3 2,,, 3 $R$ ( ), $R$ $R$ $V$, $V$ $R$,,,, 3 2 125 1 3,,, 2 ( ), $[2, 4]$, $[21, 25]$,
II (Percolation) ( 3-4 ) 1. [ ],,,,,,,. 2. [ ],.. 3. [ ],. 4. [ ] [ ] G. Grimmett Percolation Springer-Verlag New-York [ ] 3
![II (Percolation) ( 3-4 ) 1. [ ],,,,,,,. 2. [ ],.. 3. [ ],. 4. [ ] [ ] G. Grimmett Percolation Springer-Verlag New-York [ ] 3](/thumbs/94/121840826.jpg "II (Percolation) ( 3-4 ) 1. [ ],,,,,,,. 2. [ ],.. 3. [ ],. 4. [ ] [ ] G. Grimmett Percolation Springer-Verlag New-York [ ] 3")
II (Percolation) 12 9 27 ( 3-4 ) 1 [ ] 2 [ ] 3 [ ] 4 [ ] 1992 5 [ ] G Grimmett Percolation Springer-Verlag New-York 1989 6 [ ] 3 1 3 p H 2 3 2 FKG BK Russo 2 p H = p T (=: p c ) 3 2 Kesten p c =1/2 ( )
IPSJ SIG Technical Report Vol.2013-ICS-172 No /11/12 1,a), 1,b) Anomaly Detection 1. 1 Nagoya Institute of Technology 1 Presently with Nagoya In
1,a), 1,b) Anomaly Detection 1. 1 Nagoya Institute of Technology 1 Presently with Nagoya Institute of Technology a) otsuka.takanobu@nitech.ac.jp b) ito.takayuki@nitech.ac.jp Anomaly Detection 2 3 4 5 6
23 The Study of support narrowing down goods on electronic commerce sites
23 The Study of support narrowing down goods on electronic commerce sites 1120256 2012 3 15 i Abstract The Study of support narrowing down goods on electronic commerce sites Masaki HASHIMURA Recently,