Title: ?? (Search)
1??? ??? ?????? 9 ? ??
2?? (Search)
- ???? ??? ????? ???? ?? ? ?? ?? ??? ?? ??? ????
?? - ??? ?? ??? ?????? ??? ??
- ?? ?? ??? ???? ?? ?? ????? ??
- ???? ????? ??? ??? ???? ?????? ??? ??? ????
3??
- ????? ??? ?? ??? ??? ?? ??? ??
- ??????? ??? ??
- ???? ????? ??
- ?? ??? ????
- ??? ?? ???? ??? ???? ???? ?? ???? ? ?? ??? ??
??? ??? ?? - ??? ??
- Uninformed, weak method ??
- Well-informed ?
- ????? ????
- ????? ??? ???? ??? ??? ???? ????? ???? ??? ????
?? ? ?? ???? ?? ??? ?? ? ???? ?? ????? ?? - ????(Heuristic)? ??
- ????? ?? ?? ??? ?? ?? ??? ? ?? ?? ??? ?? ?? ???
??? ??? ??? ?????? ??? - ?? ??? ??? ??? ?? ?? ??? ???? ??? ??? ??? ??
????? ??????? ? - ??? ???? ?? ?? ???? ????? ??? ??? ?? ??? ?? ???
????? ??? ???? ?? ??? ?? ??? ? ?? ?? - ???? ????? ???? ?? ??
4??? ?? ?? (1)
- ??
- A ???? B ??? ??? ? ? ????
- ?? ? ? ???, ?? ???? ? ? ???, ? ? ?? ? ?? ???? ? ?
?? ?? - ??
- ???? ?????? ???? ????? ??
- ????? ??
- ? ??? ??(node), ???? ??(edge)
- aij 1, ?? i? j? ??? ??
- aij 0, ?? i? j? ???? ?? ??
??
??
??
??
??
5??? ?? ?? (2)
- ?? ??? ??? ??
- ?? ?? ?? ?? ??
- ?? 0 1 0 1 1
- ?? 1 0 1 1 0
- ?? 0 1 0 1 0
- ?? 1 0 1 0 0
- ?? 1 0 0 0 0
- ?? ?
- ?? A? B ??? ?? ???? ? ? ?? ??? ??????
- ?? ???? ???? ?????? ??? ?? ??? 2? ?? A2? ???.
- A2 A A
- Aij2 (ai1 a1j) v (ai2 a2j) v v (ain
anj) - ??? ??? ??? ? ???? ?? ??? ???? ???? ????.
- ???? ???? ??? ?? ?? ??? ???? ???!
6?? ??
- ??? ?? ????
- 12 ????? ?? ??? ? ? ??.
- ??? ? ?? ????? ??? ??
- ??? ??
- ?? ?? ?? ??? ??? ?? ?????
- ????? ??? ???? ?? ??
- ????? ??? ??(?? ?? ??? ????? ?? ?? ????)
- ????? ??
- ?? ??? ???? ?? ????? ???? ?? ??
7?? ??? ??
d1
????? ?? ??? ??
d2
?? ?? m(d1,p1) d1? p1??
???. m(d1,p2) d1? p2? ???. m(d1,p3) d1?
p3? ???. m(d2,p1) d2? p1?? ???. m(d2,p2)
d2? p2? ???. m(d2,p3) d2? p3? ???.
p1
p2
p3
(a) ????
p1
p2
p3
(b) ????
8??? ?? ?? ??
- ??? ?? ???? ?? ????? ?? ?????? ?? ????? ???? ????
- ???? ?? ?? ?? ??? ???? ?? ????? ??
- ??? ??? ??? ????? ??? ??? ?? ?? ???? ??? ??? ??
????? - ????? ??? ???? ??? ??? ?? ?? ??, ??? ???? ?? ??.
9????(state space)
- ?? ??? ?????? ??? ??(??)
- ??? ??? ????
- ????? ??? ??? ???? ????? ??
- ???? ((d1,d2)()())
- ???? (()()(d1,d2))
- ????? ???(??) ?? ? ?? ?? ? ????
- ?????? ?????? ???? ??
- ??? ??? ????? ???? ??
- ????? ??? ???? ?? ?? ? ?????
- ??? ??? ??, ????? ?? ???
10?? ?? ?
((d1,d2)()())
m(d1,p2)
m(d1,p3)
((d2)(d1)())
((d2)()(d1))
m(d2,p2)
m(d2,p3)
m(d1,p1)
m(d1,p1)
m(d1,p2)
m(d1,p3)
((d1,d2)()())
((d2)()(d1))
(()(d1)(d2))
((d2)(d1)())
((d1,d2)()())
(()(d2)(d1))
11 ?? ??? ?
((d1,d2)()())
m(d1,p1)
m(d1,p1)
m(d1,p2)
m(d1,p3)
m(d1,p3)
((d2)(d1)())
((d2)()(d1))
m(d1,p2)
m(d2,p1)
m(d2,p3)
m(d2,p2)
m(d2,p1)
(()(d1)(d2))
(()(d2)(d1))
12????
- ??? ????
- ?? ??? ????
- ?? ??? ??? ??.
- ??? ?? ??? ??? ??.
- ?? ??? ???? ??? ??? ??.
- ?? ???? ??? ? ?? ?? ??
- ?? ??(path finding) ??
- 8-puzzle
- ??(game) ??
- chess/??
- ???? ??(constraint satisfaction) ??
- 8-queen
13????
- ??? ?? (random search or blind search)
- ???? ?? ??
- ????? ??? ????? ?????,
- ?? ??? ??(??? ? ??) ???? ??? ?? ??.
- ??? ?? ??
- ???? ????
- ????(depth-first) ??, ??????(breadth-first) ??
-
1
1
a
a
5
3
2
2
b
c
b
c
3
4
6
7
4
5
6
7
d
e
f
g
d
e
f
g
14???? ??(depth-first search DFS)
- ?? ??? ?????? ?? ?? ??? ????? ?? ??? ???
??(backtracking? ??) - ??
- ????? ??? ??? ??
- ????? ?? ??? ?? ?? ?? ?? ?? ?? ??
- ??
- ?? ?? ??? ?? ?? ??(depth bound??)
- ?? ??? ??? ??? ?? ??? ?? ?? ??? ??? ??? ??
- OPEN ? ????, CLOSED ? ,
- OPEN ??? ???? ? ??? ?? ??? ????
- OPEN ??? ??? ?? E? ????, OPEN ????? ????. E? ????
??? ??? ????. - E? ?????? ??? ???? ??E? ?????.
- E? ????? ??? ?? ????? ???? E? ????? ?? ???? ?, E?
CLOSED ???? ?????. - ??????? ???? ?? E? ???? ??? OPEN ?? CLOSED ??? ??
?? ???? ??? ??? ????? OPEN ??? ???? ?????.
15???? ??(breadth-first search BFS)
- ????? ?????? ????? ?? ??? ???? ????? ??? ??? ???
?? - ??
- ?? ??? ??? ??? ???? ????? ??
- ?? ???? ??? ?? ? ??
- ??? ?? ?? ?? ??? ?? ??? ? ??
- ??
- ??? ?? ???? ????? ??????
- ????? ?? ??? ??
- OPEN ? ????, CLOSED ? ,
- OPEN ??? ???? ? ??? ?? ??? ????
- OPEN ??? ??? ?? E? ????, OPEN ????? ????. E? ????
??? ??? ????. - E? ?????? ??? ???? ??E? ?????.
- E? ????? ??? ?? ????? ???? E? ????? ?? ???? ?, E?
CLOSED ???? ?????. - ??????? ???? ?? E? ???? ??? OPEN ?? CLOSED ??? ??
?? ???? ??? ??? ????? OPEN ??? ???? ?????.
16??? ??
- ????(forward reasoning)
- ?????? ????? ??
- ????(backward reasoning)
- ?????? ????? ??
- ??? ??? ??? ?? ????
- ?? ??? ???? ??? ??
- ? ??? ?? ??? ???? ? ??
-
???
????
????
????
????
???
????
??
????
????
??
????
????
?????
?????
17????(Heuristic) ??
- ????? ?? ????? ??? ? ??? ???? ??? ?? ????? ?? ??
? ????? ??? ?? ?? ?? ??? ?? ?? - ??
- ???? ?? ?? ?) ????, ????
- ???? ??(blind search)?? ???? ????? ??
- ??? ????? ??? ??????
- ??? ???? ???, ??? ?? ??? ? ??
- ?? ??? ???? ???? ??? ????
- ?) TSP(Traveling Salesman Problem)
- n?? ?? ?? ?? ? (n-1)!/2
- n3 ? 3, n20 ? 60,822,550,204,416,000
- ? 2.2 ??? ?? ?
- ?? ???? ?? ??? ???? ?? ??
18??????(hill-climbing)
- ????(evaluation function or objective function)
?? - ?????? ??(??)??? ???? ??? ????
- ?????(valley declining)
- ???? ????? ????? ??? ??
- ??? ??? ?? ??? ??
- ????? ??? ? ??(plateau)
- ?? ?? ???(irrevocable)
- ?) 8-?? ??
-
198-????? ??
1
2
3
- ?????
- ????? ?? ??? ?? ?? ?
1
8
4
5
7
6
5
2
3
2
3
2
2
8
3
1
8
4
1
4
1
8
4
6
5
4
7
6
5
7
6
5
7
6
5
3
3
2
3
2
1
8
4
1
8
4
7
5
7
6
5
7
6
5
4
1
3
2
3
1
2
3
2
7
8
4
1
8
4
8
4
6
6
8
6
5
7
6
5
7
6
5
20??????? ??
2
8
3
1
4
5
7
6
5
2
8
3
2
3
2
8
3
2
8
3
1
4
1
8
4
1
4
1
6
4
5
4
5
4
7
6
5
7
6
5
7
6
5
7
5
21Local Search
- Simulated Annealing
- Tabu Search
- Genetic Algorithm
22??????(best-first)
- ?? ?? ??? ???? ???? ?? ???? ??
- ?? ? ? ??? ?? ?? ??
- ????? ??? ??? ????
- ???? ??? ?? ????? ????
- ??? ??? ??? ? ??
- ?? ?? ? ? ??? ? ??? ?? ? ?? ?? ?? ??
23Best-first ?? ????
- OPEN ? ????, CLOSED ? ,
- OPEN ??? ???? ? ??? ?? ??? ????
- OPEN ??? ?? ?? ???? ?? ?? Sc? ????.
- Sc? ?????? ??? ???? Sc? ??? ??? ?????.
- Sc? ??????? ???? ???? SN1, SN2, SN3,,SNk? ????.
- Sc? ?? ?? SNj (1 lt j lt k)? ?? ??? ?? ????.
- SNj? OPEN ?? CLOSED ??? ???? ?? ??? SNj ???? ???
? OPEN ??? ?????. - SNj? ?? OPEN ??? ???? ?? ?? ? SNj ??? ????
OPEN??? ?? SNj ??? ????? ? ?? ??, ?? OPEN ??? ??
SNj ??? ???? ??? ??? ?????. - SNj? CLOSED ??? ???? ?? ?? ? SNj ??? ????
CLOSED ??? ?? SNj ??? ????? ? ?? ??, CLOSED ???
?? SNj ??? OPEN ???? ????? ??? ????? ????. - Sc? CLOSED ???? ?????.
24???? ??? ? (?? 2.12)
s
s
2
2
a
b
c
a
b
c
4
1
5
4
1
5
d
e
2
3
s
s
2
2
a
b
c
a
b
c
4
1
5
4
1
5
d
e
f
g
d
e
f
g
2
3
5
6
2
3
5
6
h
i
j
5
3
8
25A ????
- ? ??(beam search)
- ???????? ????? ?? ???? ??
- ????? ????? ?? ?? ????? ??
- ???? ????
- ??? ?? ?????? ??????? ?? ??
- ?????? ? ??? ???? ?? ??
- f(n)? ??? ??? ????? ??? ??? ??? ?????? ????? ???
??? ??? ??, ??? ??? ??? ???? ?? - ??? ?? N? ?? ???? ????
- f(n) g(n) h(n)
- g(n) ?????? n ??? ?????? ??? ??
- h(n) n ???? ????? ?????? ???? ??? ???
- h(n) ????? ?????? ?? ??????
- A ????
- Best-first ?? ????? ???? ????? f(n)? ??? ??
- ? ??? ???? ?? ?? ??? ???? ?? ? ??? ???? ??? ??.
26A ????
- A ?????? ?? N? ?? h(N) ? h(N)? ????? ?? ???? ??
? A ???? - ???(admissibility) ??? ??? ???? ??
- f(N) g(N)? ??(h(N)0), ??? ??? ??
- ????? ?????? ???? ?? ? ?? ?? ??? ?? ?? ? BFS
- BFS? ?? ??? ????? ?? ?? ??
- BFS? ???? ? ?? ? ???? ??? ??? ??? ??? ?? ????
???? ???? ?? - (cf. 8-puzzle?? ?? n??? ?? ??)
- f(N) g(N) W(N)
- ??? W(N)? ????? ?? ??? ??? ??
272 8 3
1 6 4
7 5
?? A f(A)4, h(A) 4
g(n)0
2 8 3
1 6 4
7 5
2 8 3
1 6 4
7 5
2 8 3
1 4
7 6 5
g(n)1
?? D f(D)6, h(D) 5
?? B f(B)6, h(B) 5
?? C f(C)4, h(C) 3
2 8 3
1 4
7 6 5
2 3
1 8 4
7 6 5
2 8 3
1 6 4
7 5
g(n)2
?? E f(E)5, h(E) 3
?? F f(F)5, h(F) 3
?? G f(G)6, h(G) 4
2 8 3
1 4
7 6 5
2 8 3
7 1 4
6 5
2 3
1 8 4
7 6 5
2 3
1 8 4
7 6 5
g(n)3
?? H f(H)6, h(H) 3
?? I f(I)7, h(I) 4
?? J f(J)5, h(J) 2
?? K f(K)7, h(K) 4
1 2 3
8 4
7 6 5
g(n)4
?? L f(L)5, h(L) 1
1 2 3
8 4
7 6 5
1 2 3
7 8 4
6 5
g(n)5
?? M f(M)5, h(M) 0
?? N f(N)7, h(N) 2
28OPEN, CLOSED
- ???? A ()?? ?? ? ??? ???
- 1. OPEN ? A(4), CLOSED ?
- 2. A ?? OPEN ? B(6), C(4), D(6), CLOSED ?
A(4) - 3. C ?? OPEN ? B(6), D(6), E(5), F(5), G(6),
CLOSED ? C(4), A(4) - 4. E ?? OPEN ? B(6), D(6), F(5), G(6),H(6),
I(7), CLOSED ? E(5), C(4), A(4) - 5. F ?? OPEN ? B(6), D(6), G(6),H(6),
I(7),J(5),K(7), CLOSED ? F(5), E(5), C(4),
A(4) - 6. J ?? OPEN ? B(6), D(6), G(6),H(6), I(7),
K(7),L(5), CLOSED ? J(5), F(5), E(5), C(4),
A(4) - 7. L ?? OPEN ? B(6), D(6), G(6),H(6), I(7),
K(7),M(5), N(7), CLOSED ? L(5), J(5), F(5),
E(5), C(4), A(4) - 8. M ?? ???? ??, ???? ??
29???? ??
- ??? ?? ??
- ??(2?)? ?? ???? ???? ???? ?? ??? ??
- ?? ?????? ?? ??? ??? ????
- ???? ???? ??? ????? ??? ??? ? ????? ??? ?? ??? ??
?? - ??(last-one-loses) ??
- ????
- ?? 2.13 ??
30???? ??
- ????(minimax) ???
- ????? ????? ???? ??? ???? ??? ??? ??
- ? ? ?? ?????? ??? ?? ??
- ?????? ???? ??? ??? ?? ??
- ?? ??? ? ???? ???? ???? ??? ??? ???? ??? ?? ???
? ??? ?? ???? ?? - ? ?-? pruning(??-?? ????)
31???? ?? ? (1)
- ???? ?? ??? ???? ???? A? ??
A
c1 f0.8
c2 f0.3
c3 f-0.2
32???? ?? ? (2)
A
c1 f0.8
c2 f0.3
c3 f-0.2
????(B) ??
c11 f0.9
c12 f0.1
c13 f-0.6
c31 f-0.1
c32 f-0.3
c21 f0.1
c22 f-0.7
33???? ????
- ??? ???? ??? ??? ?(?? ?)? ???? ???? ??? ??? ?(??
?)? ??? ?? ??? - ????? DFS? ?? ??
-
c0 ?0.2
c1 f0.2
c2 f -0.1
c21 f -0.1
c22
c23
c11 f0.2
c12 f0.7
C21? ??? -0.1? C2? ???? ??? ???(C22, C23)? ? ??
??? ??? ??
34???? ?? ?? (1)
- ?? ??? ???? ??? ??-?? ?? ??? ????
35???? ?? ?? (2)
????? ???? ?? 1. ?? ?????? ???? ???? ??(????)?
?? ?? ??? ??? ????? ??? ?? ?, ? ??? ??? ????
??. 2. ?? ?????? ???? ???? ??(????)? ?? ?? ???
??? ????? ??? ?? ?, ? ??? ??? ???? ??. 3. ????
?????? ???? ????? ??? ?(backed-up value)? ????.
36?????? ??(Constraint Satisfaction Problem)
- n?? ??? ??? CSP ??
- ???? ???? ???? D1, D2, ,Dn???? ?? ??? ? ?? ???
???? pk(xk1,xk2,xkl)? ???? ?? - ??? NP-complete ??
- 8-Queen ??
- ?? xi, xj? ????? ????
x2 7
37????? ??
- ?????, ???? ????
- NASA? GPSS(ground processing scheduling system)
- ??? ???? ???
- ??
- chess - Deep Thought
- ?? - ?? ??? ?? (?)
- ??? ??? ???? ??? ??? ????? ??? ??
- ?? ??? ?? ??? ?? ?? ? ?? ??? ?? ??? ?? ???
38?? ??? ??? ??? ?? ?? ?? ??? ?? ??
- ??? (????? ????? ????)
- ??? (????? ????)
- ??? (????? ????????)
39Objective
- Marketing Intelligence Solution (MIS) ??
- ??? ?? ?? ??
- ??? ?????
- ??? ????? ??
- ?? ??? ??, ????, ??? ?
- ??
- ??? ??
- ?????? ??
- ?????
- ?????
40???
??? NG
???
??? OK
?? ??(???) ??
?? ??
- ???? ??
- ????
- ???? Format Setting ? Option ??
- ?? ?? Link
- Tabu Search/ Genetic Algorithm
???? ???? ??
?????
????
STP
??????????
?? Running
???? Running
??? ??
- ??? View
- ??? ?? ??(????)
- ??? ??
??? ??
- ??? View
- ??? ?? ??(????)
- ??? ??
41Model 1 ??? ????
- Lat a0 a1 GRPt a2 DISTt
- Lat ??? ???,
- ?, Lat ln (1 - At) / (1 - At-1)
- At t? ???(),
- ?, At 1 - exp(Lat) x (1 - At-1)
- GRPt TV ??? ?? Reach Frequency 100
- DISTt ?? ???? ???? ??? ???? ???? ???? ??Â
42Model 2 ?????? ????
- Kit (hit lit)/2 mi
- Â
- Kit ???? ??? ()
- hit pit Vit
- lit pit Uit
- Vit 1 / (1 exp(-citdi))
- Uit 1 / (1 exp(-bitei))
- pit ?? ?? i? ?? ???? ????
- cit i?? ????? ?? t?? ????? ?? ??? ??
- bit i?? ????? ?? t?? ???? ??? ??
- mi ??? - ??(0.25), ??(0.46) (??)
43Model 3 ????? ????
- Tit Kit x At x DISTt fi
- Â
- Tit i?? ????? ?? t?? ????? ?? ????? ?????
- Kit ???? ??? ()
- At t? ???()
- DISTt ?? ???? ???? ??? ???? ???? ???? ??
44Model 4 ????? ????
- TVit TMt x Tit x Qit x µit
- Â
- TVit i?? ????? ?? t?? ????? ?? ?? ??? ??? ? ??
??? - TMit ??????? t?? ?? ????? ?? ?
- Tit i?? ????? ?? t?? ?? ????? ??? ?????()
- Qit i?? ???? t? ???? ? 1?? ??? ?????
45?? ?? ????? ????
- Model 1 ??? ????
- Ln((1 - At) / (1 - At-1)) a0 a1 GRPt a2
DISTt - Model 2, 3, 4 ????? ????
- TVit TMt x (pit x (1 / (1 exp(-citdi)) 1 /
(1 exp(-bitei))) / 2 mi) x At x DISTt fi) x
Qit x µit
46Model Update
- Model 1
- ???? Ln((1 - At) / (1 - At-1))
- ???? GRPt, DISTt
- ?? a0, a1, a2
- ?? Update Linear
- Regression
- Model 2, 3, 4
- ???? TVit
- ???? TMt, pit, cit, bit, At, DISTt, Qit
- ?? di, ei, fi, µit
- ?? mi
- ?? Update Non-linear
- ?? Simulated Annealing, Tabu Search, Genetic
Algorithm, - ?? ??
- Artificial Neural Network
47Automatic Model Update
???
48Simulated Annealing
- s ? s0 e ? E(s) // Initial state, energy.
- sbest ? s ebest ? e // Initial "best
- solution k ? 0 // Energy evaluation count.
- while k lt kmax and e gt emax // While time left
not good enough - snew ? neighbour(s) // Pick some neighbor.
- enew ? E(snew) // Compute its energy.
- if enew lt ebest then // Is this a new best?
- sbest ? snew ebest ? enew // Save 'new
neighbor' to 'best found'. - if P(e, enew, temp(k/kmax)) gt random() then //
Should we move to it? - s ? snew e ? enew // Yes, change state.
- k ? k 1 // One more evaluation done
- return sbest // Return the best solution found.
49Neighbor Structure
(, , , ) (, , , 0) (, , , -) (, , 0, ) (, , 0, 0) (, , 0, -) (, , -, ) (, , -, 0) (, , -, -) (, 0, , ) (, 0, , 0) (, 0, , -) (, 0, 0, ) (, 0, 0, 0) (, 0, 0, -) (, 0, -, ) (, 0, -, 0) (, 0, -, -) (, -, , ) (, -, , 0) (, -, , -) (, -, 0, ) (, -, 0, 0) (, -, 0, -) (, -, -, ) (, -, -, 0) (, -, -, -) (0, , , ) (0, , , 0) (0, , , -) (0, , 0, ) (0, , 0, 0) (0, , 0, -) (0, , -, ) (0, , -, 0) (0, , -, -) (0, 0, , ) (0, 0, , 0) (0, 0, , -) (0, 0, 0, ) (0, 0, 0, 0) (0, 0, 0, -) (0, 0, -, ) (0, 0, -, 0) (0, 0, -, -) (0, -, , ) (0, -, , 0) (0, -, , -) (0, -, 0, ) (0, -, 0, 0) (0, -, 0, -) (0, -, -, ) (0, -, -, 0) (0, -, -, -) (-, , , ) (-, , , 0) (-, , , -) (-, , 0, ) (-, , 0, 0) (-, , 0, -) (-, , -, ) (-, , -, 0) (-, , -, -) (-, 0, , ) (-, 0, , 0) (-, 0, , -) (-, 0, 0, ) (-, 0, 0, 0) (-, 0, 0, -) (-, 0, -, ) (-, 0, -, 0) (-, 0, -, -) (-, -, , ) (-, -, , 0) (-, -, , -) (-, -, 0, ) (-, -, 0, 0) (-, -, 0, -) (-, -, -, ) (-, -, -, 0) (-, -, -, -)
- Neighbor of di, ei, fi, µit
- E.g.
- Current
- 1.0, 1.0, 1.0, 1.0
- Precision
- 0.1
- Neighbor
- , , 0, -
- 1.1, 1.1, 1.0, 0.9
50Evaluation
- ??? Tvit
- ??? Tvit' TMt x (pit x (1 / (1
exp(-citdi)) 1 / (1 exp(-bitei))) / 2 mi)
x At x DISTt fi) x Qit x µit - ???
- ((Tvit - Tvit')2 / n)1/2
- n number of data
51Tabu Search
- Notation
- o    S, the current solution,
- o    S, the best-known solution,
- o    f, value of S,
- o    N(S), the neighborhood of S,
- o    Ñ(S), the admissible subset of N(S)
(i.e., non-tabu or allowed by aspiration). - Initialization
- Choose (construct) an initial solution S0.
- Set S S0 , f f(S0), S S0 , T Ø.
- Search
- While termination criterion not satisfied do
- o      Select S in argmin f(S') for S'e Ñ(S)
- o      If f(S) lt f, then set f f(S), S
S - o      Record tabu for the current move in T
(delete oldest entry if necessary) - Endwhile.
52Genetic Algorithm
- Choose the initial population of individuals
- Evaluate the fitness of each individual in that
population - Repeat on this generation until termination
(time limit, sufficient fitness achieved, etc.) - Select the best-fit individuals for reproduction
- Breed new individuals through crossover and
mutation operations to give birth to offspring - Evaluate the individual fitness of new
individuals - Replace least-fit population with new individuals
53Genetic Algorithm for MIS
- Size of population 100
- Fitness ??? ((Tvit - Tvit')2 / n)1/2
- Crossover
- di' (diF diM) / 2
- ei' (eiF eiM) / 2
- fi' (fiF fiM) / 2
- µit' (µitF µitM) / 2
- Mutation
- Normal(0, precision)
54Experiment
- 20?? ??? ?? ?? ??
- ?? ?? ? 100?? ?? ??
- ??, ???? ?? ??
- ?? mi 0.46
- ???? ??
- TVit TMt x (pit x (1 / (1 exp(-citdi)) 1 /
(1 exp(-bitei))) / 2 mi) x At x DISTt fi) x
Qit x µit - 100? ? ??? 50?, ??? 50? ??
- ???
- ((Tvit - Tvit')2 / n)1/2
55Experiment Simulated Annealing
?? No. ??? ??? ??
di ei fi µit di ei fi µit sqrt(ESS/n)
1 0.9365 0.4382 0.0436 0.7803 0.9365 0.4382 0.0436 0.7803 0.00
2 0.9199 0.8765 0.7002 0.9718 0.9199 0.8765 0.7002 0.9718 0.02
3 0.5142 0.5995 0.0900 0.6093 0.5142 0.5995 0.0900 0.6093 0.00
4 0.3713 0.2902 0.8729 0.1717 0.3713 0.2902 0.8729 0.1717 1.10
5 0.7137 0.3485 0.1205 0.7633 0.7137 0.3485 0.1205 0.7633 0.02
6 0.7359 0.7618 0.1242 0.1640 0.7359 0.7618 0.1242 0.1640 0.00
7 0.4120 0.2482 0.3735 0.3031 0.4120 0.2482 0.3735 0.3031 0.02
8 0.0885 0.5231 0.8920 0.5658 0.0885 0.5231 0.8920 0.5658 0.04
9 0.4185 0.8520 0.4351 0.0868 0.4185 0.8520 0.4351 0.0868 0.04
10 0.8392 0.6276 0.5864 0.7889 0.2598 0.5387 0.5915 0.8919 587,559.69
11 0.8382 0.5747 0.6915 0.5206 0.8382 0.5747 0.6915 0.5206 0.08
12 0.6600 0.7270 0.2438 0.1101 0.6600 0.7270 0.2438 0.1101 3.00
13 0.5018 0.1308 0.6828 0.9445 0.5592 1.0000 0.6870 0.7471 5,914,427.69
14 0.4362 0.7303 0.9359 0.6039 0.4362 0.7303 0.9359 0.6039 0.07
15 0.2453 0.4088 0.2786 0.7923 0.2453 0.4088 0.2786 0.7923 0.01
16 0.1878 0.6660 0.7074 0.5474 0.1878 0.6660 0.7074 0.5474 0.03
17 0.9065 0.2633 0.8208 0.2182 0.9065 0.2633 0.8208 0.2182 1.22
18 0.2501 0.4298 0.4895 0.8934 0.2501 0.4298 0.4895 0.8934 0.01
19 0.1664 0.7223 0.4325 0.3814 0.1664 0.7223 0.4325 0.3814 0.18
20 0.8440 0.4600 0.8055 0.1501 0.2469 0.6389 0.8077 0.1633 649,919.28
?? 357,595.63
56Experiment Genetic Algorithm
?? No. ??? ??? ??
di ei fi µit di ei fi µit sqrt(ESS/n)
1 0.9365 0.4382 0.0436 0.7803 0.9365 0.4382 0.0436 0.7803 0.11
2 0.9199 0.8765 0.7002 0.9718 0.8922 0.8949 0.6999 0.9740 92,054.96
3 0.5142 0.5995 0.0900 0.6093 0.5142 0.5995 0.0900 0.6093 0.21
4 0.3713 0.2902 0.8729 0.1717 0.3713 0.2902 0.8729 0.1717 1.38
5 0.7137 0.3485 0.1205 0.7633 0.7137 0.3485 0.1205 0.7633 0.24
6 0.7359 0.7618 0.1242 0.1640 0.3021 0.9638 0.1225 0.1777 24,725.34
7 0.4120 0.2482 0.3735 0.3031 0.4120 0.2482 0.3735 0.3031 0.39
8 0.0885 0.5231 0.8920 0.5658 0.0885 0.5231 0.8920 0.5658 5.25
9 0.4185 0.8520 0.4351 0.0868 0.4701 0.7436 0.4402 0.0853 17,522.20
10 0.8392 0.6276 0.5864 0.7889 0.2598 0.5387 0.5915 0.8919 587,559.69
11 0.8382 0.5747 0.6915 0.5206 0.8382 0.5747 0.6915 0.5206 2.29
12 0.6600 0.7270 0.2438 0.1101 0.6458 0.6955 0.2442 0.1106 4,932.28
13 0.5018 0.1308 0.6828 0.9445 0.5018 0.1308 0.6828 0.9445 3.22
14 0.4362 0.7303 0.9359 0.6039 0.4362 0.7303 0.9359 0.6039 2.94
15 0.2453 0.4088 0.2786 0.7923 0.2453 0.4088 0.2786 0.7923 0.76
16 0.1878 0.6660 0.7074 0.5474 0.1878 0.6660 0.7074 0.5474 2.14
17 0.9065 0.2633 0.8208 0.2182 0.9384 0.9306 0.8189 0.1928 849,636.53
18 0.2501 0.4298 0.4895 0.8934 0.2501 0.4298 0.4895 0.8934 0.88
19 0.1664 0.7223 0.4325 0.3814 0.9386 0.6772 0.4235 0.3290 287,461.00
20 0.8440 0.4600 0.8055 0.1501 0.2470 0.6389 0.8077 0.1633 649,919.28
?? 125,691.56
57Summary
- Marketing Intelligence Solution ??
- ??? ?? ??? ?? ????? ??
- Model 1 ??? ?? ??
- Model 2 ?????? ?? ??
- Model 3 ????? ??
- Model 4 ????? ??
- Model 2, 3, 4 ????
- ?? Update
- Non-linear
- ?? ??? ?? ?? Update
- Simulated Annealing
- Tabu Search
- Genetic Algorithm
- ?? ? ???? ?? ?