-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathCriteriaElicitation.R
198 lines (155 loc) · 5.42 KB
/
CriteriaElicitation.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
# Author: Javier Martínez-López (javi.martinez.lopez # gmail.com)
library(googlesheets)
library(vegan)
library(ade4)
library(Prize)
library(ggplot2)
library(scales)
gs_ls()
data<-gs_title('PT AQ stakeholder assessment of ES in CS5 (Responses)')
df<-gs_read(data)
ndim<-11
# add stakeholders groups
g1<-'Política / Governança (ex: Governança Ambiental, da Pesca e Agricultura, Marinha, agências nacionais)'
g2<-'Administração Pública (ex: Administração Regional, Municípios, Freguesias)'
g3<-'Cidadãos (ex: moradores, proprietários de residências, grupos sub-representados e vulneráveis)'
g4<-'Ciência (ex: professores, investigadores ou técnicos em universidades / instituições locais ou independentes)'
g5<-'Grupos de interesse (ex: associações locais, organizações não governamentais (ONGs), organizações profissionais)'
g6<-'Negócios (ex: Indústria, Turismo, Agricultura, Pesca, pequenas empresas, empresas nacionais ou multinacionais com interesses locais)'
g7<-'Serviços (ex: Portos e transporte comercial, transporte público de passageiros)'
dfnum<-as.data.frame(df)
dfnum[df==g1]<-'PolGov'
dfnum[df==g2]<-'AdmPub'
dfnum[df==g3]<-'Cidad'
dfnum[df==g4]<-'Cienc'
dfnum[df==g5]<-'GrpInt'
dfnum[df==g6]<-'Neg'
dfnum[df==g7]<-'Serv'
mli<-'muito menos importante'#'much less important'
li<-'menos importante'#'less important'
ei<-'igualmente importante'#'equally important'
mi<-'mais importante'#'more important'
mmi<-'muito mais importante'#'much more important'
dfnum[df==mli]<-1/4
dfnum[df==li]<-1/2
dfnum[df==ei]<-1
dfnum[df==mi]<-2
dfnum[df==mmi]<-4
dff<-as.data.frame(matrix(NA,1,(ndim+1)))
names(dff)<-c('id','s1','s2','s3','s4','s5','s6','s7','s8','s9','s10','s11')
#CorrectScores<-function(){
# To be done for each individual in a loop
for(n in 1:dim(dfnum)[1]){
#n<-1
print(n)
dfnum2<-dfnum[n,-(1:2)]
a<-matrix(NA,ndim,ndim,byrow=T)
combs<-factorial(ndim)/(factorial(ndim-2)*factorial(2))
ind<-(ndim-1):1
l<-NULL
for(k in 1:length(ind)){l[k]<-sum(ind[1:k])}
init0<-l[1:(length(l)-1)]+1
init<-c(1,init0)
init[length(init)]<-combs
for(j in 1:(ndim-1)){
a[j,j]<-1
a[j,-(1:j)]<-as.numeric(dfnum2[1,init[j]:l[j]])
}
a[ndim,ndim]<-1
dfa<-as.data.frame(a)
names(dfa)<-names(dff)[2:12]
rownames(dfa)<-names(dff)[2:12]
write.table(dfa,paste('ind_',n,'.tsv',sep=''),sep='\t',quote=F)
A<-ahmatrix(a)
ea<-eigen(A@ahp_matrix)
dff[n,1]<-dfnum[n,2]
dff[n,2:(ndim+1)]<-rescale(as.numeric(ea$vectors[,1]),to=c(1,5)) # 2:6
}
mat <- matrix(nrow = 17, ncol = 1, data = NA)
mat[,1] <- c('ind_1.tsv',
'ind_2.tsv',
'ind_3.tsv',
'ind_4.tsv',
'ind_5.tsv',
'ind_6.tsv',
'ind_7.tsv',
'ind_8.tsv',
'ind_9.tsv',
'ind_10.tsv',
'ind_11.tsv',
'ind_12.tsv',
'ind_13.tsv',
'ind_14.tsv',
'ind_15.tsv',
'ind_16.tsv',
'ind_17.tsv'
)
rownames(mat) <- c('ind1','ind2','ind3', 'ind4', 'ind5', 'ind6', 'ind7', 'ind8', 'ind9', 'ind10', 'ind11', 'ind12', 'ind13', 'ind14', 'ind15', 'ind16', 'ind17')
colnames(mat) <- c('individual_judgement')
# non-weighted aggregation
res <- gaggregate(srcfile = mat, method = 'geometric', simulation = 500)
# consistency ratio of the aggregated group judgement
gcr<-GCR(res)
incons<-ICR(res)
okind<-incons<=0.15
okind[2]<-FALSE
crplot(ICR(res), angle = 45)
ggsave('individual_consistency_ratio.png')
# Distance between individual opinions and the aggregated group judgement
dplot(IP(res))
ggsave('Distance2Group.png')
#dmh<-dist(dff[,-1])
dff2<-dff[okind,]
dmh<-dist(dff2[,-1])
mds<-metaMDS(dmh)
hclust(dmh,"ward.D2")->mds_hclust
coph<-cophenetic(mds_hclust)
cophval<-cor(coph,dmh)
q25<-quantile(mds_hclust$height)[2]
q50<-quantile(mds_hclust$height)[3]
q75<-quantile(mds_hclust$height)[4]
#cutree(mds_hclust,h=q75)->mds_hclust_mean # Dissimilarity threshold
cutree(mds_hclust,k=2)->mds_hclust_mean # Number of classes
#km<-kmeans(dmh,3)
#mds_hclust_mean<-km$cluster
ncl<-length(unique(mds_hclust_mean))
png('hclust_classes.png')
plot(mds_hclust,hang=-1)
rect.hclust(mds_hclust,k=ncl)
dev.off()
#mds_km<-kmeans(dmh,3)
mds_xy <- data.frame(mds$points)
#mds_xy$cluster<-as.vector(mds_km$cluster)
mds_xy$cluster<-as.vector(mds_hclust_mean)
mds_xy$groups<-dff2$id
png('nmds_classes.png')
ggplot(mds_xy, aes(MDS1, MDS2, color = as.factor(cluster),label = rownames(mds_xy))) + geom_point() + theme_bw() + geom_label()
dev.off()
png('nmds_initial_groups.png')
ggplot(mds_xy, aes(MDS1, MDS2, color = as.factor(groups),label = rownames(mds_xy))) + geom_point() + theme_bw() + geom_label()
dev.off()
dff2$classes<-mds_hclust_mean
lig<-length(unique(dff2$id))
h<-0
z<-matrix(NA,ndim,ncl,byrow=T)
z2<-matrix(NA,ndim,lig,byrow=T)
z4<-matrix(NA,ndim,1,byrow=T)
for(n in 2:(dim(dff2)[2]-1)){
h<-h+1
x<-aggregate(dff2[,n]~classes,data=dff2, FUN=function(x) mean(x))
x2<-aggregate(dff2[,n]~id,data=dff2, FUN=function(x) mean(x))
z[h,]<-x[,2]
z2[h,]<-x2[,2]
z4[h,]<-mean(dff2[,n])
}
z3<-as.data.frame(z2)
names(z3)<-x2[,1]
z5<-as.data.frame(z)
z6<-as.data.frame(z4)
write.table(z5[-2,],'weights_new_groups.csv',sep=',',row.names=F)
write.table(z3[-2,],'weights_initial_groups.csv',sep=',',row.names=F)
write.table(z6[-2,],'weights_nogroups.csv',sep=',',row.names=F)
#png('nmds.png')
#plot(mds)
#s.class(mds$points,fac=as.factor(1:dim(dff)[1]))#,col=1:length(unique(dff$id)))
#dev.off()