Article-cluster size (# of articles)
new weight
old weight
lower/upper
Articles count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
2
0.00
3.13
0.16/0.00
2.00
0.39
0.20
191.00
1.00
0.00
All clustering variants
Old ranking - len = 2
2 ■ 0.99/0.01 ■ 2 ■ 4.1
Best clustering variants
2 0.99/0.01 2 4.1
Parameters [4, 0.99, 0.01, 2] average weight 4.1 execution time 0.0s
NEW: semantic (merge_dist=0.35, adaptive=off) — 4 clusters
Cluster# (4)
new weight
old weight
lower/upper
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
0
10.57
18.75
0.41/0.30
3.00
1.84
0.61
732.00
2.00
0.67
1
5.73
11.74
0.00/0.20
2.00
1.50
0.75
174.00
4.00
1.00
3
0.00
0.00
0.00/0.00
1.00
0.00
0.00
60.00
1.00
0.00
2
0.00
0.00
0.00/0.00
1.00
0.00
0.00
81.00
4.00
0.00
Sentence-cluster ID (count= 4)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
0
0.41/0.30
10.57
18.75
3.00
1.84
0.61
732.00
2.00
0.67
Simi | Wt/Ln | TimeC | Datetime | S# | S-Id | Text
0.85 | 13.13 | 2.00 | 26-10-13 10:40 | 2 | 5 | Нето задолжувањето на развиените економии треба да достигне 5,2 проценти од БДП, додека де...ина, предупреди Меѓународниот монетарен фонд.
0.85 | 13.13 | 2.00 | 26-10-13 10:40 | 3 | 6 | Нето задолжувањето на развиените економии треба да достигне 5,2 проценти од БДП, додека де... ќе се приближи до нивото од 100 проценти од.
0.14 | 2.20 | 1.00 | 26-10-13 10:40 | 1 | 4 | Оваа година, водечките светски економии би можеле да ги забележат најголемите буџетски деф...ина, предупреди Меѓународниот монетарен фонд.
Sentence-cluster ID (count= 4)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
1
0.00/0.20
5.73
11.74
2.00
1.50
0.75
174.00
4.00
1.00
Simi | Wt/Ln | TimeC | Datetime | S# | S-Id | Text
0.75 | 8.75 | 1.00 | 26-10-14 08:16 | 3 | 2 | Притисокот за зголемување на одбранбените трошоци е уште Тековно покривање на најновите економски и финансиски вести.
0.75 | 8.75 | 1.00 | 26-10-14 08:16 | 5 | 3 | Притисокот за зголемување на одбранбените трошоци е уште.
Sentence-cluster ID (count= 4)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
3
0.00/0.00
0.00
0.00
1.00
0.00
0.00
60.00
1.00
0.00
Sentence-cluster ID (count= 4)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
2
0.00/0.00
0.00
0.00
1.00
0.00
0.00
81.00
4.00
0.00
OLD: TF-IDF K-Means (k=2) — 2 clusters
Cluster# (2)
new weight
old weight
lower/upper
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
0
8.03
18.30
0.83/0.30
4.00
1.96
0.49
813.00
2.50
1.25
1
4.16
13.11
0.43/0.20
3.00
1.62
0.54
234.00
3.00
2.67
Sentence-cluster ID (count= 2)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
0
0.83/0.30
8.03
18.30
4.00
1.96
0.49
813.00
2.50
1.25
Sentence-cluster ID (count= 2)
lower/upper
new weight
old weight
Sentence count
Total similarity
Avg. similarity
Total chr length
Sentence # average
Sentence # variation
1
0.43/0.20
4.16
13.11
3.00
1.62
0.54
234.00
3.00
2.67