-
Notifications
You must be signed in to change notification settings - Fork 1
/
final-hierarchical-measures
99 lines (80 loc) · 2.96 KB
/
final-hierarchical-measures
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
1. Summarization
summary-out-comb-top200-laura-cand-train-run
map all 0.0108
Rprec all 0.0084
recip_rank all 0.0234
2. HAC with wordnet similarity
hacsim-comb-top200-laura-cand-train-run
map all 0.0029
Rprec all 0.0016
recip_rank all 0.0054
3. HAC with word2vec vectors
hacwv-comb-top200-laura-cand-train-run
map all 0.0027
Rprec all 0.0020
recip_rank all 0.0059
4. KMeans with word2vec vectors
kmwv-comb-top200-laura-cand-train-run
map all 0.0034
Rprec all 0.0022
recip_rank all 0.0073
5. LDA topic model with wordnet synonyms
topic-model-expanded-sec-train-run
map all 0.0073
Rprec all 0.0030
recip_rank all 0.0117
6. LDA topic model
topic-model-train-run
map all 0.0060
Rprec all 0.0025
recip_rank all 0.0121
7. Datamuse + Concept net + BM25
query-expansion-data-muse-and-CN-BM25-run
map all 0.0168
Rprec all 0.0109
recip_rank all 0.0225
9. Weighted PageRank with Common Entities
map all 0.0072
Rprec all 0.0044
recip_rank all 0.0146
10. Weighted PageRank with Common Words
map all 0.0083
Rprec all 0.0054
recip_rank all 0.0167
------ Candidate sets ----------
8. Candidate Set Generation using Query Expansion with entities from DBPedia
MAP : 0.0567
Rprec : 0.0985
recip_rank : 0.3478
11. BM25+(BM25+KNN-INC)+(BM25+KNN-EXT)
map all 0.0877
Rprec all 0.1449
recip_rank all 0.4501
12. LM-DS+(LM-DS+KNN-INC)+(LM-DS+KNN-EXT)
map all 0.0933
Rprec all 0.1520
recip_rank all 0.4349
13. LM-JM+(LM-JM+KNN-INC)+(LM-JM+KNN-EXT)
map all 0.0699
Rprec all 0.1202
recip_rank all 0.3397
11. ALL
map all 0.0988
Rprec all 0.1578
recip_rank all 0.4996
12. Shubham cand + laura combined using Rlib
map all 0.1355
Rprec all 0.1947
recip_rank all 0.6721
13. Datamuse + BM25
map all 0.0183
Rprec all 0.0332
recip_rank all 0.0792
14. Added ConceptNet with DataMuse + BM25
map all 0.0586
Rprec all 0.0999
recip_rank all 0.2940
15. DataMuse + BM25 + ConceptNet for hierarchical sections
map all 0.0206
Rprec all 0.0134
recip_rank all 0.0275