Farseen0 commited on
Commit
90e8ba4
·
verified ·
1 Parent(s): 7b923be

Upload evaluation results: comparison

Browse files
Files changed (1) hide show
  1. evaluation/comparison_results.json +52 -292
evaluation/comparison_results.json CHANGED
@@ -1,339 +1,99 @@
1
  {
2
- "h_curated": {
3
  "with_idf": {
4
- "recall@1": 0.8078758716583252,
5
- "recall@5": 0.9498806595802307,
6
- "recall@10": 0.964200496673584,
7
- "recall@20": 0.9761336445808411,
8
- "mrr": 0.871401846408844,
9
- "ndcg@10": 0.8934720082918809
10
  },
11
  "no_idf": {
12
- "recall@1": 0.560859203338623,
13
- "recall@5": 0.7088305354118347,
14
- "recall@10": 0.7482100129127502,
15
- "recall@20": 0.7720763683319092,
16
- "mrr": 0.6307040452957153,
17
- "ndcg@10": 0.6560224477581484
18
- },
19
- "hybrid_no_idf": {
20
- "best_alpha": 0.5,
21
- "best_metrics": {
22
- "recall@1": 0.577565610408783,
23
- "recall@5": 0.6909307837486267,
24
- "recall@10": 0.7338902354240417,
25
- "recall@20": 0.7780429720878601,
26
- "mrr": 0.6334040760993958,
27
- "ndcg@10": 0.6522560542504828
28
- }
29
- },
30
- "hybrid_idf": {
31
- "best_alpha": 0.5,
32
- "best_metrics": {
33
- "recall@1": 0.7995226979255676,
34
- "recall@5": 0.9439140558242798,
35
- "recall@10": 0.9773269891738892,
36
- "recall@20": 0.9952267408370972,
37
- "mrr": 0.8643569350242615,
38
- "ndcg@10": 0.8908862503996242
39
- }
40
- },
41
- "bootstrap_ci": {
42
- "hybrid_idf_r10": {
43
- "mean": 0.9773269891738892,
44
- "ci_lower": 0.9665871262550354,
45
- "ci_upper": 0.9868735074996948,
46
- "se": 0.005134093975245222
47
- },
48
- "hybrid_idf_ndcg10": {
49
- "mean": 0.8908862503996242,
50
- "ci_lower": 0.8750708682822845,
51
- "ci_upper": 0.9066549773221649,
52
- "se": 0.008092329911784334
53
- },
54
- "standalone_idf_r10": {
55
- "mean": 0.964200496673584,
56
- "ci_lower": 0.9510740041732788,
57
- "ci_upper": 0.9761336445808411,
58
- "se": 0.006408560281645167
59
- }
60
- }
61
- },
62
- "h_curated_s2": {
63
- "with_idf": {
64
- "recall@1": 0.79474937915802,
65
- "recall@5": 0.9510740041732788,
66
- "recall@10": 0.9725536704063416,
67
- "recall@20": 0.9832935333251953,
68
- "mrr": 0.8646395206451416,
69
- "ndcg@10": 0.8905745725645697
70
- },
71
- "no_idf": {
72
- "recall@1": 0.4618138372898102,
73
- "recall@5": 0.6002386808395386,
74
- "recall@10": 0.638424813747406,
75
- "recall@20": 0.6587111949920654,
76
- "mrr": 0.5274040699005127,
77
- "ndcg@10": 0.5508962789654486
78
- },
79
- "hybrid_no_idf": {
80
- "best_alpha": 0.5,
81
- "best_metrics": {
82
- "recall@1": 0.5441527366638184,
83
- "recall@5": 0.6527446508407593,
84
- "recall@10": 0.7016706466674805,
85
- "recall@20": 0.7446300983428955,
86
- "mrr": 0.5978065729141235,
87
- "ndcg@10": 0.6173909723302361
88
- }
89
- },
90
- "hybrid_idf": {
91
- "best_alpha": 0.5,
92
- "best_metrics": {
93
- "recall@1": 0.8031026124954224,
94
- "recall@5": 0.9594271779060364,
95
- "recall@10": 0.9832935333251953,
96
- "recall@20": 0.9940333962440491,
97
- "mrr": 0.8701260685920715,
98
- "ndcg@10": 0.8971763733153765
99
- }
100
- },
101
- "bootstrap_ci": {
102
- "hybrid_idf_r10": {
103
- "mean": 0.9832935333251953,
104
- "ci_lower": 0.9737470149993896,
105
- "ci_upper": 0.9916467666625977,
106
- "se": 0.004433608641789609
107
- },
108
- "hybrid_idf_ndcg10": {
109
- "mean": 0.8971763733153765,
110
- "ci_lower": 0.8821949014944158,
111
- "ci_upper": 0.9123868220432041,
112
- "se": 0.007695017042453908
113
- },
114
- "standalone_idf_r10": {
115
- "mean": 0.9725536704063416,
116
- "ci_lower": 0.9606205224990845,
117
- "ci_upper": 0.9832935333251953,
118
- "se": 0.005762715650875535
119
- }
120
- }
121
- },
122
- "h_rand7552": {
123
- "with_idf": {
124
- "recall@1": 0.45704057812690735,
125
- "recall@5": 0.6360381841659546,
126
- "recall@10": 0.7315036058425903,
127
- "recall@20": 0.819809079170227,
128
- "mrr": 0.54668128490448,
129
- "ndcg@10": 0.5825007922082269
130
- },
131
- "no_idf": {
132
- "recall@1": 0.2887828052043915,
133
- "recall@5": 0.42124104499816895,
134
- "recall@10": 0.4630071520805359,
135
- "recall@20": 0.4976133704185486,
136
- "mrr": 0.35064178705215454,
137
- "ndcg@10": 0.3728723791192576
138
  },
139
  "hybrid_no_idf": {
140
  "best_alpha": 0.3,
141
  "best_metrics": {
142
- "recall@1": 0.5107398629188538,
143
- "recall@5": 0.6396181583404541,
144
- "recall@10": 0.6897374987602234,
145
- "recall@20": 0.7279236316680908,
146
- "mrr": 0.5710219144821167,
147
- "ndcg@10": 0.594996674033728
148
  }
149
  },
150
  "hybrid_idf": {
151
  "best_alpha": 0.5,
152
  "best_metrics": {
153
- "recall@1": 0.6229116916656494,
154
- "recall@5": 0.8042959570884705,
155
- "recall@10": 0.861575186252594,
156
- "recall@20": 0.9021479487419128,
157
- "mrr": 0.7051689624786377,
158
- "ndcg@10": 0.739109985922535
159
- }
160
- },
161
- "bootstrap_ci": {
162
- "hybrid_idf_r10": {
163
- "mean": 0.861575186252594,
164
- "ci_lower": 0.8377088308334351,
165
- "ci_upper": 0.8842481970787048,
166
- "se": 0.011970612565884334
167
- },
168
- "hybrid_idf_ndcg10": {
169
- "mean": 0.739109985922535,
170
- "ci_lower": 0.7137405788253148,
171
- "ci_upper": 0.7641767890478637,
172
- "se": 0.012806377598930505
173
- },
174
- "standalone_idf_r10": {
175
- "mean": 0.7315036058425903,
176
- "ci_lower": 0.7016706466674805,
177
- "ci_upper": 0.7613365054130554,
178
- "se": 0.01526008860083923
179
- }
180
- }
181
- },
182
- "h_notier4": {
183
- "with_idf": {
184
- "recall@1": 0.45942720770835876,
185
- "recall@5": 0.6694510579109192,
186
- "recall@10": 0.7613365054130554,
187
- "recall@20": 0.843675434589386,
188
- "mrr": 0.5589096546173096,
189
- "ndcg@10": 0.6001309074358909
190
- },
191
- "no_idf": {
192
- "recall@1": 0.2732696831226349,
193
- "recall@5": 0.3806682527065277,
194
- "recall@10": 0.42124104499816895,
195
- "recall@20": 0.4510739743709564,
196
- "mrr": 0.3235103487968445,
197
- "ndcg@10": 0.34176754966492734
198
- },
199
- "hybrid_no_idf": {
200
- "best_alpha": 0.2,
201
- "best_metrics": {
202
- "recall@1": 0.5107398629188538,
203
- "recall@5": 0.6348448395729065,
204
- "recall@10": 0.6885441541671753,
205
- "recall@20": 0.7279236316680908,
206
- "mrr": 0.5687209367752075,
207
- "ndcg@10": 0.5927738891990314
208
- }
209
- },
210
- "hybrid_idf": {
211
- "best_alpha": 0.5,
212
- "best_metrics": {
213
- "recall@1": 0.6420047879219055,
214
  "recall@5": 0.836515486240387,
215
- "recall@10": 0.8866348266601562,
216
- "recall@20": 0.9248210191726685,
217
- "mrr": 0.7279645800590515,
218
- "ndcg@10": 0.7632779478919515
219
- }
220
- },
221
- "bootstrap_ci": {
222
- "hybrid_idf_r10": {
223
- "mean": 0.8866348266601562,
224
- "ci_lower": 0.8651551604270935,
225
- "ci_upper": 0.9069212675094604,
226
- "se": 0.010968634185269548
227
- },
228
- "hybrid_idf_ndcg10": {
229
- "mean": 0.7632779478919515,
230
- "ci_lower": 0.7392506147251328,
231
- "ci_upper": 0.7868044559342753,
232
- "se": 0.0121621765042722
233
- },
234
- "standalone_idf_r10": {
235
- "mean": 0.7613365054130554,
236
- "ci_lower": 0.7326968908309937,
237
- "ci_upper": 0.7911694645881653,
238
- "se": 0.014857916253111171
239
- }
240
- }
241
- },
242
- "v12_e15": {
243
- "with_idf": {
244
- "recall@1": 0.8603818416595459,
245
- "recall@5": 0.9546539187431335,
246
- "recall@10": 0.9701670408248901,
247
- "recall@20": 0.9856801629066467,
248
- "mrr": 0.9037026762962341,
249
- "ndcg@10": 0.9190719076440962
250
- },
251
- "no_idf": {
252
- "recall@1": 0.5477327108383179,
253
- "recall@5": 0.70525062084198,
254
- "recall@10": 0.7315036058425903,
255
- "recall@20": 0.7541766166687012,
256
- "mrr": 0.616544246673584,
257
- "ndcg@10": 0.6413623750146198
258
- },
259
- "hybrid_no_idf": {
260
- "best_alpha": 0.5,
261
- "best_metrics": {
262
- "recall@1": 0.5465393662452698,
263
- "recall@5": 0.6873508095741272,
264
- "recall@10": 0.7315036058425903,
265
- "recall@20": 0.7684963941574097,
266
- "mrr": 0.6118062138557434,
267
- "ndcg@10": 0.6360296473664324
268
- }
269
- },
270
- "hybrid_idf": {
271
- "best_alpha": 0.5,
272
- "best_metrics": {
273
- "recall@1": 0.7959427237510681,
274
- "recall@5": 0.9546539187431335,
275
- "recall@10": 0.9844868779182434,
276
- "recall@20": 0.9940333962440491,
277
- "mrr": 0.8643589615821838,
278
- "ndcg@10": 0.8930407819276659
279
  }
280
  },
281
  "bootstrap_ci": {
282
  "hybrid_idf_r10": {
283
- "mean": 0.9844868779182434,
284
- "ci_lower": 0.9761038124561312,
285
- "ci_upper": 0.9916467666625977,
286
- "se": 0.004230828890499356
287
  },
288
  "hybrid_idf_ndcg10": {
289
- "mean": 0.8930407819276659,
290
- "ci_lower": 0.8778117702491244,
291
- "ci_upper": 0.9081692573752533,
292
- "se": 0.00784255541184393
293
  },
294
  "standalone_idf_r10": {
295
- "mean": 0.9701670408248901,
296
- "ci_lower": 0.9582338929176331,
297
- "ci_upper": 0.9809069037437439,
298
- "se": 0.005905478481174235
299
  }
300
  }
301
  },
302
  "_significance_tests": {
303
  "recall@10": {
304
- "delta": 0.2959427237510681,
305
  "p_value": 0.0,
306
  "significant": true,
307
- "ci_lower": 0.26491647958755493,
308
- "ci_upper": 0.3269689679145813
309
  },
310
  "ndcg@10": {
311
- "delta": 0.3019232174052843,
312
  "p_value": 0.0,
313
  "significant": true,
314
- "ci_lower": 0.27495923197187516,
315
- "ci_upper": 0.3285048309635248
316
  },
317
  "mrr": {
318
- "delta": 0.29798927903175354,
319
  "p_value": 0.0,
320
  "significant": true,
321
- "ci_lower": 0.2706925220787525,
322
- "ci_upper": 0.32507591545581815
323
  },
324
  "standalone_recall@10": {
325
- "delta": 0.28162291646003723,
326
  "p_value": 0.0,
327
  "significant": true,
328
- "ci_lower": 0.24821002781391144,
329
- "ci_upper": 0.31503579020500183
330
  },
331
  "standalone_ndcg@10": {
332
- "delta": 0.3279543431217147,
333
- "p_value": 0.0,
334
  "significant": true,
335
- "ci_lower": 0.295606120573298,
336
- "ci_upper": 0.3597483916727818
337
  }
338
  },
339
  "_bm25": {
 
1
  {
2
+ "h_extended": {
3
  "with_idf": {
4
+ "recall@1": 0.5011933445930481,
5
+ "recall@5": 0.715990424156189,
6
+ "recall@10": 0.7720763683319092,
7
+ "recall@20": 0.8532219529151917,
8
+ "mrr": 0.5991758108139038,
9
+ "ndcg@10": 0.6341397571131019
10
  },
11
  "no_idf": {
12
+ "recall@1": 0.3663484454154968,
13
+ "recall@5": 0.49164676666259766,
14
+ "recall@10": 0.5453460812568665,
15
+ "recall@20": 0.5680190920829773,
16
+ "mrr": 0.4280979037284851,
17
+ "ndcg@10": 0.45219292759059226
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  },
19
  "hybrid_no_idf": {
20
  "best_alpha": 0.3,
21
  "best_metrics": {
22
+ "recall@1": 0.5155131220817566,
23
+ "recall@5": 0.6455847024917603,
24
+ "recall@10": 0.6921241283416748,
25
+ "recall@20": 0.7315036058425903,
26
+ "mrr": 0.5743579864501953,
27
+ "ndcg@10": 0.5978281095200982
28
  }
29
  },
30
  "hybrid_idf": {
31
  "best_alpha": 0.5,
32
  "best_metrics": {
33
+ "recall@1": 0.6694510579109192,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
  "recall@5": 0.836515486240387,
35
+ "recall@10": 0.8818615674972534,
36
+ "recall@20": 0.9081145524978638,
37
+ "mrr": 0.7436834573745728,
38
+ "ndcg@10": 0.7742396474131288
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39
  }
40
  },
41
  "bootstrap_ci": {
42
  "hybrid_idf_r10": {
43
+ "mean": 0.8818615674972534,
44
+ "ci_lower": 0.8603818416595459,
45
+ "ci_upper": 0.9033412933349609,
46
+ "se": 0.011077928874225752
47
  },
48
  "hybrid_idf_ndcg10": {
49
+ "mean": 0.7742396474131288,
50
+ "ci_lower": 0.7506954504355015,
51
+ "ci_upper": 0.7977965140781702,
52
+ "se": 0.012106710016022492
53
  },
54
  "standalone_idf_r10": {
55
+ "mean": 0.7720763683319092,
56
+ "ci_lower": 0.7434367537498474,
57
+ "ci_upper": 0.800715982913971,
58
+ "se": 0.014512963508524598
59
  }
60
  }
61
  },
62
  "_significance_tests": {
63
  "recall@10": {
64
+ "delta": 0.19331742823123932,
65
  "p_value": 0.0,
66
  "significant": true,
67
+ "ci_lower": 0.16348448395729065,
68
+ "ci_upper": 0.2219570428133011
69
  },
70
  "ndcg@10": {
71
+ "delta": 0.1831220828907474,
72
  "p_value": 0.0,
73
  "significant": true,
74
+ "ci_lower": 0.1592187369075159,
75
+ "ci_upper": 0.20782378544404753
76
  },
77
  "mrr": {
78
+ "delta": 0.17731375992298126,
79
  "p_value": 0.0,
80
  "significant": true,
81
+ "ci_lower": 0.15270963460206985,
82
+ "ci_upper": 0.20239105299115182
83
  },
84
  "standalone_recall@10": {
85
+ "delta": 0.08353222161531448,
86
  "p_value": 0.0,
87
  "significant": true,
88
+ "ci_lower": 0.044152744114398956,
89
+ "ci_upper": 0.12291169166564941
90
  },
91
  "standalone_ndcg@10": {
92
+ "delta": 0.04302219259072037,
93
+ "p_value": 0.0066,
94
  "significant": true,
95
+ "ci_lower": 0.009433026997052838,
96
+ "ci_upper": 0.07666100482019714
97
  }
98
  },
99
  "_bm25": {