Table 4 Evaluation results over the SemEval-2007 dataset

Bi-LSTM + Attention (Our Method)

52.60

0.7140

0.5506

0.1700

0.0915

Deep learning baselines

 sent2affect [48]

37.20

0.3339

0.1075

0.2241

0.1428

 SS-BED [44]

50.40

0.6139

0.5098

0.1771

0.1090

 Kim’s CNN [64]

47.20

0.5437

0.4451

0.1987

0.1200

 Bi-LSTM [48]

49.89

0.6007

0.5059

0.1812

0.1074

 LSTM [9]

49.20

0.6015

0.5248

0.1842

0.1089

 GRU

46.00

0.5673

0.5003

0.2005

0.1098

Lexicon based baselines

 SWAT [11]

46.00

0.4945

0.3981

0.2453

0.1354

 Emotion Term Model [12]

49.40

0.5642

0.0167

0.3031

0.1975

 Synesketch [33]

35.86

0.3705

0.3570

0.2470

0.1510

Problem transformation baselines

 WMD [39]

40.50

0.1447

0.0459

0.2430

0.1143

 TF-IDF [39, 48]

45.60

0.4954

0.4039

0.2080

0.1135

 N-Grams [32, 44] (\(N=1\))

45.00

0.4992

0.3931

0.2089

0.1189

 TEC [32]

45.20

0.5451

0.4219

0.2028

0.1219

 TEI [32]

45.60

0.5900

0.4635

0.2985

0.1228

 MEI [32]

45.60

0.4884

0.4071

0.2051

0.1257

 GEC [32] (\(\delta =0.25\))

40.80

0.4643

0.3398

0.2113

0.1251

 GEI [32] (\(\delta =0.25\))

44.00

0.4416

0.3207

0.2136

0.1291

 Sentiment word count [32, 65]

39.04

0.5604

0.3820

0.2089

0.1208

 SSWE_u [63] (\(d=50\))

34.56

0.3130

0.1152

0.2300

0.1272

 GloVe [44] (\(d=100\))

33.12

0.2605

0.1088

0.2378

0.1152

Algorithm adaptation baselines

 TF-IDF [39, 48]

46.40

0.4799

0.3941

0.2059

0.1206

 N-Grams [32, 44] (\(N=1\))

46.80

0.5135

0.4140

0.2027

0.1171

 TEC [32]

46.40

0.5639

0.4270

0.2021

0.1204

 TEI [32]

49.60

0.6034

0.4993

0.2005

0.1122

 MEI [32]

46.40

0.4949

0.4103

0.2062

0.1306

 GEC [32] (\(\delta = 0.25\))

46.00

0.4861

0.3622

0.2089

0.1229

 GEI [32] (\(\delta = 0.25\))

46.70

0.4722

0.3531

0.2099

0.1248

 Sentiment word count [32, 65]

40.00

0.5732

0.3798

0.2023

0.1193

 \(\hbox {SSWE}_{\mathrm{u}}\) [63] (\(d=50\))

40.80

0.2071

0.0595

0.4032

0.1641

 GloVe [44] (\(d=100\))

42.40

0.2261

0.0777

0.4022

0.1643