Table 4 Comparison results of the overall performance (percentage) of different methods on ICESW14, ICEWS05-15, ICEWS18 and GDELT.

From: A contrastive learning framework with dual gates and noise awareness for temporal knowledge graph reasoning

Model

ICEWS14

ICEWS05-15

ICEWS18

GDELT

MRR

H@1

H@3

H@10

MRR

H@1

H@3

H@10

MRR

H@1

H@3

H@10

MRR

H@1

H@3

H@10

DistMult(2014)

15.44

10.91

17.24

23.92

17.95

13.12

20.71

29.32

11.51

7.03

12.87

20.86

8.68

5.58

9.96

17.13

ComplEx(2016)

32.54

23.43

36.13

50.73

32.63

24.01

37.50

52.81

22.94

15.19

27.05

42.11

16.96

11.25

19.52

32.35

ConvE(2018)

35.09

25.23

39.38

54.68

33.81

24.78

39.00

54.95

24.51

16.23

29.25

44.51

16.55

11.02

18.88

31.60

Conv-TransE(2019)

33.80

25.40

38.54

53.99

33.03

24.15

38.07

54.32

22.11

13.94

26.44

42.28

16.20

10.85

18.38

30.86

RotatE(2019)

21.31

10.26

24.35

44.75

24.71

13.22

29.04

48.16

12.78

4.01

14.89

31.91

13.45

6.95

14.09

25.99

TTransE(2016)

13.72

2.98

17.70

35.74

15.57

4.80

19.24

38.29

8.31

1.92

8.56

21.89

5.50

0.47

4.94

15.25

TA-DistMult(2018)

25.80

16.94

29.74

42.99

24.31

14.58

27.92

44.21

16.75

8.61

18.41

33.59

12.00

5.76

12.94

23.54

DE-SimlE(2020)

33.36

24.85

37.15

49.82

35.02

25.91

38.99

52.75

19.30

11.53

21.86

34.80

19.70

12.22

21.39

33.70

TNTComplEx(2020)

34.05

25.08

38.50

50.92

27.54

9.52

30.80

42.86

21.23

13.28

24.02

36.91

19.53

12.41

20.75

33.42

RE-NET(2020)

36.93

26.83

39.51

54.78

43.32

33.43

47.77

63.06

28.81

19.05

32.44

47.51

19.62

12.42

21.00

34.01

xERTE(2020)

40.02

32.06

44.63

56.17

46.62

37.84

52.31

63.92

29.98

22.05

33.46

44.83

18.09

12.30

20.06

30.34

TANGO(2021)

–

–

–

–

42.86

32.72

47.14

62.34

28.97

19.51

32.61

47.51

19.66

12.50

20.93

33.55

RE-GCN(2021)

40.39

30.66

44.96

59.21

48.03

37.33

53.85

68.27

30.58

21.01

34.34

48.75

19.64

12.42

20.90

33.69

TiRGN(2022)

44.04

33.83

48.95

63.84

50.04

39.25

56.13

70.71

33.66

23.19

37.99

54.22

21.67

13.63

23.27

37.60

HisMatch(2022)

46.42

35.91

51.63

66.84

52.85

42.01

59.05

73.28

33.99

23.91

37.90

53.94

22.01

14.45

23.80

36.61

RETIA(2023)

42.76

32.28

47.77

62.75

47.26

36.64

52.90

67.76

32.43

22.23

36.48

52.94

20.12

12.76

21.45

34.49

CENET(2023)

39.02

29.62

43.23

57.49

41.95

32.17

46.93

60.43

27.85

18.15

31.63

46.98

20.23

12.69

21.70

34.92

\(\text {L}^2\text {TKG}^{(2023)}\)

47.40

35.36

–

71.05

57.43

41.86

–

80.69

33.36

22.15

–

55.04

20.53

12.89

–

35.83

LogCL(2024)

48.87

37.76

54.71

70.26

57.04

46.07

63.72

77.87

35.67

24.53

40.32

57.74

23.75

14.64

25.60

42.33

BH-TDEN(2024)

39.3

29.7

44.0

58.1

43.0

32.7

47.9

62.7

28.9

19.1

32.6

48.7

19.4

12.5

21.4

33.6

DNCL(ours)

51.18

40.37

56.86

71.89

59.02

48.06

65.50

80.88

37.43

25.83

42.70

60.01

23.59

14.78

25.40

42.39

+ Improve

4.73

6.91

3.93

1.82

2.77

4.31

2.79

0.24

4.93

5.30

5.83

3.93

–

0.96

–

0.14

  1. The best result is highlighted in black font and the second best result is underlined. The result of TANGO is from41, the results of \(\text {L}^2\text {TKG}\), BH-TDEN and HisMatch are from the original paper, and the rest of the results are from9.
  2. The model results with the best performance under each evaluation indicator.