Similarity | Rank | |
---|---|---|
(_n_=_[number] | 0.740 | 87204 |
,_n_=_[number]_) | 0.716 | 94537 |
(_n_=_[number]_) | 0.715 | 17896 |
=_[number]_) | 0.697 | 38861 |
[number]_±_[number] | 0.671 | 35628 |
=_0 | 0.666 | 22510 |
(_N_=_[number]_) | 0.663 | 65149 |
be_[number]_±_[number] | 0.646 | 88408 |
−_1 | 0.642 | 87297 |
>_0 | 0.642 | 88458 |
(_0_._[number] | 0.628 | 41794 |
=_2 | 0.626 | 50110 |
P_[number]_) | 0.624 | 51048 |
[number]_(_0_. | 0.611 | 62641 |
(_p | 0.610 | 54896 |
=_4 | 0.606 | 91534 |
mean_value | 0.606 | 48998 |
[number]_+_/_- | 0.601 | 49829 |
=_3 | 0.597 | 71012 |
(_p_=_0_._[number]_) | 0.594 | 70282 |
(_P_=_0_. | 0.593 | 86219 |
(_P_=_[number]_) | 0.592 | 44450 |
control_group | 0.591 | 13992 |
(_P_[number]_) | 0.589 | 29952 |
be_randomize | 0.588 | 85991 |