Table 2 Learning of LML models at different levels of error tolerance and its effect on accuracy of the model
From: Lifelong aspect extraction from big data: knowledge engineering
\(t_{mustlink}\) | \(t_{cannotlink}\) | Total rules used | Topic coherence |
|---|---|---|---|
−1 | 1 | 3 | −855 |
−1 | 0.9 | 5 | −851 |
−0.9 | 0.9 | 6 | −842 |
−0.9 | 0.8 | 9 | −839* |
−0.8 | 0.8 | 11 | −843 |
−0.8 | 0.7 | 14 | −852 |