How to evaluate the performance of your text mining data?
Precision and recall
The presentation of the proposed method was evaluated in comparison to Hu and Liu et al. The two commonly used statistical classification for performance evaluations are precision and recall. For classification purposes, terms such as true positive(TP), true negative (TN), false positive (FP) and false-negative (FN) are utilized to compare the given classification of the item with the required correct classification. The given classification of an item means the class label allotted to the item by a classifier and the required correct classification means the class the item really included. This is well explained in the table below.
Table 2: Explanation of terms for precision and recall
| Obtained result/classification | Obtained result/classification | ||
| E1 | E2 | ||
| Obtained result/classification | E1 | TP(true positive) | FP(false positive) |
| Obtained result/classification | E2 | FN(false negative) | TN(true negative) |
Source: Adopted from Wu (2010)
Precision=tp/(tp+fp)
Recall=tp/(tp+fn)
F-MEASURE
The traditional F-measure or balanced F-score is the harmonic mean of precision and recall, which combines both precision and recall.
F=2(PrecisionRecall)/(Precision+Recall)
he precision-recall and F-measure values calculated using the above formulas for various methods. Besides, other commonly used performance evaluated tool is the Receiver Operating Characteristic (ROC),
Contact us for more information about how we can help you to develop CER report.
Comment here