Open azhe825 opened 7 years ago
In UPDATE, we import previous knowledge to boost current review. Two options for the type of knowledge being imported:
UPDATE_pos saves more on storage and communications.
Presumptive non-relevant examples made it happen to train a learner with only "positive" ("relevant") examples in imbalanced data sets.
Hall data set:
Wahono data set:
Danijel data set:
UPDATE_pos performs quite similarly with UPDATE_all. It is reasonable to use UPDATE_pos and save a lot in storage and communications.
UPDATE_pos!
What
In UPDATE, we import previous knowledge to boost current review. Two options for the type of knowledge being imported:
Why
UPDATE_pos saves more on storage and communications.
How
Presumptive non-relevant examples made it happen to train a learner with only "positive" ("relevant") examples in imbalanced data sets.
Result
Hall data set:![](https://github.com/ai-se/ML-assisted-SLR/blob/master/time_decay/figure/UPDATE_Hall2.png?raw=yes)
Wahono data set:![](https://github.com/ai-se/ML-assisted-SLR/blob/master/time_decay/figure/UPDATE_Wahono2.png?raw=yes)
Danijel data set:![](https://github.com/ai-se/ML-assisted-SLR/blob/master/time_decay/figure/UPDATE_Danijel2.png?raw=yes)
Conclusion
UPDATE_pos performs quite similarly with UPDATE_all. It is reasonable to use UPDATE_pos and save a lot in storage and communications.