Table 1.
Source shares | Target has | Target shares | Best methods |
---|---|---|---|
Labeled text | Labeled text | – | |
Raw text | – | ||
Labeled features | Labeled text | – | Feature augmentation6 |
Raw text | – | Feature correspondence learning13–15 | |
Trained Models | Labeled text | – | |
Models | – | ||
Raw text | – | Online self-training21 | |
Models | Pseudo in-domain data selection26 |
Notes: It is assumed that there is always labeled data available in the source domain. “Source shares” describes what the source site is able to share with the target site. “Target has” describes what data are available at the target site. “Target shares” describes what the target site is able to share with the source site. “Methods” gives names for the types of methods in each configuration, and citations to examples of such work