Update README.md
Browse files
README.md
CHANGED
@@ -11,6 +11,8 @@ metrics:
|
|
11 |
widget:
|
12 |
- text: "1820 [DATE] We received a letter from [MASK] Majesty."
|
13 |
- text: "1850 [DATE] We received a letter from [MASK] Majesty."
|
|
|
|
|
14 |
|
15 |
---
|
16 |
<img src="https://upload.wikimedia.org/wikipedia/commons/5/5b/NCI_peas_in_pod.jpg" alt="erwt" width="200" >
|
@@ -40,9 +42,9 @@ For example, we would format a snippet of text taken from the [Londonderry Senti
|
|
40 |
|
41 |
|
42 |
|
43 |
-
## Intended uses
|
44 |
|
45 |
-
Exposing the model to temporal metadata allows us to investigate **language change** and perform **date prediction**.
|
46 |
|
47 |
### Language Change 👑
|
48 |
|
@@ -50,5 +52,11 @@ Also in the nineteenth-century Britain had Queen on the throne for a very long t
|
|
50 |
|
51 |
### Date Prediction
|
52 |
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
## Data Description
|
54 |
|
|
|
11 |
widget:
|
12 |
- text: "1820 [DATE] We received a letter from [MASK] Majesty."
|
13 |
- text: "1850 [DATE] We received a letter from [MASK] Majesty."
|
14 |
+
- text: "[MASK] [DATE] The Franco-Prussian war is a matter of great concern."
|
15 |
+
- text: "[MASK] [DATE] The Second Schleswig war is a matter of great concern."
|
16 |
|
17 |
---
|
18 |
<img src="https://upload.wikimedia.org/wikipedia/commons/5/5b/NCI_peas_in_pod.jpg" alt="erwt" width="200" >
|
|
|
42 |
|
43 |
|
44 |
|
45 |
+
## Intended uses
|
46 |
|
47 |
+
Exposing the model to temporal metadata allows us to investigate **historical language change** and perform **date prediction**.
|
48 |
|
49 |
### Language Change 👑
|
50 |
|
|
|
52 |
|
53 |
### Date Prediction
|
54 |
|
55 |
+
## Limitations
|
56 |
+
|
57 |
+
ERWT models were trained for evaluation purposes, and cary critical limitations. First of all, as explained in more detail below, this model is trained on a rather small subsample of British newspapers, with a strong Metropolitan and liberal bias.
|
58 |
+
|
59 |
+
Secondly, we only trained for one epoch, which suggests. For the evaluation purposes we were interested in the relative performance of our models.
|
60 |
+
|
61 |
## Data Description
|
62 |
|