Skip to content

Commit 52bd874

Browse files
committed
update README file
1 parent 14ce3a9 commit 52bd874

1 file changed

Lines changed: 27 additions & 2 deletions

File tree

README.md

Lines changed: 27 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,31 @@
1-
## Named entities
1+
## Automatic generator of semi-structured documents (SSDs)
2+
This project generates semi-structured documents ( invoices, payslips, and receipts).
23

3-
#### Annotations in GEDI File :
4+
This is the Java implementation of our two papers :
5+
6+
Blanchard, J., Belaïd, Y., & Belaïd, A. (2019, September). Automatic generation of a custom corpora for invoice analysis and recognition. In 2019 International Conference on Document Analysis and Recognition Workshops (ICDARW). IEEE.
7+
8+
and
9+
10+
Belhadj, D., Belaïd, Y., & Belaïd, A. (2021, September). Automatic generation of semi-structured documents. In 2021 International Conference on Document Analysis and Recognition Workshops (ICDAR-OST21). IEEE.
11+
12+
### SSDs generation :
13+
- We can generate locally the three SSDs as well as the cloned invoices separately by launching :
14+
15+
Testlanuch
16+
17+
- We can choose to generate these documents via the API web interface by launching :
18+
mvn quarkus:dev
19+
and then by accessing :
20+
http://localhost:9080/api/ws/
21+
22+
### Diversity evaluation :
23+
- We can evaluate the diversity of the local generated SSD datasets using 4 metrics :
24+
Alignement, overlapping, SCR_score, and SELF-BLEU :
25+
TestDiversityLaunch
26+
27+
28+
#### Annotations in Invoices GEDI File :
429

530
* Invoice Number - IN
631
* Seller(Company) name - SN

0 commit comments

Comments
 (0)