From e2a3c1d0989c6d63df3184bf7dbc9d9a4cdda4ed Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Mika=20H=C3=A4m=C3=A4l=C3=A4inen?= Date: Sat, 12 Sep 2020 00:23:13 +0300 Subject: [PATCH] Update README.md --- README.md | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index d0dc2f2..6d27592 100644 --- a/README.md +++ b/README.md @@ -100,6 +100,14 @@ By default, caching is enabled. If you want to use the method with multiple diff natas.is_correctly_spelled("cat") #The result will be served from the cache natas.is_correctly_spelled("cat", cache=False) #The word will be looked up again +# Business solutions + + +Rootroo logo + +Non-standard historical or OCRed data can be a mess to deal with when you want to squeeze all the juice out of your corpora. Let us help! [Rootroo offers consulting related to non-standard data](https://rootroo.com/). We have a strong academic background in the state-of-the-art AI solutions for every NLP need. Just contact us, we won't bite. + + # Cite If you use the library, please cite one of the following publications depending on whether you used it for normalization or OCR correction. @@ -143,4 +151,4 @@ Mika Hämäläinen, and Simon Hengchen. 2019. [From the Paft to the Fiiture: a F doi = "10.26615/978-954-452-056-4_051", pages = "431--436", abstract = "A great deal of historical corpora suffer from errors introduced by the OCR (optical character recognition) methods used in the digitization process. Correcting these errors manually is a time-consuming process and a great part of the automatic approaches have been relying on rules or supervised machine learning. We present a fully automatic unsupervised way of extracting parallel data for training a character-based sequence-to-sequence NMT (neural machine translation) model to conduct OCR error correction.", - } \ No newline at end of file + }