Thanks for the sample data.
Are you planning to add a different source, or will your corpus consist of this single journal?
To be honest, I am not able to give you a lot of feedback, because your source makes public only poor metadata and text with no linking to facsimiles.
variable ocr quality
The OCR quality varies a lot. In your metadata, you have
<ocr_quality_mean>0.7797</ocr_quality_mean>
<ocr_quality_sd>0.2681</ocr_quality_sd>
per each article. Can you assign labels from this taxonomy? https://github.com/clarin-eric/PressMint/blob/data/Samples/PressMint-SI/PressMint-taxonomy-OCR.xml
repeated title
one title is taken from metadata and one from text
<div type="article" xml:id="PressMint-GB_1876-09-02_WidnesExaminer_art0017" n="art0017">
<head>CO .V[MEBOIAL IVIARKETS.</head>
<p>CO .V[MEBOIAL IVIARKETS.</p>
Thanks for the sample data.
Are you planning to add a different source, or will your corpus consist of this single journal?
To be honest, I am not able to give you a lot of feedback, because your source makes public only poor metadata and text with no linking to facsimiles.
variable ocr quality
The OCR quality varies a lot. In your metadata, you have
per each article. Can you assign labels from this taxonomy? https://github.com/clarin-eric/PressMint/blob/data/Samples/PressMint-SI/PressMint-taxonomy-OCR.xml
repeated title
one title is taken from metadata and one from text