In order to investigate certain issues I wanted to find out the journal categories directly from database.
Below is the select I came up with. This is for MySQL. The string operations on the second line try to extract the actual content that is inside the extra markup. It is not perfect, but you get an idea about the title (which was the goal).
SELECT replace(substr(ja.title, LENGTH(ja.title) - LOCATE('"',REVERSE(ja.title)) + 3),"</Title></root>","") as Title ,ac.name as Category, ac.categoryId, ac.vocabularyId FROM journalarticle ja, journalarticleresource jar, assetentry ae, classname_ cn, assetentries_assetcategories anac, assetcategory ac WHERE jar.resourcePrimKey = ja.resourcePrimKey AND jar.groupId = ja.groupId AND jar.uuid_ = ae.classUuid AND cn.classNameId = ae.classNameId AND anac.entryId = ae.entryId AND ac.categoryId = anac.categoryId AND ac.groupId = jar.groupId AND ja.title like '%Your title here%' ORDER by ja.id_ ;