Discrepancies bar chart and "show resources" in BoE

[eva-andersen]

Affected corpora: AMP & SMMB

Main problems:

• Bar chart shows X number of hits for a keyword but the 'show resources with specific keyword' shows different information. The keyword hits within the resources pane are (much) lower or(much) higher. Sometimes it also displays "no resources found" although their should be resources according to the bar chart. For example: AMP year 1867 has within the bar chart 41 hits for the keyword "Mundy". The displayed resources for "Mundy" in that year is only 12. The opposite happens for the year 1866: bar chart displays 2 hits. The resources pane shows 29 hits.

• When bar chart is displayed by year, the 'show resources with specific keywords' contains sources of multiple years. For example: in AMP the year 1866 is selected in the bar chart but displays documents from 1866 and 1867.

• In the "show resources with specific keyword" the visualisation of the amount of pages is also sometimes displayed faulty. For example: "5 of 2 pages" instead of "2 of 5 pages".

Examples: (see also screenshots)

• http://10.240.4.161/explorer?step=1&from=1884-10-24&to=1887-01-01&filters=eyJrZXl3b3JkUHJlc2VuY2VGcmVxdWVuY3ktMCI6eyJrZXl3b3JkIjoiYW52ZXJzIEFORCBwaHJlbm9wYXRoaWUifX0%3D&explorables=keywordPresenceFrequency&binMethod=year&resourcesSubfilter=keywordPresenceFrequency-0

• http://10.240.4.161/explorer?step=32&filters=eyJrZXl3b3JkUHJlc2VuY2VGcmVxdWVuY3ktMCI6eyJrZXl3b3JkIjoibXVuZHkifX0%3D&explorables=keywordPresenceFrequency&resourcesSubfilter=keywordPresenceFrequency-0

• http://10.240.4.163/explorer?step=23&filters=eyJrZXl3b3JkUHJlc2VuY2VGcmVxdWVuY3ktMCI6eyJrZXl3b3JkIjoibXVuZHkifX0%3D&explorables=keywordPresenceFrequency&binMethod=year&resourcesSubfilter=keywordPresenceFrequency-0

• http://10.240.4.163/explorer?step=20&filters=eyJrZXl3b3JkUHJlc2VuY2VGcmVxdWVuY3ktMCI6eyJrZXl3b3JkIjoibXVuZHkgQU5EIGxhbmd1ZSJ9fQ%3D%3D&editPlotId=keywordPresenceFrequency-0&explorables=keywordPresenceFrequency&binMethod=year

[theorm]

Hi @eva-andersen . How critical is this issue for your work. I can see that there is a problem with a database query under certain circumstances. I need to think about a proper solution to fix it. Fixing it quickly will most likely break queries elsewhere.

[eva-andersen]

hi @theorm . It does mess up the research a bit, especially because I'm then not sure if I missed any useful sources due to the discrepancies or that I can't give the correct statistical numbers of how many times a word/person is present in the different corpora. But if a quick fix will maybe break other things, then lets leave it as it is for now. I'll be working with some other queries the upcoming days. I'll see how severe the problem is with those keywords.