Skip to main content
Have a personal or library account? Click to login
Using Text Mining to Search for Neolithic Vlaardingen Culture Sites in the Rhine-Meuse-Scheldt Delta Cover

Using Text Mining to Search for Neolithic Vlaardingen Culture Sites in the Rhine-Meuse-Scheldt Delta

Open Access
|Mar 2025

Figures & Tables

Figure 1

Distribution of the Vlaardingen Culture and Stein group (after: Verhart 2010; base map: © EuroGeographics 2024, map made in QGIS).

Table 1

List of queries entered in AGNES with number of hits.

START DATE (BCE)END DATE (BCE)FREE TEXT QUERYENGLISH TRANSLATIONNUMBER OF HITS
“vlaardingen cultuur”vlaardingen culture834
38002000“vlaardingen*”vlaardingen2483
“vlaardingen stein wartburg”vlaardingen-stein-wartburg11
“vlaardingen stein wartberg”vlaardingen-stein-wartburg4
“vlaardingen groep”vlaardingen group265
“vlaardingen stein”vlaardingen-sStein/vlaardingen stein98
“vlaardingencultuur”vlaardingen culture712
“vlaardingengroep”vlaardingen group125
Table 2

Relevance of AGNES hits.

NR.RELEVANCE
1Relevant (report about a Vlaardingen Culture site) unknown
2Relevant (report about a Vlaardingen Culture site) known
3Relevant (previously unknown Vlaardingen site mentioned in the text)
4Relevant (previously unknown Vlaardingen site mentioned in the literature list)
5Semi-relevant (Stein site publication, mentioning Vlaardingen Culture in discussion)
6Semi-relevant (Vlaardingen Culture mentioned in a discussion)
7Semi-relevant (a different Vlaardingen Culture site mentioned in the text based on previous research)
8Not relevant (not a report about a Vlaardingen Culture site)
Table 3

Types of irrelevant hits.

NUMBERTYPE OF IRRELEVANT DOCUMENT
1Wrong time period
2Page listing abbreviations
3Page containing research plan (plan van aanpak)
4Unknown time period
5Page containing list of time periods
6Negation (‘no vlaardingen culture’)
7Other
8Literature list (only)
9Coring chart
10Database structure
11Vlaardingen as a location on a map
12Vlaardingen as place name in text
Table 4

Relevance of AGNES hits totals.

RELEVANCECOUNT%
Relevant (report about a Vlaardingen Culture site) unknown1653.6
Relevant (report about a Vlaardingen Culture site) known2595.7
Relevant (previously unknown Vlaardingen site mentioned in the text)90.2
Relevant (previously unknown Vlaardingen site mentioned in the literature list)60.1
Semi-relevant (a different Vlaardingen Culture site mentioned in the text based on previous research)139830.8
Semi-relevant (Stein site publication, mentioning Vlaardingen Culture in discussion)651.4
Semi-relevant (Vlaardingen Culture mentioned in a discussion)67014.8
Not relevant (not a report about a Vlaardingen Culture site)196043.2
Total453299.8
Table 5

Reasons for irrelevant hits AGNES.

IRRELEVANCE REASONCOUNTPERCENTAGE
Page listing abbreviations40.2%
Page containing research plan (plan van aanpak)20.1%
Page containing list of time periods25713.1%
Negation (‘no vlaardingen culture’)160.8%
Literature list (only)46523.7%
Database structure30215.4%
Vlaardingen as place name in text90546.2%
Vlaardingen as place name on a map80.4%
Figure 2

Network representation; two-mode network visualising the relevance of different queries. Network visualising the different queries (grey) and relevant (red), irrelevant (blue), and semi-relevant hits. Nodes are scaled according to their centrality degree (std), links are ranked by weight, visualized in stress minimization layout (graph made in Visone).

Figure 3

Network representation; two-mode network visualising irrelevance types (blue) for different queries (grey), Nodes are scaled according to their centrality degree (std), links are ranked by weight, visualized in stress minimization layout (graph made in Visone).

Table 6

Results per site, newly found sites, previously known sites and sites of which the reports are not in AGNES.

RESULT PER SITECOUNTPERCENTAGE
Found exclusively in AGNES2717.1%
Found exclusively indirectly in AGNES31.9%
Found previously and in AGNES3924.7%
Not found in AGNES queries (pdf not present in DANS or ARCHIS)7648.1%
Not found in AGNES queries (pdf is present in DANS)138.2%
Total158100%
Figure 4

Vlaardingen Culture sites plotted according to whether or not they were found in AGNES (basemap: © EuroGeographics 2024, map made in QGIS).

Figure 5

Vlaardingen Culture sites plotted according to their cultural attribution, on the right the supposed border area between the Vlaardingen and Stein group (basemap: © EuroGeographics 2024, map made in QGIS). The sites are plotted according to their cultural attribution, on the left the supposed border area between the Vlaardingen and Stein group.

DOI: https://doi.org/10.5334/jcaa.205 | Journal eISSN: 2514-8362
Language: English
Page range: 110 - 124
Submitted on: Feb 10, 2025
Accepted on: Feb 13, 2025
Published on: Mar 24, 2025
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2025 Lasse Van den Dikkenberg, Alex Brandsen, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.