Queries

From TracingNetworksWiki

Overall:

archaeologists are looking for non-random patterns in their data that can then be further interpreted

important for our analyses are constant feedback loops, meaning that the results of a query instantly prompt further questions and queries to explore; some query results might make an additional dataset we would like to add on to the dataset

- queries comparing two and more different variables (e.g. show all loomweights from Metaponto with cross decoration)

- numerical and statistical expression of the results (e.g. 12 in total, 5% of the Metaponto set, 3 % overall)

- spatial display of the results (geographical distribution, density, distance, random or non-random distribution - finding non-random distributions and then being able to add a tripple for 'non random' and analyse this further)

- across time, how do patterns change

- mapping to ontologies of concepts (e.g. chaine operatoire, networks, processes)

- what is the absolute/relative difference between objects in male/female graves

so far, we see the advantage of being able to map different related data sets together without having to change the individual data entries (e.g. one site records fabric in one way, another site records it another way - find commonalities and analyse across both datasets)

other than that, we need to come up with more examples of advantages of using ontologies over xcel, access and GIS applications to make it worthwhile

Sara:

I'd like to lend my voice to queries combining two or more variables (what in Access is called crosstab query, I believe), e.g. distribution of fabric types by chronological categories, or of shape types across different sites, with the option of expressing that in numbers (existential quantifier, if that's the phrase).

Alessandro:

I thought some queries we can use in order to text the dataset show skills and advantages of our ontological dataset. We can start with the very basic following queries:

-which is the most attested type of loom weight (conical, pyramidal etc... ) in percentage ? -what is the distribution of these different types in relation with the nature of the site (household context, tombs, place of cult etc...) ? - what is the distribution of these different types in relation with the sites, i.e. which are the different percentages of the types (discoid, etc...) in the different site (using the property -has location-)? Is it possible to show that in Google earth using the coordinates? -What is the distribution of these types in relation with the time spam? - what is the range of the weight (I mean only the exact values) in comparison with the different types (discoid, pyramidal etc...) - there is any relation between two different dimensional parameters, for instance weight versus diameter in the discoid loom weights and weight versus height in the pyramidal loom weights, using only the exact values? - what is the distribution of the type of decoration (stamp, stamp on relief, graffito, ect...) in relation with the different types (conical, discoid, etc...) -What is the distribution of a single decoration (for instance Gorgoneion_on_relief, Two_heads_facing_each_other_on_relief,and rosette impressed) in the different sites (using the property 'has location')