[x] visualize formalism on clustering example (#21)
[x] visualize formalism on topic model example
Questions and Answers
[x] I still got the problem with ternary relationships
=> solved by having association entity types
[x] and the one of n coding scheme
=> this is a special case where the weak entity of the domain of the summarization constraint is no longer weak
[x] I need to look at the translation of covered plates in the book (1:1?)
=> 1:n (i.e. 1:1 : 1:* and weak)
[x] Do I need to observe a variable with constraints in order to translate them? When am I allowed to remove variables and convert them to relationships with constraints (cardinalities)? => observation not necessary. It is only necessary to fill the database (and thus the foreign key e.g.).
[x] What about the cardinalities? Why do they have to be (1,). They can be more specific depending on the cardinality of the index sets of the corresponding plates. Thus instead of (1,) : (1,) it could be (n,n) : (m:m)? => it's ok to leave it as (1,)
[x] Are entity types w/o keys possible? => usually this means that I can have arbitrary entries in the table. If I define that there is only one global entity it is fine.
[x] Should association entity types have names? => Yes (A_B between A and B)
[x] I need to remove the cardinalities on the weak relationships (1,1)
[x] give association entities names
[x] is it called association entity or association entity type? => types besser
[x] what are the conditions for merging relationships in the reduction step? => express the same thing (semangtically)
[x] there is a problem with covered plates in (pm2erm_covered.dia:covered_3) => created separate issue for that
ToDo
Questions and Answers
Depends on: #12