CODEWITHSUNDEEP
sortingprolog
SteakStyles
SteakStyles

Reputation: 87

SWI Prolog homework assistance

I'm kind of at a loss on how to tackle this homework project based on what I know about prolog (which, admittedly isn't a whole lot)

Basically, we have a group of diners that rank their preferences on a certain dish. to make things easy, we name the dishes A, B, C, D, and E

The goal is to assign seats at a table based on what the diners selected as their preference, so if diner1 had their rankings as [A, E, D, C, B], he would be seated somewhere near Diner2 who has [E, C, A, B, D] because of the similar rankings.

each dish will have a point value, and will be multiplied by a modifier based on the dishes position in the list.

I know I'm going to need a list of Diners (which are going to be lists themselves), and I'm going to need to sort that list based on a score that will be calculated based on the dish rankings. The sorted list will be output in decreasing order (highest score at the top)

My questions are, how would I calculate the score based on prologs math functions, and how would I associate a Diner with a score, and then sort the list of Diners based on the score? Furthermore, is it possible to do this all in one query?

thank you in advance

Upvotes: 0

Views: 114

Answers (1)

Daniel Lyons
Daniel Lyons

Reputation: 22803

Well, I would first create a small database with the scores for each dish:

dishScore(a, 5).
dishScore(b, 7.3).
% ...

Then I'd set up the location coefficients:

coefficients([1, 0.8, 0.7, 0.4, 0.1]).

Now you're prepared to write a predicate to score a ranking:

scoreRanking(Ranking, Score) :-
  coefficients(Coeff),
  scoreRankingLoop(Ranking, Coeff, Score).

scoreRankingLoop([], [], 0).
scoreRankingLoop([Dish|Dishes], [C|Coeff], FinalScore) :-
  dishScore(Dish, DishScore),
  scoreRankingLoop(Dishes, Coeff, RemainingScore),
  MyScore is DishScore * C,
  FinalScore is RemainingScore + MyScore.

Now you need to sort the list of diners. The ISO predicate for doing this kind of thing is keysort/2, and the idea is that you build X-Y terms where X is the key and Y is the value, and it sorts on X. SWI has its own special predicate, predsort/3 which we will use instead because it's less work.

compareOnScore(Delta, Left, Right) :-
  scoreRanking(Left, LScore),
  scoreRanking(Right, RScore),
  compare(Delta, LScore, RScore).

Now you can just call predsort/3 with compareOnScore as an argument:

predsort(compareOnScore, Diners, Result).

Of course, this is making a lot of bad assumptions. I hope it's helpful anyway!

Upvotes: 1

Related Questions