Logo alt

A event of the International Conference on Case-Based Reasoning (ICCBR 2015)
28th September 2015, Frankfurt, Germany

Challenges

The contest offers three challenges:

  • the cocktail challenge on suggesting cocktail recipes with a limited set of ingredients and managing the ingredient quantities
  • the sandwich challenge on suggesting sandwich recipes with no restriction on ingredients, but managing the preparation steps (or at least the order of the ingredients)
  • the open challenge on novel ideas and positions on computer adaptation cooking.

The cocktail and sandwich recipes will be prepared in reality and will be tasted by the ICCBR audience! A team may participate to many challenges.

Cocktail challenge on making real cocktails

In this challenge, the system should be able to suggest a tasty cocktail recipe that matches a user query including a set of desired ingredients from a limited set of available ingredients and avoiding unwanted ones (not necessarily from the limited set of ingredients). In addition, the system should adapt the ingredient quantities. Without information on the ingredient quantities, the original quantities will be used to prepare the cocktail in real.

Evaluation criteria: scientific quality, culinary quality

Assessment procedure: paper evaluation and comparison of the results of the systems on a same set of queries by the jury (cocktail jury prize) and public vote after tasting in real the recipes of all the system on a same query chosen by the jury (cocktail public prize).

Material provided by the organizers (see resources section):

Call of paper: see rules page.

Sandwich challenge on making real sandwiches

In this challenge, the system should be able to suggest a tasty cold sandwich recipe that matches a user query including a set of desired ingredients and avoiding unwanted ones. In addition, the system should adapt the recipe preparation, at least the order in which ingredient will be put in the sandwich. Without information on the ingredient quantities, the original input procedure will be used to prepare the sandwich in real. The recipe will be interpreted by a chef, to correct some usual missing preparation step, for example, put the tomatoes without mentioning that the tomatoes must be sliced.

Evaluation criteria: scientific quality, culinary quality

Assessment procedure: paper evaluation and comparison of the results of the systems on a same set of queries by the jury (sandwich jury prize), and public vote after tasting in real the recipes of all the system on a same query chosen by the jury (sandwich public prize).

Material provided by the organizers (see resources section):

Call of paper: see rules page.

Open challenge on adapting cooking recipes

In this challenge, you may propose whatever you want about the adaptation of cooking recipes, e.g. workflow adaptation, text adaptation, community-based adaptation, recipes combination, explanations, similarity computation, recipe personalized recommendation, etc. The evaluation will take into account the originality aspect and the scientific aspect of the work. A running system implementing the work is optional.

Evaluation criteria: scientific quality, originality, culinary quality

Assessment procedure: usual scientific review process; program commitee vote

Material provided by the organizers (see resources section):

  • No particular resources are provided for this challenge, depending of the what problem is addressed, but feel free to use resources provided for the other challenges.

Call of paper: see rules page.