Dedicated "Learn to rank" algorithms are not currently available in the visual machine learning interface. We will add this as a feature request.
You can easily view this problem by a regression (pointwise approach) or classification (pairwise approach). In this case, you can use Dataiku's visual ML interface to train models on the rank. You will need some custom code to evaluate these models according to rank-specific metrics (NDCG for instance).
If you want to apply dedicated "learn to rank" algorithms, you would use the coding capabilities of Dataiku, using either Python, R, or Spark-Scala. There are many open source libraries in these languages (see https://github.com/topics/learning-to-rank).
Note that popular "learn to rank" algorithms (RankNet, LambdaRank and LambdaMART) do transform the ranking problem into a classification or regression problem. But they have custom cost functions compared to "classic" ML. See https://www.microsoft.com/en-us/research/publication/from-ranknet-to-lambdarank-to-lambdamart-an-overview/?from=http%3A%2F%2Fresearch.microsoft.com%2Fpubs%2F132652%2Fmsr-tr-2010-82.pdf for more details.