Drug Recommendations


In this section, you get drug recommendations based on the list of drugs that were prescribed for the patients clinical status. You can try either the Pearson Correlation or the Cosine Similarity measure to compute similarities between clinical cases.


Pearson Correlation measures the statistical relationship, or association, between two continuous variables.

Cosine similarity measures how similar two vectors (patients profile) are irrespective of their size.

Step 1/3

Which clinical study are you interested in providing treatment recommendations?



Common diseases related with this clinical study:



Step 2/3

A meta path is a sequence of different node and edge types, which capture a specific relation among graph entities.

Which of the following meta path would you like to select to get recommended?



Step 3/3

Select either the Cosine Similarity or the Pearson Correlation

Cosine Similarity
Pearson Correlation


Select the minimum number of co-interactions that are needed to measure the similarity score




Select the number of recommended drugs