Retrieval is one of the stages in case-based reasoning system which find a solution to new problem or case by measuring the similarity between the new case and old cases in the case base. Some of the similarity measurement techniques are involving feature weights that show the importance of the feature in a case. Feature weights can be obtained from a domain expert or by using a feature weighting method either locally or globally. Gradient descent is the feature weighting method which computes global weights for each feature. This research implemented gradient descent to obtain feature weights in case-based reasoning for hepatitis diagnosis and the similarity measurement using weighted Euclidean distance. There were four variations of case base size and test data used in the research; i.e., 50% of case base and 50% of test data, 60% of case base and 40% of test data, 70% of case base and 30% of test data, and 80% of case base and 20% of test data were variation 1, 2, 3, and 4 respectively. In addition, each variation used four scenarios based on the way how to mark the test data. In scenario 1, 2, 3 and 4, the test data were respectively marked at the end, the beginning, the half of beginning and half of end, and the middle. The result showed that the accuracy of the system reaches 100% at scenario 1 in variation 4. Overall of all four variations and four kinds of scenario, the average accuracy of the system was 77.55%, average recall of system was 69.74%, and the average of precision was 78.39%. In addition, the level of accuracy was also influenced by the number of case base and the scenario of case selection for the case base. This is because more cases in the case base, the bigger chance for the system to find similar cases.