Lambda value analysis on Weighted Minkowski distance model in CBR of Schizophrenia type diagnosis

Autor: Akmar Efendi, Ause Labellapansa, Evizal Abdul Kadir, Ana Yulianti
Rok vydání: 2016
Předmět:
Zdroj: 2016 4th International Conference on Information and Communication Technology (ICoICT).
DOI: 10.1109/icoict.2016.7571898
Popis: According to The Global Burden of Disease conducted by Murray in collaboration with WHO and the World Bank predicts that mental illnesses will occupy the second position after cardiovascular disease in 2020. One of the mental illnesses is Schizophrenia Psychosis disorder. This research uses Artificial Intellegence case-based reasoning (CBR) method for diagnosing types of Schizophrenia disorders. Each Schizophrenia new case will be calculated using Weighted Minkowski similarity method. This study focuses on determining the most efficient Lambda (r) between the values of 1, 2 and 3 on Minkowski distance model. Data obtained from the medical records of 95 cases of which 80 cases used as training data and 15 cases are used as test data. Based on the results of tests performed, lambda value that has the best accuracy rate is at 3. In order to diagnose the type of schizophrenia, Minkowski Distance Model with a lambda value of 3 can be used.
Databáze: OpenAIRE