<p>The economic stability of a country can be determined from the changes in the rate of inflation. Inflation is measured by the annual percentage change in the Consumer Price Index. Since there exists some uncertainties in the inflation data, fuzzy logic is one of the ways to analyse the data. Decisions in fuzzy logic can be made using the fuzzy rule-based inference system. Fuzzy rule-based inference can be obtained from expert knowledge, but the knowledge from the experts on the working of a system is not always available. Therefore, the use of association rules<em> </em>approach could solve the problem. Using three methods of fuzzy inferences; namely the Mamdani Methods, zero-order Sugeno method, and the first-order Sugeno method, this study was carried out to determine which method fits to predict the general monthly inflation data in Indonesia. The Inflation data were derived from the inflation of foodstuff price, <em>X<sub>1</sub></em>; inflation of food, drinks, cigarettes and tobacco prices, <em>X<sub>2</sub></em>; inflation of housing, water, electricity, gas, and fuel prices, <em>X<sub>3</sub></em>; inflation of clothing price, <em>X<sub>4</sub></em>; inflation of health care price, <em>X<sub>5</sub></em>; inflation of education, recreation, and sports prices, <em>X<sub>6</sub></em>; and inflation of transportation, communication, and financial services prices, <em>X<sub>7</sub></em>. The performance of the three methods was compared using mean squared error (MSE) and mean absolute percentage error (MAPE) as the accuracy measurement to establish the best fuzzy inference method that fits the inflation value. It was found that the most appropriate method which generated the most accurate results to fit the fuzzy inference system to the inflation data was the first-order Sugeno method.</p>
Genetic fuzzy system for prediction of respiratory rate of chicks subject to thermal challenges
