ABSTRACT
Hepato Cellular Carcinoma (HCC) is one of the primary liver cancer. It typically affects persons who have chronic liver diseases like cirrhosis, which is brought on by hepatitis B and C. It is the most common type of liver cancer also it is of long-term disease. The motivation of the work is to in early stages wants to identify the diseases because the annual mortality rate is increasing. By recognizing early on, we can lower the death rate. The objective of this work is to reduce the features with identify the diseases in their early stages. In the first phase, we find out the attribute weight for each feature by using Information Gain (IG). In the second phase, we applied the hybrid of Fire Fly with Harmony Optimization Algorithm (FFHOA) to reduce the high-level feature space to low dimensional space to improve the performance. In the third phase, Multi-Layer Perceptron (MLP) classifier is used for evaluating the performance with 12 metrics. It is justified that the abovementioned FFOHA-MLP outperform well on reduced feature with good performance. Datasets are taken from the UCI machine learning repository. The dataset contains 165 instances and 50 attributes.
