Kmeans - Chimpanzee Leader Election Optimization Algorithms-Based Data Analysis in Clustering Model
Akhmad Dahlan, Ferry Wahyu Wibowo · 2023
The data analysis process includes data grouping techniques based on data transformation, data cleaning, model building, and similar characteristics to understand important information from the data. The clustering method is an unsu-pervised learning model in machine learning. The principle of the clustering method is to divide, or group data points with similar characteristics into several small groups. The clustering method may provide optimal results in some instances so that not all cases can get appropriate results. It is what underlies the research in this paper. This paper analyzes data similarity according to the characteristics found and groups similar data objects into one cluster. The method used in this research uses the Chimpanzee Leader Election Optimization (CLEO) algorithm as an optimization algorithm in determining clustering and also a hybrid method with the K-means algorithm. The data analyzed to determine the performance of the model is iris data. Even though the data used is classification data, data object analysis is used to look for similarities between existing data. The elbow method functions to find the number of clusters that will be the input K to K-means. The model comparison in this paper uses three methods: data clustering based on the K-means method, the CLEO algorithm, and the combined Kmeans-CLEO. These results are evaluated and analyzed based on the input data and the method's performance. The results of the research that has been carried out show good modeling using the CLEO algorithm, while the hybrid algorithm is still below accuracy.