ANALSYING THE PARALLEL FUSION LOGIC AND FUSION K-MEANS (FKM) WITH LANGUAGE DOCUMENT

Text summarization is a method of retrieving information, and it is very effective. Text summarization approach techniques fall into two categories: abstract and abstract. This white paper focuses on extraction techniques so that the output of the application is the key sentence taken verbatim from the original text without verbal modification. There are four main processes in this research. These are the preprocessing phase, the scoring function, the optimization of summarization by two methods (logic and fused K-Means), and the extraction of summarization results. In this study, seven features are used to calculate a score for each sentence. The core of this study is to compare the reliability of fused logic and fused k-means methods in optimizing the aggregation results process. The results showed that the Fusion Logic method outperformed the Fusion K-Means method and had the highest similarity to human summarization, with a percentage accuracy of 48.25% for Fusion Logic and 44% for Fusion K-Means. It was %.33%. Accuracy improves with the additional compression of 0.67% for Fusion Logic Method and 3.5% for Fusion K-Means.