Classification Accuracy and Speed Search Using Career Site Page Focus-Based Web Crawler and Harmony Search Algorithm

Career site pages contain valuable information for job seekers, but effectively classifying these pages into relevant categories remains a challenge. This study aims to develop a classification system for career website pages using the Harmony Search algorithm, enhanced by a Focused Web Crawler for data collection. The Focused Web Crawler extracts relevant job tags and features from career site pages, which are then used as training data for the classification algorithm. Experiments were conducted with various parameters, including job tags, the number of features (50), and Harmony Search algorithm settings. The results show that with 500 web pages and optimized algorithm parameters, the system achieved a classification accuracy of up to 95%. The system outperforms alternative parameter configurations, demonstrating that the combination of 50 features and specific Harmony Search parameters, such as HMCR and PAR set at 0.5, leads to the best performance. This system can assist users in efficiently finding job information tailored to their needs and skills. Moreover, it serves as a reference for future classification systems using similar techniques and algorithms.