CAJ | 학술논문

　　使用调和均值的 KHM 聚类算法，不像 KM 聚类算法，具有对初始值不敏感的优点。但它作为一个基于中心聚类算法，难以摆脱早熟收敛的问题。为了克服 KHM 算法的不足，本文提出结合 ABC 和 KHM 的 ABC-KHM 混合聚类算法。在混合算法中，聚类行为可以分为两个阶段：全局搜索的 ABC 聚类阶段和局部求精的 KHM 聚类阶段。通过仿真实验，并与 KHM 聚类算法进行了比较，结果表明：ABC-KHM 混合聚类算法，不仅对聚类初始值不敏感，而且具有较快的聚类速度、良好的全局聚类效果，是一个不错的聚类算法。
　　사용조화균치적 KHM 취류산법，불상 KM 취류산법，구유대초시치불민감적우점。단타작위일개기우중심취류산법，난이파탈조숙수렴적문제。위료극복 KHM 산법적불족，본문제출결합 ABC 화 KHM 적 ABC-KHM 혼합취류산법。재혼합산법중，취류행위가이분위량개계단：전국수색적 ABC 취류계단화국부구정적 KHM 취류계단。통과방진실험，병여 KHM 취류산법진행료비교，결과표명：ABC-KHM 혼합취류산법，불부대취류초시치불민감，이차구유교쾌적취류속도、량호적전국취류효과，시일개불착적취류산법。
The K-Harmonic Means (KHM) is a center-based clustering algorithm which uses the harmonic averages of the distances from each data point to the centers as components to its performance function. Unlike K-means, KHM is less sensitive to initial conditions. However, KHM as a center-based clustering algorithm can only generate a local optimal solution. The paper presents a hybrid clustering algorithm combining Artificial Bee Colony and K-Harmonic Means (ABC-KHM) for solving this problem. This hybrid clustering algorithm has been implemented and tested on several simulated and real datasets. The performance of this algorithm is compared with KHM. Our computational simulations reveal the ABC-KHM clustering algorithm has the advantage of global searching, fast convergence and less sensitive to initial conditions. The ABC-KHM is a robust clustering algorithm.