Multi-Agent System (MAS) appears to be an efficient, low cost, flexible, and reliable form of system, these features turns the MAS as a perfect solution for resolving complicated jobs. Personalisation is defined as the process of addressing the learner-specific techniques and their intentions or ideologies for assisting and promoting the process of an individual’s learning success. The process of both modelling and estimating the above mentioned tasks in the internet is now turning out to be a tedious task due to the continuous growth in their sizes. Here, a decentralized technique based on a multi agent optimized clustering process has been found to work well for large data sets. Genetic Algorithms (GAs) are observed as the stochastic global optimization techniques that are meant for solving the optimization problems. The Firefly algorithm (FA) is the most efficient algorithms adopted for performing the global optimization tasks in complicated search spaces. Another type of population-oriented algorithm is the Differential Evolution (DE) algorithm. In this research article a novel combination of DE and the Firefly global optimization algorithms considered as the Hybrid Firefly Algorithm Differential Evolution (HFADE) for performing the clustering tasks in an efficient manner. The effectiveness of the HFADE was experimented with benchmark functions, the achieved results shows the Hybrid algorithm well suitable for the Learning Optimisation Problems.

多代理系统（MAS）似乎是一种高效，低成本，灵活和可靠的系统形式，这些功能使MAS成为解决复杂工作的理想解决方案。个性化被定义为解决学习者特定技术及其帮助或促进个人学习成功过程的意图或意识形态的过程。由于其大小的持续增长，现在在Internet中对上述任务进行建模和估计的过程变得非常繁琐。在这里，已经发现基于多代理优化聚类过程的分散技术可以很好地用于大型数据集。遗传算法（GA）被认为是用于解决优化问题的随机全局优化技术。Firefly算法（FA）是在复杂的搜索空间中执行全局优化任务的最有效算法。面向人群的另一种算法是差分进化（DE）算法。在这篇研究文章中，DE和Firefly全局优化算法的新颖组合被视为可有效执行聚​​类任务的混合Firefly算法差分进化（HFADE）。通过基准函数对HFADE的有效性进行了实验，所得结果表明Hybrid算法非常适合于学习优化问题。在这篇研究文章中，DE和Firefly全局优化算法的新颖组合被视为可有效执行聚​​类任务的混合Firefly算法差分进化（HFADE）。通过基准函数对HFADE的有效性进行了实验，所得结果表明Hybrid算法非常适合于学习优化问题。在这篇研究文章中，DE和Firefly全局优化算法的新颖组合被视为可有效执行聚​​类任务的混合Firefly算法差分进化（HFADE）。通过基准函数对HFADE的有效性进行了实验，所得结果表明Hybrid算法非常适合于学习优化问题。