The social networks are widely used by millions of people worldwide. The trust concept is one of the most important issues in Social Network Analysis (SNA) which highly affects the quantity and quality of the inter-connections, decisions, and interactions among the users in e-commerce or recommendation systems. Many normative algorithms are developed to calculate the trust which most of them are complicated, depend on the network structure, and need lots of critical information that makes them hard to use. The aim of this paper is proposing a descriptive, simple and effective method for calculating the maximal trust and the trust route between any two users of an Online Social Network (OSN). For this purpose, four new models for estimating the trust mechanism of the users are proposed and analyzed using Kolmogorov–Smirnov and Anderson–Darling statistical hypothesis tests to identify and validate the best-fitted model based on 20,613 empirical results gathered from 4552 social network volunteers. Due to the time–complexity of the problem, a meta-heuristic algorithm based on the Artificial Bee Colony (ABC) optimization method is also developed for solving the best-fitted model. The proposed algorithm is simulated in Matlab® over six larger test cases adopted from the Facebook dataset. In order to evaluate the performance of the developed algorithm, the Ant Colony Optimization (ACO) and Genetic Algorithm (GA) based meta-heuristics are also simulated on the same test cases. The comparison of the computational results shows that the ABC approach performs better than the ACO and GA as the size of the network increases.

中文翻译：

---

在线社交网络最大信任度计算和人工蜂群算法求解的新模型

社交网络被全世界数百万人广泛使用。信任概念是社交网络分析（SNA）中最重要的问题之一，它在很大程度上影响着电子商务或推荐系统中用户之间的相互联系，决策和交互的数量和质量。开发了许多规范算法来计算信任度，其中大多数算法很复杂，取决于网络结构，并且需要大量关键信息，因此很难使用。本文的目的是提出一种描述性，简单有效的方法，用于计算在线社交网络（OSN）的任何两个用户之间的最大信任度和信任路径。以此目的，提出了四个新的估计用户信任机制的模型，并使用Kolmogorov–Smirnov和Anderson–Darling统计假设检验进行了分析，以基于从4552个社交网络志愿者收集的20,613个经验结果来识别和验证最合适的模型。由于问题的时间复杂性，还开发了一种基于人工蜂群（ABC）优化方法的元启发式算法来求解最佳拟合模型。该算法在Matlab®中从Facebook数据集中采用的六个更大的测试案例中进行了仿真。为了评估所开发算法的性能，还在相同的测试案例上模拟了基于蚁群优化（ACO）和遗传算法（GA）的元启发式算法。