Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Memetic Spider Monkey Optimization for Spam Review Detection Problem
Big Data ( IF 4.6 ) Pub Date : 2021-06-21 , DOI: 10.1089/big.2020.0188 Sayar Singh Shekhawat 1 , Harish Sharma 1 , Sandeep Kumar 2
Big Data ( IF 4.6 ) Pub Date : 2021-06-21 , DOI: 10.1089/big.2020.0188 Sayar Singh Shekhawat 1 , Harish Sharma 1 , Sandeep Kumar 2
Affiliation
Spider monkey optimization (SMO) algorithm imitates the spider monkey's fission–fusion social behavior. It is evident through literature that the SMO is a competitive swarm-based algorithm that is used to solve difficult real-life problems. The SMO's search process is a little bit biased by the random component that drives it with high explorative searching steps. A hybridized SMO with a memetic search to improve the local search ability of SMO is proposed here. The newly developed strategy is titled Memetic SMO (MeSMO). Further, the proposed MeSMO-based clustering approach is applied to solve a big data problem, namely, the spam review detection problem. A customer usually makes decisions to purchase something or make an image of someone based on online reviews. Therefore, there is a good chance that the individuals or companies may write spam reviews to upgrade or degrade the stature or value of a trader/product/company. Therefore, an efficient spam detection algorithm, MeSMO, is proposed and tested over four complex spam datasets. The reported results of MeSMO are compared with the outcomes obtained from the six state-of-art strategies. A comparative analysis of the results proved that MeSMO is a good technique to solve the spam review detection problem and improved precision by 3.68%.
中文翻译:
垃圾评论检测问题的模因蜘蛛猴优化
蜘蛛猴优化(SMO)算法模仿蜘蛛猴的裂变-融合社会行为。通过文献可以明显看出,SMO 是一种基于群体的竞争算法,用于解决现实生活中的困难问题。SMO 的搜索过程因随机成分而有点偏差,随机成分以高探索性搜索步骤驱动它。此处提出了一种具有模因搜索的混合 SMO,以提高 SMO 的局部搜索能力。新开发的策略名为模因 SMO (MeSMO)。此外,所提出的基于 MeSMO 的聚类方法应用于解决大数据问题,即垃圾评论检测问题。客户通常会根据在线评论来决定购买某物或制作某人的形象。所以,个人或公司很有可能会撰写垃圾评论来提升或降低交易者/产品/公司的地位或价值。因此,提出了一种有效的垃圾邮件检测算法 MeSMO,并在四个复杂的垃圾邮件数据集上进行了测试。将 MeSMO 的报告结果与六种最先进策略的结果进行了比较。结果对比分析证明,MeSMO 是解决垃圾评论检测问题的好技术,精度提高了 3.68%。
更新日期:2021-06-22
中文翻译:
垃圾评论检测问题的模因蜘蛛猴优化
蜘蛛猴优化(SMO)算法模仿蜘蛛猴的裂变-融合社会行为。通过文献可以明显看出,SMO 是一种基于群体的竞争算法,用于解决现实生活中的困难问题。SMO 的搜索过程因随机成分而有点偏差,随机成分以高探索性搜索步骤驱动它。此处提出了一种具有模因搜索的混合 SMO,以提高 SMO 的局部搜索能力。新开发的策略名为模因 SMO (MeSMO)。此外,所提出的基于 MeSMO 的聚类方法应用于解决大数据问题,即垃圾评论检测问题。客户通常会根据在线评论来决定购买某物或制作某人的形象。所以,个人或公司很有可能会撰写垃圾评论来提升或降低交易者/产品/公司的地位或价值。因此,提出了一种有效的垃圾邮件检测算法 MeSMO,并在四个复杂的垃圾邮件数据集上进行了测试。将 MeSMO 的报告结果与六种最先进策略的结果进行了比较。结果对比分析证明,MeSMO 是解决垃圾评论检测问题的好技术,精度提高了 3.68%。