电子商务人才需求组合数据自适应提取仿真

Adaptive Extraction Simulation of E-Commerce Talent Demand Combination Data

对电商人才需求组合数据提取的最终目的是要将有用的人才信息推荐给企业,使企业招聘更加依赖于个性化,提出一种基于模糊规则和相似度计算的电商人才需求组合数据自适应提取方法。通过选取电商人才需求组合数据特征,并计算搜索关键词在选取特征上的隶属度。同时为了拉开电商人才需求组合数据在不同特征上的隶属度差距,突出电商人才需求组合数据某一方面或某几方面特征的重要性,赋予重要特征较大权重,总结电商人才需求组合数据提取模糊规则。在此基础上,分别计算电商人才需求组合数据文本中的词语相似度和句子相似度。根据计算结果向企业用户推荐与其搜索需求相关度较高的电商人才需求组合数据,实现电商人才需求组合数据自适应提取。仿真结果表明,上述方法克服了当前方法存在提取准确率不高、提取召回率较低、综合评价指数较差以及提取效率低下的弊端,实现了电商人才需求组合数据自适应提取,对企业招聘提供个性化推荐具有重要作用。

The purpose of extraction for combination data of e-commerce talent demand is to recommend useful talent information to enterprises, so that enterprise recruitment is more dependent on personalization. Therefore, this article focused on a method to extract combination data of e-commerce talent demand based on fuzzy rules and similarity calculation. By selecting e-commerce talent demand combination data, the membership degree of search keyword on the selected feature was calculated. In order to widen the gap of membership degree for combination data of e -commerce talent demand on different features, it was necessary to highlight the importance of one or several aspects of features of combination data of e-commerce talent demand. Meanwhile, the larger weight was given to the important feature. Moreover, the combination data of e-commerce talent demand was summarized and the fuzzy rule was extracted. On this basis, the word similarity and sentence similarity in text of combination data of e-commerce talent demand were respectively calculated. According to the calculation result, the combination data of e-commerce talent demand that was highly relevant to their search requirements was recommended to the enterprise users. Thus, the adaptive extraction for combination data of e-commerce talent demand was achieved. Simulation results show that the proposed method overcomes the shortcomings of low extraction accuracy, low recall rate, poor comprehensive evaluation index and low extraction efficiency. Meanwhile, the adaptive extraction of e-commerce talent demand data was completed. This method plays an important role in providing personalized recommendation for enterprise recruitment.