情报科学 ›› 2023, Vol. 41 ›› Issue (9): 130-137.

• 业务研究 • 上一篇    下一篇

基于数据挖掘的就业需求信息资源采集研究

  

  • 出版日期:2023-09-01 发布日期:2023-10-07

  • Online:2023-09-01 Published:2023-10-07

摘要: 【 目的/意义】随着互联网的不断普及,海量就业信息资源持续涌现,为了能够为待就业学生提供更具针对 性的就业指导,满足更多高校学生的就业需要,本文提出基于大数据挖掘技术的学生就业需求信息资源采集研究。 【方法/过程】首先,本文基于网络信息爬取模块、就业需求信息挖掘模块、资源整合模块对就业需求信息资源采集 总体架构进行优化;并结合就业专业-就业岗位-就业需求信息-高校信息的组织体系,实现就业岗位和就业需 求信息以及高校信息间的信息整合。然后,利用就业需求信息挖掘统一计算系统对已整合结果进行内容标准化处 理,再利用深度学习算法分别进行就业异常数据筛查和特定异常就业数据特征识别,构建就业需求信息资源采集 系统。最后,凭借E-S-Qual量表对系统中的就业信息采集质量进行评价。【结果/结论】结果表明,所提出的研究方 法可以向用户推送全面的就业需求信息,并能全面分析需求信息推荐效果,有效为待就业学生提供相关就业指导。 【局限/创新】但由于本次实证研究较单一,研究结果存在一定的局限性,日后可结合更多案例对所提方法进行验 证,使结果更具说服力。

Abstract: 【 Purpose/significance】 With the continuous popularization of the Internet, massive employment information resources con⁃ tinue to emerge. In order to provide more targeted employment guidance for students to be employed and meet the employment needs of more college students, this paper proposes a research point on the collection of information resources for students′ employment needs based on big data mining technology.【 Method/process】 Firstly, this article optimizes the overall architecture of employment de⁃ mand information resource collection based on the network information crawling module, employment demand information mining mod⁃ ule, and resource integration module; And combine the organizational system of employment majors, employment positions, employ⁃ ment demand information, and university information to achieve information integration between employment positions, employment demand information, and university information. Then, the unified calculation system for employment demand information mining is used to standardize the content of the integrated results, and deep learning algorithms are used to screen employment abnormal data and identify specific abnormal employment data features, respectively, to construct an employment demand information resource col⁃ lection system. Finally, evaluate the quality of employment information collection in the system using the E-S-Qual scale. 【Result/ conclusion】 The results show that the proposed research method can push comprehensive employment demand information to users, comprehensively analyze the effect of demand information recommendation, and effectively provide relevant employment guidance for students to be employed.【 Innovation/limitation】 However, due to the single empirical research, the research results have some limita⁃ tions. In the future, more cases can be used to verify the proposed methods, making the results more convincing.