Ajou University repository

A Study on Job Satisfaction Factors in Retention and Turnover Groups using Dominance Analysis and LDA Topic Modeling with Employee Reviews on Glassdoor.com
Citations

SCOPUS

0

Citation Export

Publication Year
2018-01-01
Journal
ICIS 2017: Transforming Society with Digital Innovation
Publisher
Association for Information Systems
Citation
ICIS 2017: Transforming Society with Digital Innovation
Keyword
Dominance AnalysisEmployee Review Data AnalysisInductive ReasoningJob SatisfactionLDA Topic ModelingText mining
Mesh Keyword
Data analysis techniquesDominance analysisEffective managementEmployee retentionInductive reasoningReview datumTopic ModelingUnstructured texts
All Science Classification Codes (ASJC)
Computer Science ApplicationsInformation Systems
Abstract
HR analytics is an important area for the application of big data analysis techniques, and the organizational insight that it provides enables effective management of employees. In this paper, we analyze employee review data posted on a representative third-party employee review website. We identify the relative importance of factors affecting job satisfaction and then extract topic differences after classifying employees according to retention and turnover. First, LDA Topic Modeling by adopting n-grams is performed on unstructured text data to analyze employee review data. Second, a dominance analysis is conducted to examine the relative importance of job factors. We found that the “Culture and Values†and “Senior Management†factors have the highest influence on both retention and turnover. Our model follows a novel approach in applying the analysis of reviews and text mining to the HR domain and will be of practical relevance for enhancing employee retention.
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36355
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85126507204&origin=inward
DOI
https://doi.org/2-s2.0-85126507204
Type
Conference
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Kang, Ju Young Image
Kang, Ju Young강주영
Department of Business Intelligence
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.