運用資料探勘技術分析預測員工離職研究-以A公司為例

Utilizing Data Mining Technique to Anaylze the Turnover Rate for a High Technology Company

俞治文(Chih-Wen Yu)

指導教授 ： 劉俞志

元智大學/資訊學院/資訊管理學系/碩士(2009年)

https://doi.org/10.6838/YZU.2009.00260

摘要

經過多年努力，中華民國台灣已發展成為高科技島嶼，從新竹、南部科學園區，至中部科學園區，成立了數百家工廠。相關技術難從學校或其他地方習得，除非曾經從事該項行業，因此，對於這些勞工雇主而言，如何招募到穩定員工降低離職率，將是一大挑戰。
由於近年來資訊科技發蓬勃發展，使得電腦科技技術也隨之受惠，以往需要大型電腦計算的公式，在今日已可透過一般電腦計算得知，本研究藉由類神經網路、決策樹的技術，試圖於離職資料中尋找離職人員，與離職之因素、年齡、學歷是否有相關聯性。希望能提供企業管理者，在人員管理與運用時，多一項參考資料。

關鍵字

離職, 資料探勘, 類神經網路, 決策樹

並列摘要

With many years effort, Taiwan has been transferred to Hi-Tech island and there are hundreds of factories in Shing-Chu, Southern Science Park to Middle Science Park. The Hi-Tech technique is hard to learn from schools or the other places unless employee had participated in that area. How to hire stable employees to minimize turnover rate is a big challenge to the employers.
Recently the information technologies develop rapidly and consequently associate computer sciences also get benefit. In the past, some of the formula need super computer to solve the problem. Today only need desktop can get the answer. In this research, utilize neural network and decision tree techniques to mine the quitting list. Try to find the employee’s quit reason, age, and education are any connection. To provide the manager a reference when they manger employee.

並列關鍵字

Turnover, Data Mining, Neural Network, Decision Tree

參考文獻

[1] Chien-Hung Wei and Ying Lee, “Development of Freeway Travel Time Forecasting Models by Integrating Different Sources of Traffic Data”, National Cheng Kung University, Vol.56, No.6, P3682-3694, 2007.

[2] Chrispin Ntungo and Milton Boyd, “Commodity Futures Trading Performance Using Neural Network Model Versus Arima Models”, The Journal of Futures Markets, Vol.18, No.18, P956, 1998.

[4]Gwo-Ching Liao, “A Novel Particle Swarm Optimization Approach Combined with Fuzzy Neural Networks for Short-Term Load Forecasting”, IEEE member, P1-6, 2007.

[5] Leilei Pan, Lixu Qin, Simon X. Yang and Jiangping Shuai, “A Neural Network-Based Method for Risk Factor Analysis of West Nile Virus”, Risk Analysis: The International Journal, Vol.28, No.2, P487-496, 2008.

[6] James V. Hasnsen and Ray D. Nelson, “Data mining of time series using stacked generalizes”, Brigham Young University, No.43, P173-184, 2002.