310 likes | 464 Views
I ntellectual Professional. C heerfulness Morality. Source of Knowledge Blooming Like a Lotus. Knowledge is the competitive weapon of the 21 st century. Forecasting Model for the Students’ Job Turnover in Thai Industries Pirapat Chantron Prasong Praneetpolgrang
E N D
Intellectual Professional Cheerfulness Morality SourceofKnowledgeBloomingLike a Lotus Knowledge is the competitive weapon of the 21st century
Forecasting Model for the Students’ Job Turnover in Thai Industries PirapatChantron PrasongPraneetpolgrang Master of Science Program in Information Technology Sripatum University, Bangkok, Thailand
Agenda Background of the Research 1 Research Objective 2 Theories & Related Research 3 Experiments 4 Conclusions 5 Future Works 6
A number of students transfer their majors of studies or change their majors, drop or resign from the university. Background of the Research Many students in the university are not aware whether they should choose to study, any field of studies that match for them in order to work directly with their interests.
Background of the Research (Cont…) After graduating from the university and get into work, a number of students change their work or resign for the reasons that they cannot find the appropriate or proper work with their major of studies or their interests.
These are the reasons that students do not have experience and lack of information in their majors of studies. They unknown individual disciplines well enough, and they found afterward that their studies or their majors and their work didn’t fit with them. It is too late for them to start again. Background of the Research (Cont…)
Research Objective The purpose of this study is to develop forecasting model for the students’ job turnover in Thai industries.
Data Mining • Bayesian Networks • Cross-validation • Evaluation Theories and Related Research
Data mining technique is based on statistical analysis, it has been used in finding and describing structural patterns in data segmentation and predictions (Witten and Frank,2005). • This technique has been applied extensively in many industries including banking and finances, education, medical sciences and manufacturing. Theories Data Mining
Specific type of graphical model which is a directed acyclic graph (Kijsirikul,2003). All of the edges in the graph are directed and there are no cycles. Used as a classifier that gives the posterior probability distribution of the class node given the values of other attributes. Theories (cont.) Bayesian Networks
Example of Bayesian Networks Theories (cont.) Bayesian Networks (cont.) B A C P(A,B,C) = P(A | B) P(B) P(C | B) 11
Some of the data are removed before training begins. • When training is done, the data that were removed can be used to test the performance of the learned model. • The Data set is separated into two sets, called the training set and the testing set. Theories (cont.) Cross-validation
Correct Percentage = Theories (cont.) Evaluation in this System Number of correct classification Total number of classifications • Precision = • Recall = F-measure = Number of documents relevant and retrieved Total number of documents that are retrieved Number of documents relevant and retrieved Total number of documents that are relevant 2 x Precision x Recall Precision + Recall
Related Research Research in Data Mining Techniques
Related Research (cont.) Research in Data Mining Techniques
Research Experiments System Framework for the research methodology Student Database Data Pre-processing Data Mining Data Pre-processing 1 BayesianNetworks 2 Post-processing Model 3
Data mining techniques (Data Mining) were used in this research to create a relationship model between their majors, having and changing their jobs of persons in public and private organizations by studying from academic performance, profiles, and work background. Data from the total sample set were 2,536. The table of Krejcie and Morgan was used to define the sample size Research Experiments (cont.) Dataset
Random Sample Size from the Population which based on Morgan & Krejcie Table Research Experiments (cont.)
Research Experiments (cont.) Population of this study will be referred with the calculating method of Taro Yamane Formula, n = Sample size N = Population size e = The error of sampling This study allows the error of sampling on 0.05
Research Experiments (cont.) Data were used in this study and the modeling consisted of: - Information from 6 universities: 3 public and 3 private universities, Kasetsart University. Rajabhat PranakonUniversity, Rajabhat Lopburi University and private universities including Sripatum University, Durakit Bundit University and Saint John's University. - Data from 6 organizations: The CP (Research and Development), The DTAC, The Department of Transportation, Thai International Airways (Aviation Management), the Department of Cooperative, The Auditing Office and the Office of Bangkhen District Office and The Office of Disease Prevention area 1.
Research Experiments (cont.) Data from the total sample set were 2,536 23
Research Experiments (cont.) ATTRIBUTE OF DATASET
Work Change Salary Major Position Research Experiments (cont.) Experimental Results Model of the variable that effect to the work changing.
Research Experiments (cont.) == Run information === Test mode: 10-fold cross-validation === Classifier model (full training set) === Naïve Bayes Classifier not using ADTree === Summary === Correctly Classified Instances 2280 97.2634 % Incorrectly Classified Instances 256 2.7366 % Kappa statistic 0.8633 Mean absolute error 0.0742 Root mean squared error 0.1872 Relative absolute error 25.1745 % Root relative squared error 48.9402 % Total Number of Instances 2536.0000 • The predicting model for work changing was constructed in order to prove the accuracy of data mining technique by using Bayesian Networks. The result indicated that the accuracy was 97.26%. This study suggests the graduated student to used the factors that effect to his working, those are field of study, Major, Position and Salary. These variables are suitable for model constructing to predict the changing of work opportunity.
In conclusion, it was found that variables effect the description of the factors affecting the change of the job: major, position of the job and job salary rate. CONCLUSION
Future works • Applying data mining technique for prediction. In order to increase the prediction power of classification, alternative feature selection might be applied to select importance attributes before classification. • Increase sampling size in the next research, include universities sampling and organizations in order to develop the model more effectively.
References [1] K. Waiyamai, T. Rakthanmanon and C. ngsiri, “Data Mining Techniques for Developing Education in Engineering Faculty,” NECTEC Technical Journal, volume III, no.11, 2001, pp. 134-142. [2] B. Kijsirikul, Artificial Intelligence, Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, 2003. [3] J. Yingkuachat, B. Kijsirikul and P. Praneetpolgrang, “A Prediction of higher Education Students’ Graduation with Bayesian Learning and Data Mining,” in Research and Innovations for Sustainable Development Conference, 2006. [4] T. Hsia, A. Shie and L. Chen, “Course Planning of extension education to meet market demand by using data mining techniques-an example of chinkuo technology university in Taiwan,” Expert Systems with Applications, volume 34, Issue 1, 2008, pp. 596-602.
References (cont.) [5] I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques, Second Edition, Morgan Kaufmann, San Francisco, 2005. [6] WEKA, http://www.cs.waikato.ac.nz/ml/weka, 17 September 2007. [7] P. Garcia, A. Amandi, S. Schiaffino and M.Campo, “Evaluating Bayesian networks’ precision for detecting students’ learning styles,”Computer & Education, Volume 49, Issue 3, 2007, pp. 794-808. [8] M. Xenos, “Prediction and assessment of student behaviour in open and distance education in computers using Bayesian networks,” Computer & Education, Volume 43, Issue 4, 2004, pp. 345-359.
Thank You for your kind attention Parinya.ch@spu.ac.th Prasong.pr@spu.ac.th