Email Classification Model for Workflow Management Systems

Takorn PREXAWANPRASUT, Piyanuch CHAIPORNKAEW

Abstract


The researchers observed and studied the business operations of 3 startup businesses in the export/import field. It was found that employees and their clients mostly communicate via email. Therefore, crucial business data are conveyed in email contents. Whenever employees need to find information, the first place they look for such data is email. The owners of businesses are concerned about this issue, so they proposed to buy a new workflow management system to help in managing their business transactions. The difficulty of implementing the new workflow management system is in migrating existing emails into the system. A new workflow management system should also be able to classify any incoming emails into categories. The researchers noticed that there were some keywords that frequently occurred in email contents in the same categories. Therefore, the researchers implemented a program to categorize the emails based on the words found in email messages. There are 2 parameters which affect the accuracy of the program. The first parameter is the number of words in a database compared to the sample emails. The second parameter is an acceptable percentage to classify emails. The results of this research demonstrated that the number of words in a database compared to the sample emails should be 9, and the acceptable percentage to categorize emails should be 30 %. When this rule was applied to categorize 8,751 emails, the accuracy of this experiment was approximately 73.6 %. The next phase is to order emails in each category based on their characteristics. Finally, the program extracts essential data from structured emails and prepares them for the new workflow management system.


Keywords


Business operations, startup business, import/export field, email, business data, workflow management system, business transactions, migrating

Full Text:

PDF

References


I Alsmadia and I Alhamib. Clustering and classification of email contents. J. King Saud Univ. Comput. Inform. Sci. 2015; 27, 46-57.

I Katakis, G Tsoumakas and I Vlahavas. Web Data Management Practices: Emerging Techniques and Technologies. Idea Group Publishing, Pennsylvania, 2006, p. 220-43.

T Ayodele, R Khusainov and D Ndzi. Email classification and summarization: A machine learning approach. In: Proceedings of the IET Conference on Wireless, Mobile and Sensor Networks. Shanghai, China. 2007, p. 805-8.

T Ayodele, S Zhou and R Khusainov. Email grouping and summarization: An unsupervised learning technique. In: Proceedings of the WRI World Congress on Computer Science and Information Engineering. Los Angeles, USA, 2009, p. 575-9.

N Kushmerick and T Lau. Automated email activity management: An unsupervised learning approach. In: Proceedings of the 2005 International Conference on Intelligent User Interfaces. San Diego, USA, 2005, p. 67-74.

D Schuff, O Turetken, JD Arcy and D Croson. Managing e-mail overload: Solutions and future challenges. Computer 2007; 2, 31-6.


Refbacks

  • There are currently no refbacks.




http://wjst.wu.ac.th/public/site/images/admin/image012_400

Online ISSN: 2228-835X

http://wjst.wu.ac.th

Last updated: 2 August 2017