V3I1P38

A Classification Algorithm for a Gender-Adaptive Menu-Based Chatbot for Employees’ Vacation Leave Requests Management

Mustafa Kadhim Taqi^1*

Abstract

In chatbots, a key challenge lies in generating replies adapted to users’ personalities. In this paper, first, we presented a novel Telegram bot-based employee’s vacation leave requests management system for public sector employees in Iraq. In this system, an employee has to select a vacation type from the vacation types menu. However, the vacation types in this menu are not the same for both genders. Therefore, a classification algorithm based on gender is proposed. For better classification performance, this algorithm combines three well-known online gender inference services based on the user’s first name. This is to improve the user experience. An experiment to test the accuracy of the proposed algorithm has been conducted. The result confirms the feasibility of the proposed algorithm with high gender identification accuracy. To our knowledge, this is the first attempt to create a chatbot for the public sector in Iraq.

Keywords:

Telegram bot, public and private sector, vacation management system, classification.