arXiv subjects

Abdul Rahman Sherzad

Publications and source records attributed to Abdul Rahman Sherzad.

3 recordsLinked to original sources

arXiv2018

Data is the Fuel of Organizations: Opportunities and Challenges in Afghanistan

In this paper, the author at first briefly outlines the value of data in organizations and the opportunities and challenges in Afghanistan. Then the author takes the Kankor (National University Entrance Exam) data, particularly names of participants, locations, high schools and higher education institutions into account and explains how these data, that organizations in Afghanistan do not use for anything, can be useful in several cases and areas. The application of these data is shown through cases such as Auto filling missing values, identifying names of people, locations, and institutions from unstructured text, generating fake data to benchmark the database and web application performance and appearance, comparing and matching high school data with Kankor data, producing the top-n male and female names very common in Afghanistan or province-wise, and the data mining application in education and higher education institutions.

cs.CY↗

arXiv2017

Applicability of Educational Data Mining in Afghanistan: Opportunities and Challenges

The author's own experience as a student and later as a lecturer in Afghanistan has shown that the methods used in the educational system are not only flawed, but also do not provide the minimum guidance to students to select proper course of study before they enter the national university entrance (Kankor) exam. Thus, it often results in high attrition rates and poor performance in higher education. Based on the studies done in other countries, and by the author of this paper through online questionnaires distributed to university students in Afghanistan - it was found that proper procedures and specialized studies in high schools can help students in selecting their major field of study more systematically. Additionally, it has come to be known that there are large amounts of data available for mining purposes, but the methods that the Ministry of Education and Ministry of Higher Education use to store and produce their data, only enable them to achieve simple facts and figures. Furthermore, from the results it can be concluded that there are potential opportunities for educational data mining application in the domain of Afghanistan's education systems. Finally, this study will provide the readers with approaches for using Educational Data Mining to improve the educational business processes. For instance, predict proper field of study for high school graduates, or, identify first year university students who are at high risk of attrition.

cs.CY↗

arXiv2016

Using Descriptive Analytics for the Improvement of National University Entrance Exam: A case study in the Context of Kankor in Afghanistan

In Afghanistan, High school graduates, to continue higher education, need to pass the National University Entrance Exam (Kankor). Kankor is very important and requires further research as it is used as the only means to identify the participant's competence and skills which leads to a qualified higher education and manpower, employment opportunities, allocation of human capital in appropriate fields of study, and leading to a more specialized labor force in the long run. Having a better picture of Kankor in Afghanistan is essential before employing educational data mining techniques and other social and pedagogical approaches to support the students and the educational institutions. Hence, the author carried out a research on Kankor, in which he assessed and analyzed Kankor from different angles. This research paper provides positive contribution to the Ministry of Higher Education (MoHE), particularly, the Kankor committee, and other educational institutions - introducing the importance of data as an asset and a proper structure of recording and producing data effective for research studies, paving the way for in-depth descriptive and predictive analytics which improves the situation within the context of the existing structural models in Afghanistan. The author's research target is to study Kankor using descriptive analytics supported by Kankor Results Data (KRD) of 1.5 million records from 2003-2015, Detailed Kankor Results Data which is a subset of KRD from 2004-2006 of 120,000 records, High school performance and grades data from 2011-2013 of 6,000 records, data collected through conducting more than 2500 questionnaires among academicians as well as the author's personal observations, findings, and analyses as a student and faculty member at the University in the field of computer science.

physics.ed-ph↗