期刊
EMERGING RESEARCH IN ELECTRONICS, COMPUTER SCIENCE AND TECHNOLOGY, ICERECT 2018
卷 545, 期 -, 页码 549-568出版社
SPRINGER-VERLAG SINGAPORE PTE LTD
DOI: 10.1007/978-981-13-5802-9_50
关键词
Grouping; Feed-forward neural network; Range of values; Neural networks; Machine learning; Support vector machines
The theme of this paper is to analyze how a student can perform in an examination by answering the questions by selecting from various units. This paper focuses mainly on the common units of the questions to which students can make their maximum attempts to write the answers and the units to which the students will make rare attempts. It is considered to be one of the attempts made in this regard which will enlighten teaching community as well as college management to focus where exactly risks will be there to both students and teachers. From last 4 years, AICTE has made mandatory rules to all the technical education institutes in this regard as a part of accreditation certificate. It helps in framing the syllabus as well as setting the question papers by considering number of chapters for a given unit and number of units for a given subject. The main theme behind this performance analysis lies with the units with which maximum students will make an attempt to answer which in turn improve the confidence level of the teacher in setting the type of questions and framing trickiness in the questions. Here, there were two approaches which have been suggested both in traditional manual method and soft computing method. In the manual method, a program application was written which makes the group of the students based on the total marks scored by the students. In soft computing method, it was used with the basic feed-forward neural network to group the students by taking the split marks of individual units of the student. These splits marks will be considered as features to train the network. At the end, a fruitful positive result was received in both manual methods and soft computing methods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据