Book ChapterDOI
A Neural Based Approach to Evaluate an Answer Script
M. R. Thamizhkkanal,V. D. Ambeth Kumar +1 more
- pp 1187-1207
Reads0
Chats0
TLDR
This paper gives cue of the possibility to cut down the teacher’s workload on open questions by a component managing a neural network-based model of the students’ decisions, involved in a peer-assessment task.Abstract:
Assessment of answers with particular questions is a heavy task as in assumption that all the students answers have to be awarded. In this paper we give cue of the possibility to cut down the teacher’s workload on open questions by a component managing a neural network-based model of the students’ decisions, involved in a peer-assessment task. The answer is recognized by OCR and it converted by machine readable format. The network of constraints and relations established among the answers through the students’ knowledge, allows us to compare and relate with set of possible keyword of the database. Convolution neural network plays a vital role in comparison of answer database and student database. The Receiver Operating Characteristic curve are constructed based on the accuracy of students marks. Based on the comparison result obtained, the accuracy of student answer is measured and awarded. Our computer system suggests that the subset of the answers is evaluated with database answer in which the performance of evaluation is measured. This is used to reduce workload of humans and it automatically evaluate the answer. It is mainly used in schools, colleges, university etc.read more
References
More filters
Proceedings Article
End-to-end text recognition with convolutional neural networks
TL;DR: This paper combines the representational power of large, multilayer neural networks together with recent developments in unsupervised feature learning, which allows them to use a common framework to train highly-accurate text detector and character recognizer modules.
Proceedings ArticleDOI
Convolutional Experts Constrained Local Model for 3D Facial Landmark Detection
TL;DR: To achieve best performance on the Menpo3D dense landmark detection challenge, a network that maps the output of CE-CLM to 84 landmarks called Adjustment Network, and a Deep Residual Network called Correction Networks that learns dataset specific corrections for CE- CLM are used.
Proceedings ArticleDOI
Image recognition based on deep learning
Meiyin Wu,Li Chen +1 more
TL;DR: The experiment results show that deep learning does have an excellent feature learning ability and it don't need to extract features manually, and can learn more nature features of the data.
Journal ArticleDOI
Multilingual Character Segmentation and Recognition Schemes for Indian Document Images
Parul Sahare,Sanjay B. Dhok +1 more
TL;DR: In this paper, robust algorithms for character segmentation and recognition are presented for multilingual Indian document images of Latin and Devanagari scripts, where primary segmentation paths are obtained using structural property of characters, whereas overlapped and joined characters are separated using graph distance theory.
Proceedings ArticleDOI
Bengali handwritten character recognition using deep convolutional neural network
TL;DR: A convolutional deep model to recognize Bengali handwritten characters is proposed that first learnt a useful set of features by using kernels and local receptive fields, and then it has employed densely connected layers for the discrimination task.
Related Papers (5)
Online Subjective answer verifying system Using Artificial Intelligence
G. Jagadamba,Chaya Shree G +1 more