Esma Fatima Bilgin-Tasdemir

Esma Fatima Bilgin-Tasdemir
Istanbul Medeniyet Universitesi · Faculty of Arts and Humanities

Doctor of Engineering

About

5
Publications
1,682
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
11
Citations
Citations since 2016
4 Research Items
11 Citations
201620172018201920202021202201234
201620172018201920202021202201234
201620172018201920202021202201234
201620172018201920202021202201234
Learn about citations on ResearchGate
Introduction
Skills and Expertise
Handwriting Recognition
Pattern Recognition
Machine Learning
Feature Extraction
Neural Networks and Artificial Intelligence
Classification
Pattern Classification
Supervised Learning

Publications

Publications (5)
Samples of characters with potential delayed strokes: a ‘i’ with dot, b...
Detected delayed strokes (shown in red) are embedded as the arrows...
Hat feature value is 1 for points lying below the delayed strokes (in...
Examples for sources of error with resulting incorrect embeddings: a, b...
A comparative study of delayed stroke handling approaches in online handwriting
Article
Full-text available
  • Mar 2019
Delayed strokes, such as i-dots and t-crosses, cause a challenge in online handwriting recognition by introducing an extra source of variation in the sequence order of the handwritten input. The problem is especially relevant for languages where delayed strokes are abundant and training data are limited. Studies for handling delayed strokes have ma...
Figure 1. Sample handwritten words from the Elementary Turkish dataset:...
Table 1 . BOUN Corpus details. M is used as an abbreviation for million.
Table 2 . Lexical units obtained from the BOUN corpus.
Table 4 . Coverage rates for stem+ending-based large vocabulary task...
Large vocabulary recognition for online Turkish handwriting with sublexical units
Article
Full-text available
  • Sep 2018
We present a system for large vocabulary recognition of online Turkish handwriting, using hidden Markov models. While using a traditional approach for the recognizer, we have identified and developed solutions for the main problems specific to Turkish handwriting recognition. First, since large amounts of Turkish handwriting samples are not availab...
Figure 1. Sample handwritten words from the Elementary Turkish dataset:...
BOUN Corpus details. M is used as an abbreviation for million.
Lexical units obtained from the BOUN corpus.
Coverage rates of word-based large vocabulary task.
+2 Coverage rates for stem+ending-based large vocabulary task (the chosen...
Large vocabulary recognition for online Turkish handwriting with sublexical units
Article
Full-text available
  • Sep 2018
We present a system for large vocabulary recognition of online Turkish handwriting, using hidden Markov models. While using a traditional approach for the recognizer, we have identified and developed solutions for the main problems specific to Turkish handwriting recognition. First, since large amounts of Turkish handwriting samples are not availab...
Figure 2.1: A HMM with three states and 4-symbol observation alphabet....
Figure 2.2: Word HMM as concatenation of character HMMs
Figure 2.3: HMM topologies : 1) Linear 2) Bakis-type 3) Ergodic
Figure 3.2: The simple lattice expanded with a bi-gram model
+27 Figure 3.3: The simple lattice expanded with a tri-gram model
A Large Vocabulary Online Handwriting Recognition System for Turkish
Thesis
Full-text available
  • Jun 2018
We present a system for large vocabulary recogmition of online Turkish handwriting, using Hidden Markov Models. While using a traditional approach for the recognizer, we have identified and developed solutions for the main problems specific to Turkish handwriting recognition. First, since large amounts of Turkish handwriting samples are not availab...
An Overview of Ottoman to Turkish Machine Transliteration System
Conference Paper
  • May 2012

Network

Cited
View All
Cited By
View All

Projects

Project (1)
Recognition of Ottoman Printed Naskh Script
Project
To develop a recognition system for Ottoman Printed Naskh Script from line images. It aims to achieve taht goal using limited amount of real data.