<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
<channel>
<title>Seventeenth International Conference On Computer Applications (ICCA 2019)</title>
<link>https://onlineresource.ucsy.edu.mm/handle/123456789/25</link>
<description/>
<pubDate>Mon, 08 Jun 2026 10:28:25 GMT</pubDate>
<dc:date>2026-06-08T10:28:25Z</dc:date>
<item>
<title>Myanmar Homonym Disambiguation System</title>
<link>https://onlineresource.ucsy.edu.mm/handle/123456789/1233</link>
<description>Myanmar Homonym Disambiguation System
Hlaing, Zar Zar; Thida, Aye
Natural Language Processing (NLP) is one of&#13;
the most important research areas in Human&#13;
Language. One of the challenges in Natural Language&#13;
Processing (NLP) is to resolve ambiguous homonyms&#13;
or homonym errors in sentences. Myanmar Homonym&#13;
Disambiguation System is the kind of Word Sense&#13;
Disambiguation System in Natural Language&#13;
Processing. This system is needed for Myanmar Word&#13;
Segmentation and Spell Checker System. If the&#13;
sentence contains incorrect homonyms, this sentence&#13;
cannot be segmented correctly. Moreover, incorrect&#13;
usage of homonyms is a common problem in Myanmar&#13;
to English translation. In this paper, Myanmar&#13;
Homonym Disambiguation System has been&#13;
described. This system detects homonym errors or&#13;
ambiguous homonyms and then resolves these errors&#13;
by using Corpus-Based N-Gram Model. Myanmar&#13;
Text Corpus is also needed in calculation of N-Gram&#13;
Model for this system. After resolving homonym&#13;
errors, the system will output the sentence with correct&#13;
homonyms.
</description>
<pubDate>Wed, 27 Feb 2019 00:00:00 GMT</pubDate>
<guid isPermaLink="false">https://onlineresource.ucsy.edu.mm/handle/123456789/1233</guid>
<dc:date>2019-02-27T00:00:00Z</dc:date>
</item>
<item>
<title>Mobile based Text Image Recognition using Deep Learning Approach</title>
<link>https://onlineresource.ucsy.edu.mm/handle/123456789/1232</link>
<description>Mobile based Text Image Recognition using Deep Learning Approach
Maung, Saw Zay Maung; Aye, Nyein
Recognizing text image from mobile phone is a&#13;
challenge task for limited capacity and processing&#13;
power. And also the accuracy of the system is&#13;
important for text image recognition system. In this&#13;
system, we aimed to develop a Text Image Recognition&#13;
System for mobile environment using Myanmar&#13;
Character Dataset. Firstly, the image captured from&#13;
mobile phone’s camera and then segment each&#13;
connected character using Connected Labeling&#13;
Algorithm. After that the segmented characters input&#13;
into the Convolutional Neural Network by passing&#13;
layer by layer to get feature maps for recognizing the&#13;
words in a given text image.
</description>
<pubDate>Wed, 27 Feb 2019 00:00:00 GMT</pubDate>
<guid isPermaLink="false">https://onlineresource.ucsy.edu.mm/handle/123456789/1232</guid>
<dc:date>2019-02-27T00:00:00Z</dc:date>
</item>
<item>
<title>Machine Learning Based Android Malware Detection using Significant Permission Identification</title>
<link>https://onlineresource.ucsy.edu.mm/handle/123456789/1231</link>
<description>Machine Learning Based Android Malware Detection using Significant Permission Identification
Kyaw, May Thu; Kham, Nang Saing Moon
The increasing popularity of smartphones and&#13;
tablets has introduced Android malware which is&#13;
rapidly becoming a potential threat to users. A recent&#13;
report indicates the alarming growth rate of Android&#13;
malware in which a new malware is introduced in&#13;
every second more precisely in 10 seconds. To against&#13;
this dangerous malware growth, this paper proposes&#13;
a scalable malware detection system using permission&#13;
analysis behavior that can identify malware apps&#13;
effectively and efficiently. We propose multi-level of&#13;
pruning procedures to identify the most significant&#13;
permission instead of extracting all permissions. The&#13;
propose system utilizes supervised classification&#13;
method in machine-learning to classify different&#13;
families of benign and malware apps. We found that&#13;
22 permissions are significant actually. Our&#13;
evaluation finds that the analysis time of using these&#13;
22 permissions are 4 to 32 times less than using all&#13;
permissions. The results show that most of malware&#13;
apps are located the unnecessary permission on&#13;
AndroidManifest.xml to inject the malicious codes in&#13;
the apps.
The authors are grateful for the supports&#13;
provided by University of Computer Studies, Yangon.
</description>
<pubDate>Wed, 27 Feb 2019 00:00:00 GMT</pubDate>
<guid isPermaLink="false">https://onlineresource.ucsy.edu.mm/handle/123456789/1231</guid>
<dc:date>2019-02-27T00:00:00Z</dc:date>
</item>
<item>
<title>Location-based Service Personal Navigation System for Nay Pyi Taw City</title>
<link>https://onlineresource.ucsy.edu.mm/handle/123456789/1230</link>
<description>Location-based Service Personal Navigation System for Nay Pyi Taw City
Thu, Myint; Sein, Myint Myint
With the development of mobile Internet, more&#13;
and more people begin to get convenient service by&#13;
mobile phones. Obtaining one's current location by&#13;
GPS positioning or network positioning has become&#13;
one of the important foundations in most applications&#13;
of location based service. In this paper, we designed&#13;
and implemented a personalized positioning and&#13;
navigation system based on the Android platform.&#13;
With the combination of GPS positioning and network&#13;
positioning, and using OpenStreetMap API, this&#13;
system provides the following functions: view the&#13;
current location, define the current location by user&#13;
and get the navigation route.
</description>
<pubDate>Wed, 27 Feb 2019 00:00:00 GMT</pubDate>
<guid isPermaLink="false">https://onlineresource.ucsy.edu.mm/handle/123456789/1230</guid>
<dc:date>2019-02-27T00:00:00Z</dc:date>
</item>
</channel>
</rss>
