| Volunteering in Systems Development: A Case Study |
| Pages: 259-265 (7) | [Full Text] PDF (636 KB) |
| D Lima, G Disperati |
| Lato Sensu Postgraduate in Information Systems Management, Federal Institute of Science, Education and Technology of Sao Paulo, Guarulhos, Sao Paulo, Brazil |
|
Abstract - Understanding what motivates and demotives professionals is a challenge for project leaders, especially when it comes to volunteering. Although studies with development teams in companies or open source projects present important motivational factors, it is still necessary to differentiate these professionals from volunteers. Therefore, this case study brought together a team of developers to work on a volunteer project for the Legion of Mary, an association linked to the Catholic Church, to identify the groups motivations and demotivations during the development of the project. Through interviews, training, questionnaires and meetings, it was observed that, in a team of volunteers, motivational factors such as the possibility of acquiring knowledge, helping the institution, ease with technology or time to dedicate to the project contribute to the beginning, permanence or giving up voluntary work. |
| Index Terms - Motivation, Volunteer Work, Legion of Mary, Information Technology |
|
C itation - D Lima, G Disperati. "Volunteering in Systems Development: A Case Study." International Journal of Computer Science and Software Engineering 8, no. 11 (2019): 259-265. |
| Text Mining Preprocessing In Times of Python vs MVCS |
| Pages: 266-275 (10) | [Full Text] PDF (452 KB) |
| J Quillo-Espino, RM Romero-Gonzalez, FJ Paulin-Martinez |
| Faculty of Informatics, Autonomous University of Queretaro, Santiago de Queretaro, Queretaro 76230, Mexico |
|
Abstract - This investigation illustrates in a clear and understandable manner the preprocessing execution of text mining, through two programming languages: Microsoft Visual C Sharp (MVCS) and Python. It pretends to demonstrate the usefulness of each analyzed method, mentioning which one is more efficient. The main purpose is to show the advantages and disadvantages of natural language preprocessing tools in Python, versus a personalized programming, specifically for text mining preprocesses in Microsoft Visual C Sharp. |
| Index Terms - PNLTK, Preprocesses, Python, Text-Mining, Visual- C-Sharp |
|
C itation - J Quillo-Espino, RM Romero-Gonzalez, FJ Paulin-Martinez. "Text Mining Preprocessing In Times of Python vs MVCS ." International Journal of Computer Science and Software Engineering 8, no. 11 (2019): 266-275. |
| Dynamic Data Race Detection in Java Applications |
| Pages: 276-279 (4) | [Full Text] PDF (407 KB) |
| MS Unluturk, I Sofuoglu, R Atar, E Nakilcioglu, A Unlu |
| Department of Software Engineering, Yasar University, Izmir, TurkeyEricsson ITU Ayazaga Kampusu Koru Yolu, ARI2 Teknokent Binasi B Blok 4-1, 34390, Maslak, Istanbul, Turkey |
|
Abstract - A data race is occurred when two threads concurrently access a shared variable and at least one of those threads is the write thread. Data race related bugs are hard to catch because they can happen only under very specific conditions. Even the multi-threaded application passes all the tests successfully, it does not mean that the data races are all removed from the source code. As a result, it is important to have tools to catch the existing data races and the new ones as soon as they are added into the source code. The problem of catching data race is known to be NP-hard. There are number of approaches for catching the data races. Basic ones are the static, on-the-fly and postmortem. Two multithreaded software applications developed at Ericsson Turkey called Notification Engine (NE) and Multi-Mediation (MM) have been already tested by using unit tests and regression tests. However, these two applications had never been investigated using a data race detector. Our main objective here is to implement and deploy a continuous methodology for finding data races in these kind of multithreaded applications. Furthermore, improvement in the performance of the proposed analysis tools and a significant reduction of false positives have been achieved. |
| Index Terms - Dynamic Data Race Detection, Static Data Race Detection, Multi-threaded Applications, On-the-fly Data Race Detector, Post-mortem Data Race Detector |
|
C itation - MS Unluturk, I Sofuoglu, R Atar, E Nakilcioglu, A Unlu. "Dynamic Data Race Detection in Java Applications." International Journal of Computer Science and Software Engineering 8, no. 11 (2019): 276-279. |
| Study of Cloudlet Deployment Challenges under Multiple Environment |
| Pages: 280-284 (5) | [Full Text] PDF (390 KB) |
| M Nawaz, A Ali, MS Siddiqui |
| Computer Science, UET Taxila, University of Engineering & Technology, Taxila, Pakistan3 Computer Science, UET Peshawar, University of Engineering & Technology, Peshawar, Pakistan |
|
Abstract - Cloud computing is a recently developing innovation. Cloud computing is generally being used in light of the fact that it gives chance to powerfully scale the registering resources for applications. Additionally, the resources are productively shared among clients utilizing virtualization innovation. In the course of recent decades, the world economy has quickly moved from assembling to more assistance arranged. Since cloud computing condition has huge arrangement and cost related with it, testing applications and asset portion strategies are exceptionally testing. Additionally, in todays age, cell phones are broadly utilized like cell phones, PDA, PCs, and so on., which gives open the adaptability to get to the cloud while progressing. In this way, it is hard to deal with the huge measure of information on cloud server and assign the resources proficiently to portable clients or cell phones [3], [11]. Subsequently, another building component called CLOUDLET which originates from the union of versatile registering innovation and cloud computing innovation appeared [3]. It speaks to the center level of a 3-level progressive system: cell phone - cloudlet - cloud. A cloudlet can be seen as a . |
| Index Terms - Cloud Computing, Cloudlet, Latency, Mobile Computing, Virtualization |
|
C itation - M Nawaz, A Ali, MS Siddiqui. "Study of Cloudlet Deployment Challenges under Multiple Environment." International Journal of Computer Science and Software Engineering 8, no. 11 (2019): 280-284. |
| A Dataset of Handwritten Arabic Letters for Machine Learning Applications |
| Pages: 285-293 (9) | [Full Text] PDF (1.15 MB) |
| AM Hassanein, Y Mohammed, Y Ehab |
| Systems and Information Department, Engineering Division, National Research Centre (NRC), Dokki, Giza, Egypt. Postal Code:12622Electrical Communication and Electronics Department, School of Engineering, October University for Modern Sciences and Arts (MSA), Al Wahat Road, Giza , Egypt |
|
Abstract - A lot of research is done in the Arabic letter recognition field and several datasets are created. We aim to add to the research field a new dataset that is a benchmark with the already existing ones to be used for classification of Arabic letters. This paper provides images of a new handwritten disconnected Arabic letters dataset. A group of 19 features are calculated for each letter in the dataset. The K-nearest neighbor algorithm is applied for the classification problem of the letters. When choosing a part of the dataset including 16 letters which are visually different, the successful classification results is almost 86%. However; when choosing the whole dataset of 28 letters, the successful classification result is 61%. |
| Index Terms - Arabic letters, handwritten letters, classification problem, KNN algorithm, Holland Features |
|
C itation - AM Hassanein, Y Mohammed, Y Ehab. "A Dataset of Handwritten Arabic Letters for Machine Learning Applications." International Journal of Computer Science and Software Engineering 8, no. 11 (2019): 285-293. |