Oscar Karnalim

Detecting source code plagiarism on introductory programming course assignments using a bytecode approach

Even though there are various source code plagiarism detection approaches, most of them only concern with low-level plagiarism attack with an assumption that plagiarism is only conducted by students who are not proficient in programming. However, plagiarism is often conducted not only due to student incapability, but also because of bad time management. Thus, high-level plagiarism attack should be detected and evaluated. This paper proposes source code plagiarism detection approach which can detect most introductory-programming-course plagiarism attacks at any level by utilizing low-level instructions instead of source code tokens. Several mechanisms are also introduced to improve its effectiveness such as instruction generalization, instruction reinterpretation, method-based comparison, and method linearization. Since low-level instruction is a language-dependent feature, Java is selected as target programming language with bytecode as its low-level instruction. Based on evaluation, it can be concluded that our approach is more effective to detect most plagiarism attack types than raw source code approach on introductory programming course. This evaluation is based on plagiarism attack types that are collected through controlled experiment.

Semi-Supervised Keyphrase Extraction on Scientific Article using Fact-based Sentiment

Even though most keyphrase extraction works are focused on arbitrary text to keep its wide-applicability, several works are focused on a particular domain due to the availability of several domain-specific features. This paper proposes a new learning feature for extracting keyphrase on scientific articles. This feature is called fact-based sentiment which is implicitly occurs on novelty-based scientific articles. According to our dataset, this feature is quite effective for extracting keyphrase on scientific articles. In addition, our keyphrase extraction approach also combines unsupervised and supervised approach to exploit their combined benefits. Hidden pattern can be detected dynamically whereas candidate importance is still comparable to each other. Our approach is implemented by ranking all keyphrase candidates based on their TF-IDF score and feeding them to Deep Belief Network until N keyphrases are selected. Based on our evaluation, our approach yields moderate F-measure which is 13.22% when retrieving Top-5 keyphrases. Thus, it can be stated that our approach is quite considerable to be applied for extracting keyphrases on scientific articles.

A Quasi-Experimental Design to Evaluate the Use of PythonTutor on Programming Laboratory Session

Abstract—Educational tool is one of the prominent solutions for aiding students to learn course material in Information Technology (IT) domain. However, most of them are not used in practice since they do not properly fit student necessity. This paper evaluates the impact of an educational tool, namely PythonTutor, for completing programming laboratory task regarding data structure materials. Such evaluation will be conducted in one semester by implementing a quasi-experimental design. As a result, six findings can be deducted which are: 1) PythonTutor might positively affect student performance when the students have used such tool before; 2) Sometimes, student perspective regarding the impact of educational tool is not always in-sync with actual laboratory result; 3) the impact of PythonTutor might be improved when similar data representation is used consequently for several weeks; 4) the correlation between the use of PythonTutor and student performance might not be significant when the control and intervened group share completely different characteristics; 5) the students might experience some difficulties when they are asked to handle a big task for the first time; and 6) the students might be able to complete a particular weekly task with a promising result if the students have understood the material well.

Medical Specialists Retrieval System Using Unified Medical Language System

A large number of doctors and wide range of medical specialties can cause confusion in choosing the right medical specialist. This research aims to build a medical specialists retrieval system that corresponds with the users disease. To make the system whole, it requires the ability to differentiate a query from common words and relate it to a disease, then associate the disease to related medical specialties. The Unified Medical Language System (UMLS) is used in query handling and finding relations between a disease and medical specialties. Additionally, the search results are sorted by the nearest medical practices based on users location. This system has been evaluated by two internists which revealed an average score of 4.625 out of 5, which means relevant, of all points evaluated. Thus, provided a positive feedback to overall system performance.