Philippine Computing Journal

Includes bibliographical references.

Educational Data Mining: Current Research and Open Questions -- Specifying Melodic Features for Classical and Non-Classical Melody Evaluation -- Exploring the Implications of Tutor Negativity Towards a Synthetic Agent in a Learning-by-Teaching Environment -- A Study on the Effect of Exit Widths and Crowd Sizes in the Formation of Arch in Clogged Crowds.

[Article Title: Educational Data Mining: Current Research and Open Questions/ Ma. Mercedes T. Rodrigo, p. 1-8]

Abstract: Educational data mining (EDM) refers to the application of statistical and machine learning methods to educational data in order to achieve one of four typical ends: improvement of student models, improvement of subject matter domain structures, studying pedagogical support and refining educational theories.
An interdisciplinary field, EDM draws on mathematics, computer sciences, cognitive psychology, education theory, sociology and others. This paper walks the reader through the EDM process and then discusses recent work and open questions in the first three application areas. The paper hopes to introduce young researchers to the field and suggest problems that are still open for investigation.

https://pcj.csp.org.ph/index.php/pcj/issue/view/16/PCJ%20V8%20N1%20pp1-8%202013

[Article Title: Specifying Melodic Features for Classical and Non-Classical Melody Evaluation/ Andrei D. Coronel, Ariel A. Maguyon, and Proceso L. Fernandez, p. 9-14]

Abstract: Algorithmic Composition for music is a progressive field of study. The success of an automated music generating algorithm depends heavily on the fitness function that is used to scote the generated music. This fitness function is generally based on music features that a given algorithm is programmed to measure. This study explores features that are important for melody generation by investigating those that can 1) separate classical from non-classical music based on melody, and 2) help distinguish between two specific subgenres of both classical and non-classical music - Chopin vs. Bach, and jazz vs. contemporary compositions. Use Symbolic tool was used to collect 160 standard features from 400 music files. C4.5 was then used to select significant features. A comparative analysis between the feature sets suggested by the C4.5 algorithm and suggested in a previous study of Towsey et al. was performed by running Naïve-Bayes and SVM classifiers on each feature set. The results show that the features that have been. identified in is study are better able to classily classical Com Ton-classical music. These features may, therefore, be considered when formulating melody-based fitness functions for automated. classical music generation.

https://pcj.csp.org.ph/index.php/pcj/issue/view/16/PCJ%20V8%20N1%20pp9-14%202013

[Article Title: Exploring the Implications of Tutor Negativity Towards a Synthetic Agent in a Learning-by-Teaching Environment/Ma. Mercedes T. Rodrigo, Regina Ira Antonette Geli, Aaron Ong, Gabriel Jose Vitug, Rex Bringula, Roselle Basa, Cecilio Dela Cruz, and Noburo Matsuda, p. 15-20]

Abstract: We examine the implications of negativity in free-form dialogue between student tutors and a synthetic agent in APLUS, a learning-by-teaching online learning environment for Algebra. We attempt to determine whether the negativity of a student tutor's discourse with the agent indicates that the student is learning more or less of the material and whether the feedback they give the synthetic agent is more or less accurate. We found a weak negative correlation between tutor negativity and learning gains and a strong negative correlation between tutor negativity and accuracy of feedback Negativity might indeed indicate that student tutors lack mastery of the subject matter and need assistance themselves and detecting negativity during tutoring and providing appropriate assistance might enhance the effectiveness of APLUS and other intelligent tutoring systems.

https://pcj.csp.org.ph/index.php/pcj/issue/view/16/PCJ%20V8%20N1%20pp15-20%202013

[Article Title: A Study on the Effect of Exit Widths and Crowd Sizes in the Formation of Arch in Clogged Crowds/Ma. Mercedes T. Rodrigo, Regina Ira Antonette Geli, Aaron Ong, Gabriel Jose Vitug, Rex Bringula, Roselle Basa, Cecilio Dela Cruz, and Noburo Matsuda, p. 15-20]

Abstract: The arching phenomenon is an emergent pattern formed by a c-sized crowd of intelligent, goal-oriented, autonomous, heterogeneous individuals moving towards a w-wide exit along a long W-wide corridor, where W > w. We collected empirical data from microsimulations to identify the combination reflects of c and w to the time I of the onset of and the size S of the formation of the arch. The arch takes on the form of the perimeter of a half ellipse halved along the minor axis. We measured the S with respect to the lengths of the major M and minor m axes of the ellipse, respectively. The mathematical description of the formation of this phenomenon will be an important information in the design of walkways to control and easily direct the flow of large crowds, especially during panic egress conditions.

https://pcj.csp.org.ph/index.php/pcj/issue/view/16/PCJ%20V8%20N1%20pp21-29%202013