A Fuzzy-Based Dynamic Bandwidth Allocation Approach for Campus Area Networks

Paredes, Rolysent K.; Fajardo, Arnel C.; Sison, Ariel M.

doi:10.1109/icetas.2018.8629213

Cited by 2 publications

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It develops the model to discover hidden patterns, relationships, and phenomena helpful in decision-making. Data mining uses various algorithms and techniques for decision-making (Paredes, Fajardo, Sison, 2018;Sugiyarti et al, 2018;Lumasag et al, 2020); prediction and forecasting (Orong, Sison, & Hernandez, 2018;Rady & Anwar, 2019); detection (Souri & Hosseini, 2018); classification (Jalota & Agrawal, 2019); clustering (Kausar et al, 2018); association (Caroro, Sison, & Medina, 2018;Caroro, Sison, & Medina, 2019), and other purposes. The KDD methodology exploits a vast collection of useful data to determine varied data mining results to aid decisions using various effective algorithms such as pre-processing, transformation, and data mining.…”

Section: Data Miningmentioning

confidence: 99%

Modeling Teaching Brand using Multisource Feedback

Caroro,

Orong,

Piquero

2021

JMDS

View full text Add to dashboard Cite

An individualized teaching performance evaluation identified the teacher’s handling of his class. Using multisource feedback in evaluating the teaching performance defined the teacher’s teaching performance as observed only by several evaluators. However, knowing the brand of teachers from a certain group is vitally important to define the group’s identity in their delivery of instruction. Hence, modeling the teaching performance to determine the brand of teaching enables academic administration to easily monitor and evaluate the teaching performance. This study developed a model for teaching performance to determine the brand of teaching. The model used the faculty performance evaluation dataset in a university in the southern Philippines. The model pre- processed the data using the expectation-maximization algorithm and then classified them using the J48 classifier running in WEKA to generate decision trees. Furthermore, the model partitioned the original data set into ten partitions, making nine training sets, and the remaining are for validation or test sets using the 10-fold cross-validation testing. Based on the dataset used, the model was able to identify prominent teaching brands and disregard attributes with no noticeable contribution to the classification. Keywords : classification algorithm, data mining, teaching performance, teacher brand model, policy enhancement

show abstract