INTRODUCTION

Background of the Study

In recent years, the field of education has increasingly turned to data-driven

methodologies to enhance teaching and learning outcomes. Educational data mining

(EDM) is frequently used to identify trends and important facts gathered from academic

data, including teacher and student evaluations [1]. This process often involves data

analysis, which is the systematic collection, purification, transformation, description,

modeling, and interpretation of data [2]. By applying these techniques, institutions can

analyze large volumes of educational data to identify patterns and trends that inform

decision-making processes.

Faculty Development Programs (FDPs), recognized as an independent

educational approach to enhance faculty members' professional competencies and

knowledge [3], represent a key area where data-driven approaches can have a profound

impact. By applying methods such as data mining techniques, institutions can transform

faculty evaluation into a more objective and actionable process that supports continuous

professional development, ultimately improving teaching practices and student outcomes

[4].

Several studies have been conducted on data mining and data-driven approaches

in education. In China, a study examined the application of data mining techniques to

improve the management and evaluation of college faculty. Using K-means clustering,

the research developed a system to analyze teacher performance and offer tailored
development recommendations. The system integrated teacher data and evaluation

processes, demonstrating that data mining tools, with a minimum support of 5% and a

confidence level of 90%, can effectively aid human resource planning and enhance

faculty competitiveness in higher education [5]. The findings suggest that data-driven

approaches can significantly optimize faculty development and institutional management

practices.

In the Philippine context, there has also been research on data mining and data-

driven approaches in education. At Taguig City University, a study developed a faculty

performance evaluation system using data analytics, employing descriptive and

developmental research methods to design and assess its effectiveness. Feedback from

faculty and administrators, collected through a structured questionnaire, showed high

acceptance of the system with a mean rating of 4.33 for functionality, reliability, and

usability, and a moderate usefulness rating of 4.41. The findings highlight the potential of

data analytics to improve faculty evaluation processes and foster professional

development in educational institutions [6]. This suggests that such systems could lead to

more efficient and objective assessments.

In a study conducted at the University of Mindanao in Davao City, Philippines,

data mining algorithms such as Naïve Bayes, C4.5, and K-Nearest Neighbor (KNN) were

employed to assess student satisfaction with a Learning Management System (LMS). The

Naïve Bayes algorithm achieved a prediction accuracy of 100%, demonstrating the

effectiveness of data mining techniques in analyzing educational datasets [7]. This

suggests that applying similar data-driven methodologies can also enhance the evaluation

processes in other areas, such as faculty performance assessment.

 Recognizing the effectiveness of data mining in the field of education and the

impact of data-driven approaches, this study aims to address the limitations in current

faculty evaluation processes at Notre Dame of Midsayap College (NDMC). The existing

manual evaluation method lacks the analytical capability to provide actionable insights

and targeted recommendations for faculty development. To address this gap, the study

proposes the development of a tool called ElevatEd, an automated evaluation analysis

system designed to enhance instructor performance at NDMC. This tool will utilize data

mining techniques to analyze historical faculty evaluation data, predict future

performance, and generate personalized recommendations for professional growth.

Statement of the Problem

This study aims to develop strategies for improving faculty performance by

analyzing historical evaluation data of the faculties at Notre Dame of Midsayap College.

Specifically, the study seeks to utilize data-mining techniques to assess historical faculty

performance data and create a data-driven tool to enhance and predict faculty teaching

performance.

Specifically, this study aims to answer the following:

1. What are the trends in the performance of the faculties from 2015 to 2023 across

different educational levels at NDMC?

a. Primary Education (Primary Ed)

b. Junior High School (JHS)

 c. Senior High School (SHS)

d. College

2. How do faculties perform in different categories based on the evaluations?

a. Classroom Management

b. Impact on Student Learning and Personal Relationships

c. Knowledge of Subject Matter

d. Methodology

e. Personal Qualities

f. Regular Instruction

3. What is the overall rate of the faculty's teaching performance over time?

a. Integrated Basic Learning Education (IBEd)

b. College

4. What predictive model can most accurately forecast faculty performance

evaluation rates based on historical data?

5. How proficient is the ElevatEd tool in predicting results and recommending

training needs in terms of:

a. Efficiency

b. Usability

Definition of Terms

The following terms are defined operationally for common understanding:

Algorithm. Refers to the sequential process used to predict faculty performance

based on historical evaluation data in NDMC.

Data Mining. Refers to the research approach used in developing the tool of this

study.

ElevatEd. The tool designed to analyze faculty performance data, predict future

performance, and provide recommendations for professional growth at NDMC.

Faculty Evaluation. The primary data source used to analyze and predict faculty

performance trends.

Notre Dame of Midsayap College (NDMC). The specific setting where the study is

conducted and where faculty evaluation data is collected.

Performance Categories. These categories are analyzed to differentiate performance

trends among faculty members.

Predictive Analytics. Used to forecast future faculty performance and inform

development strategies.

Professional Development Context in Research: The study aims to provide

recommendations for targeted professional development based on data analysis.

Reliability. To ensure that the tool provides consistent outcomes for faculty

evaluations.
Usability. A key metric to evaluate the practicality and user-friendliness of the

developed tool.
Review of Related Study

Recommender Tool

A study proposed MoodleRec, a hybrid recommender system integrated into the

Moodle Learning Management System. This tool helps educators find relevant learning

materials by searching across multiple standard-compliant repositories with keyword-

based queries. MoodleRec ranks Learning Objects by their relevance and quality, and

utilizes social features to illustrate how these resources have been employed in other

courses, thus simplifying the search process for teachers [8]. Another study developed the

Automated Recommendation Tool (ART) for synthetic biology. ART employs machine

learning and probabilistic modeling to provide recommendations for optimizing

biological production levels. By analyzing experimental data, ART generates predictive

models and suggests specific strains for future development to enhance production

outcomes [9].

The literature mentioned above presents different approaches to developing

recommendation tools. The studies demonstrated the use of machine learning and

probabilistic modeling in bioengineering, as seen in ART, and a hybrid recommender

system for educational resources, as implemented in MoodleRec. However, this study

focuses on developing a recommender tool specifically tailored for faculty development

in Notre Dame of Midsayap College.

Personalized Theory
 A study explored how personalized content and the social role of a voice-based

conversational agent can impact user attitudes toward products. It investigated whether

recommendations tailored to customer preferences and framed by the agent as a friend,

rather than a formal assistant, could enhance user engagement and satisfaction in a voice

shopping scenario. The findings indicated that both personalization and the agent's role

significantly influenced user attitudes, with effects varying based on the level of product

involvement [10]. Additionally, a systematic review of 72 studies on personalized

persuasive technologies examined how adapting interventions to individual user traits and

contexts can improve their effectiveness. This review emphasized the importance of

personalization in persuasive systems, particularly focusing on how personality traits and

other individual differences are used to tailor persuasive interventions across different

domains [11].

These studies showcase different uses of personalization theory. One study

focuses on enhancing user attitudes through personalized recommendations, while

another examines adapting persuasive technologies to individual differences. In contrast,

this study applies personalization theory to develop a recommender tool for faculty

development in Notre Dame of Midsayap College.

Reference:

[1] Z. Alamgir, H. Akram, S. Karim, and A. Wali, “Enhancing Student Performance

Prediction via Educational Data Mining on Academic Data,” Informatics in Education,

vol. 23, no. 1, pp 1, 2024.

[2] S. Eldridge, “Data Analysis,” Encyclopædia Britannica, Aug. 30, 2023. [Online].

Available: https://www.britannica.com/science/data-analysis. [Accessed: 01-Sep-2024]

[3] S. Bilal, S. Y. Guraya, and S. Chen, “The impact and effectiveness of faculty

development program in fostering the faculty’s knowledge, skills, and professional

competence: A systematic review and meta-analysis,” Saudi Journal of Biological

Sciences, vol. 26, no. 4, pp 688, 2019. doi: 10.1016/j.sjbs.2017.10.024.

[4] D. J. Martin, “Engagement: The Secret Sauce to Effective Faculty Professional

Development,” Faculty Focus. [Online]. Available:

https://www.facultyfocus.com/articles/academic-leadership/engagement-the-secret-sauce-

to-effective-faculty-professional-development/. [Accessed: 31-Aug-2024]

[5] Y. Zhou, “Development and Training Strategies of College Teachers Based on Data

Mining Technology,” Mobile Information Systems, vol. 2023, Article ID 7103391, 11

pages, 2023. doi: 10.1155/2023/7103391.

[6] R. O. D. Fuente and R. O. D. Fuente, “Faculty performance evaluation system with

application of data analytics,” South Asian Journal of Engineering and Technology, vol.

12, no. 1, pp. 24, 2022.

[7] A. J. P. Delima, et al., “The use of Schoology as learning management system in the

college of computing education: A response assessment using data mining techniques,”

International Journal of Advanced Trends in Computer Science and Engineering, vol. 9,

no. 3, pp. 3619-3621, 2020.

[8] C. De Medio, C. Limongelli, F. Sciarrone, and M. Temperini, “MoodleREC: A

recommendation system for creating courses using the moodle e-learning platform,”

Computers in Human Behavior, vol. 104, p. 106168, 2020.

[9] T. Radivojević, Z. Costello, K. Workman, and H. Garcia Martin, “A machine learning

Automated Recommendation Tool for synthetic biology,” Nature Communications, vol.

11, no. 1, p. 4879, 2020.

[10] C. E. Rhee and J. Choi, “Effects of personalization and social role in voice shopping:

An experimental study on product recommendation by a conversational voice agent,”

Computers in Human Behavior, vol. 109, p. 106359, 2020.

[11] . Alslaity, G. Chan, and R. Orji, “A panoramic view of personalization based on

individual differences in persuasive and behavior change interventions,” Frontiers in

Artificial Intelligence, vol. 6, p. 1125191, 2023.