//

This activity challenged my understanding of predictive modeling, target variables, and machine learning methods. I was able to answer all the questions and explain my ideas in detail, which helped me practice organizing my thoughts and applying the concepts discussed in class. I also tried to justify my answers clearly by relating the attributes in the dataset to possible prediction models. However, while reviewing my answers, I realized that some of my choices were incorrect. I selected “Year Built” as the target attribute, but later understood that it was not the best choice for creating an impactful predictive model in the given scenario. A more appropriate target variable could have been something more directly connected to prediction outcomes, such as property value or another dependent factor. Because of this mistake, my explanations and reasoning became less accurate. I also chose K-Nearest Neighbors (KNN) as the model to use, but I later realized that my justification was weak because “Year Built” is a numeric variable. Linear Regression would have been a more suitable choice since it is commonly used for predicting continuous numerical values. Although KNN can still work with numeric data, it was not the best model for the situation compared to regression techniques. This made me realize that choosing the correct algorithm depends not only on the data type but also on the relationship between the variables and the prediction goal.

Taking this exam was both challenging and a learning experience for me. I think I did fairly okay in Test I because I was able to analyze the data, observe the relationship between Unit Price and Quantity Sold, and answer some of the questions about correlation and trends. I understood that as the price increased, the quantity sold decreased, which showed a negative relationship between the variables. I also tried my best to explain my observations clearly and organize my answers properly. However, I also made several mistakes during the exam. One thing I forgot to do was add labels to the graph, which is important because labels help make the scatterplot easier to understand and interpret. I realized that even small details like proper labeling matter in presenting data correctly.

In this exercise, I have learned that the study effectively balances technical execution with ethical responsibility. It highlights that while an algorithm like KNN can reach near-perfect accuracy on a standard dataset, its real-world application that is especially in government that requires more than just a high score; it requires human oversight to ensure transparency and fairness in automated decision-making.

Creating this IEEE paper about the k-Nearest Neighbor (k-NN) algorithm became both a learning experience and a realization of my weaknesses in technical writing and research presentation. While I was able to complete the paper and explain the general concepts of k-NN, I realized that I struggled in many aspects of making a proper IEEE-format research paper. One of the biggest challenges I faced was organizing my ideas academically. I understood the basic concepts of the Iris Dataset, machine learning, and k-NN classification, but translating those ideas into a formal and professional research structure was difficult for me. Some parts of the paper became repetitive, especially in the methodology section where I repeatedly explained the same concepts in different subsections. Because of this, the flow of the paper sometimes felt inconsistent and redundant instead of concise and technical as expected in IEEE writing. I also realized that my grammar, sentence construction, and technical wording were weak in several areas. Some explanations were unclear, overly long, or contained grammatical mistakes that reduced the professionalism of the paper. There were moments where I mixed informal explanations with technical discussions, making the writing less polished than a real academic journal article. I learned that IEEE papers require precision, clarity, and a more formal tone, which I still need to improve on significantly.

DID NOT SUBMIT

DID NOT SUBMIT

Creating my IEEE about Logistic Regression and construction safety prediction became a difficult but valuable learning experience for me. While I was able to create a working model and present performance metrics such as accuracy, recall, and confusion matrices, I realized that I struggled in properly presenting the report in an academic and professional manner. My paper was not fully IEEE compliant, and there were several issues with formatting, structure, grammar, and consistency throughout the report. Some explanations became repetitive, while other sections lacked deeper technical analysis and proper interpretation of the results. I also failed to include enough proper references and citations, which reduced the credibility and professionalism of the paper. Looking back, I understand that I focused too much on completing the output and obtaining good results instead of fully understanding the standards required in technical research writing. Another major weakness in my work was forgetting to add meaningful comments in the code. At the time, I only focused on making the program function correctly, but I now understand that code comments are important because they show proof of understanding and actual learning from the exercise. Without proper comments, my work lacked evidence that the explanations written in the report came from my real experience while building and testing the model. I also realized that I struggled in critically analyzing the behavior of the model since I mostly described the outputs instead of deeply explaining the reasoning behind the results and the limitations of the Logistic Regression model. Despite these weaknesses, this activity still helped me improve my understanding of preprocessing, feature selection, classification, and machine learning evaluation metrics. More importantly, it taught me that creating a proper research paper requires not only technical results, but also clear documentation, deeper analysis, proper formatting, and professional presentation.

Presenting our research about using Logistic Regression to evaluate SDG 4.7 integration in the Philippine education system became a challenging experience for our group because we realized that our understanding of the topic was not as deep as we initially thought. During the presentation, we struggled to clearly explain several parts of the study, especially when our professor asked questions about the methodology, the variables, and how the Logistic Regression model would actually work in practice. Although we were able to explain the general concept of SDG 4.7, sustainability education, and the importance of global citizenship education in the Philippines, we had difficulty defending the technical aspects of the research design. We also noticed that some of our explanations sounded theoretical rather than evidence-based because we relied heavily on concepts and references instead of actual implementation and experimentation