The original pagination navigation is primarily based on page numbers, which is generally sufficient for everyday browsing. However, as the website’s content gradually grows and the number of pages increases, it becomes impractical to display every page number in the navigation bar. Relying solely on page numbers can therefore make navigation somewhat inconvenient.

For this reason, I decided to make a small modification to LoveIt’s pagination navigation. While keeping the original page-number navigation, I added Previous Page, Next Page, and page number jump functionality, making pagination more convenient and complete.

Cover image was generated by ChatGPT.

The cover image was generated by ChatGPT.

The year 2023 marked the rapid rise of generative artificial intelligence. From large language models to a wide range of AI-powered applications, artificial intelligence gradually became part of everyday life, making deep learning one of the most prominent and widely discussed technologies of the year.

For me, 2023 also marked the beginning of my journey into the world of deep learning, as I built and experimented with my first deep learning models.

The cover image is a downloaded typhoon satellite image. It is a satellite water vapor cloud image produced by the Central Weather Administration (CWA) using CCU/SSL. The image was captured at 07:10 on July 11, 2026, and was obtained from the Weather Image Archive of the Department of Atmospheric Sciences at Chinese Culture University (Atmospheric Science Research and Application Databank).

This is a small program that was originally written to download typhoon satellite images and related data. Its initial purpose was to collect image data for image recognition research. Since the program was developed at an earlier stage, it still contains several relatively immature implementation approaches. Therefore, this article serves only as a record and reference for the data download process, and does not provide an in-depth discussion of the program architecture or optimization methods. The programming language used in this project is Python.

The cover image was generated by Google Gemini.

In 2026, with the rapid development of AI, Vibe Coding has become a commonly used approach for assisting software development.

The cover image was generated by Google Gemini.

While taking the graduate course Simulation Study, the instructor mentioned that the random numbers used in modern computers and programming are mostly not truly random in the strict sense, but rather pseudo-random numbers. Pseudo-random numbers are generated through mathematical algorithms that compute the next value based on the previous state. Therefore, as long as the algorithm, initial value (Seed), and current state are known, the entire random number sequence can theoretically be reproduced. This raises an interesting question: if every pseudo-random number can be calculated from the previous value, where does the very first number come from?

The cover image was generated by ChatGPT.

While attending The 35th South Taiwan Statistics Conference, I first learned about control charts from an experienced professor who had spent many years working in manufacturing environments. Intrigued by the concept, I decided to spend some time organizing my notes and gaining a foundational understanding of what control charts are.

The Forbidden City, now home to the Palace Museum in Beijing, served as the imperial palace of the Ming and Qing dynasties. It is the largest and best-preserved ancient palace complex in the world, consisting of nearly 10,000 rooms and representing the pinnacle of traditional Chinese imperial architecture.

The cover photo was taken at the 2024 Joint Conference of University Administrators for Teacher Education, held on November 1, 2024.

The Joint Conference of University Administrators for Teacher Education is an important annual event for Taiwan’s teacher education system, where administrators from teacher-training universities across the country come together to discuss education policies, share experiences, exchange insights on policy implementation, and engage in face-to-face dialogue with officials from the Ministry of Education. The conference functions as a convergence point, gathering perspectives, expectations, and questions from all sides, gradually shaping the future direction of education. For me, it represents one of the most essential spaces for dialogue in the education sector, not only to review the current situation, but also to explore future strategies and confirm policy directions.

Time complexity is used to describe how the computation time of an algorithm grows as the input size increases during its execution. Because the processing capabilities, hardware architecture, and system environment differ across computer devices, comparing algorithms solely based on actual execution time often fails to yield representative and portable conclusions. Therefore, time complexity is usually measured by the “number of basic operations executed,” and asymptotic complexity is employed to analyze the efficiency of algorithms, establishing a hardware-independent and universally applicable performance evaluation method.

The cover image was generated by ChatGPT.

Programming has been incorporated into Taiwan’s 12-Year Basic Education Curriculum as an essential skill for future talents. Through coding, students can not only develop systems thinking, logical reasoning, and problem-solving abilities, but also cultivate aesthetics, digital literacy, and innovative adaptability during the creative process. The Hualien County Information Education Competition is designed with these goals in mind, aiming to stimulate students’ computational thinking and problem-solving potential by requiring them to design animations and games using the intuitive Scratch programming language.

The cover photo was taken on the way to the national competition with another team from our school on January 26, 2024.

Seeing my wholehearted dedication and enthusiasm for scientific inquiry in the Student Research Competition, the school invited me to join the guidance team for this year’s Youth Invention Exhibition. It was my first time serving as a mentor, and I felt a mix of excitement and nervousness. With a careful and attentive mindset, I led the students through brainstorming sessions, encouraging them to identify problems in daily life and design innovative solutions. From the initial conception to practical implementation, every step required careful thought and effort, but also brought a great sense of accomplishment.

The cover photo shows unprocessed Taiwan jade materials taken during an interview at the Rufeng Jade Workshop on September 7, 2023. The photograph was taken by the author.

“Positive thinking will let you do everything better than negative thinking will, because attitude is contagious.” This quote by American author Zig Ziglar has always reminded me that the way we engage often determines the final outcome. I also hope that my own positive attitude can bring constructive influence to the educational setting.

The cover image shows Zhixue Station and the surrounding parking facilities of Taiwan Railway Co., Ltd., captured from Google Maps on December 29, 2024.

The cover photo was taken by the author on December 25, 2025.

The Sports Day is the most ceremonial event in the school year. It is not only a gathering of track and field events, team competitions, and fun activities, but also a unifying ceremony for all students and staff. Shoufeng Junior High School holds its Sports Day biennially, combining it with the school anniversary and hosting a “Joint Sports Day with Surrounding Villages,” inviting local community members to participate, giving the event a unique community flavor.

This is a classic proof in mathematical statistics. That is, whether the variance of the sum of two random variables equals the sum of their variances.

Cover image was generated by ChatGPT.

Gradient descent is an iterative method used in most modern models to help find optimal parameters. However, in today’s era of big data, the massive amount of data results in significant time costs for gradient descent. One key goal is to compute parameters quickly without sacrificing accuracy.

Cover image generated by ChatGPT.

Why do we need sampling? What are the benefits for data analysis? When building models, we sometimes find that the amount of data required for training is too large, which prolongs the training time. The purpose of sampling is to select a representative finite sample from a large population and perform statistical analysis on it, thereby inferring the characteristics or properties of the overall population.

From this, we can see that a sample is a subset of the population. Sampling methods refer to how we reasonably select a subset of the population as a sample. Since the sample size is smaller than the population, we hope that the sample statistics calculated from the selected sample approximate the population statistics, minimizing the bias introduced by sampling.

The cover image was generated by ChatGPT.

Introduction

Logistic regression is a type of binary classification model derived from linear regression. Unlike linear regression, which fits a regression line that passes through the data points by minimizing the sum of squared distances, known as the least squares method. The goal of logistic regression is to find a decision boundary that clearly separates the data into two distinct classes.

Introduction

Recently, while reading articles, I noticed that it’s quite troublesome to find similar posts after finishing one. While researching, I also saw that many websites have sections like “Related Posts” or “You May Also Like,” which inspired me to create a “Related Posts” section on my personal site.

Since it’s about related posts, we need a way to determine which ones are similar, right? But re-tagging all posts or going through the trouble of using a classification model seems exhausting (or just plain silly?). Then I realized my articles already use tags and categories. So I thought: “Why not make use of what’s already there?” That’s why I decided to use existing metadata for classification.

Below is a guide on how to add a “Related Posts” section to each article.

Cover image generated by ChatGPT.

Introduction

How to calculate distance is an important aspect of modern statistical analysis. Choosing different distance metrics may lead to different results in statistical analysis methods. The following introduces commonly used distance metrics.

The cover image about Anscombe’s Quartet was generated by ChatGPT, and it used the following prompt: “The digital design highlights the title ‘Anscombe’s Quartet’ in bold, white sans-serif letters, centered against a dynamic abstract backdrop. The image is split into four colorful quadrants, each showcasing unique textures and patterns—ranging from painterly hues and curved lines to scattered circles and dots.”

Introduction

While listening to a presentation today, I happened to hear a term——Anscombe’s Quartet, a term I had never encountered before, yet it has significant implications in both statistics and data visualization.

Career exploration is an opportunity for junior high school students who are still feeling uncertain about their future to discover their interests.

By participating in activities from various career fields, students can develop their interests while also gaining an understanding of different industries, their job responsibilities, operations, and underlying principles. They also learn through hands-on experiences.

This time, the chemical engineering career exploration event was organized by National Hualien Industrial Vocational Senior High School, which held a weekend workshop on making photoengraved stamps. Through this activity, students not only learned how textbook knowledge applies to real life but also used their imagination to create their own unique stamps, experiencing the joy of hands-on work and creative expression.

The cover image is generated by ChatGPT, depicting data analysis. The prompt used is “A modern data analysis concept illustration featuring a futuristic workspace. The image includes multiple data charts, graphs, and dashboards displayed on transparent holographic screens. A diverse team of analysts and data scientists collaborate, analyzing trends and insights on large monitors. The scene has a sleek, high-tech atmosphere with glowing blue and purple hues, reflecting a professional and cutting-edge environment.” 。

Introduction

The dataset used in this analysis is the Swiss bank notes dataset, which is utilized by banks to distinguish between genuine and counterfeit old Swiss Francs. This study downloads the bank2.dat dataset from https://github.com/QuantLet/MVA/tree/master/QID-1530-MVAscabank56 and conducts the analysis using the R programming language.

Sometimes, a specific article may have flaws or contain information that we don’t want everyone to see, but we still want the article to be accessible to a select group of people. This is where article passwords come in handy!