Environmental Setup The experiment was conducted in a flexible environment, utilizing the following configurations on a Kaggle notebook. - Baseline Configuration: Intel(R) Xeon(R) CPU @ 2.20 GHz with 4 cores 30 Gigabytes of RAM. - TPU Accelerated Configuration (TPU VM v3-8): Intel(R) Xeon(R) CPU @ 2.00 GHz with 96 cores 330 Gigabytes of RAM. Experimental Dataset This experiment utilizes the ThaiSum dataset, a large-scale collection of Thai text summarization pairs. These articles are all sourced from reputable online news sources, namely Thairath, Thai PBS, Prachatai, and The Standard. This dataset encompasses 358,868 articles and their journalist-written summaries. The Use of Open-source Libraries - PyThaiNLP for Thai natural language processing. - Modin and Pandas for data analysis and manipulation. - NumPy for numerical computing. - scikit-learn for constructing a TF-IDF matrix and similarity matrix. - NetworkX for graph formation and TextRank algorithm. - rouge provides metrics that compare the generated summaries to human-written references. - seaborn and Matplotlib for data visualization