Gunita: Visualizing the evolution of the Philippines’ languages with a historical Philippine text corpora

Gunita: Visualizing the evolution of the Philippines’ languages with a historical Philippine text corpora Amiel Bornales Jonn Yuvallos De La Salle University-Manila De La Salle University-Manila 2401 Taft Ave, Manila 2401 Taft Ave, Manila Philippines Philippines amiel_bornales@dlsu.edu.ph jonn_yuvallos@dlsu.edu.ph Courtney Ngo De La Salle University-Manila 2401 Taft Ave, Manila Philippines courtney.ngo@dlsu.edu.ph Abstract While there are many culturomic studies in other countries, only a few studies focus on the culturomics unique to the Philippines. This study developed a Philippine news sources scraper and used a pre-existing Tagalog corpora containing books and poems across 100 years to build a continuously growing corpus. This study introduces Gunita, a web application that allows users to visualize how an n-gram is used over time and know which article, book, or poem the n- gram is used in to shed light on how Filipinos communicate through written text. 1 Introduction Culture is an ever-evolving aspect of society. It provides each person their own cultural identities and heritages, and is an essential area of human society that is ripe with research. This field of research is known for its cultural studies, and it aims to explore the connections between gender, race, class, etc. and their effects on culture (Barker, 2003). While cultural studies are able to draw powerful conclusions on human culture, these conclusions are based off of a collection of carefully chosen works that represent only a minority of the media available at the time (Michel et al., 2011). Analysis of cultural trends has always been hampered by the lack of suitable data, and so, to help further research along, a corpus containing 5,195,769 digitized books was created and analyzed (Michel et al., 2011). This led to the creation of the field of culturomics, which is a form of computational lexicology that studies human behavior and cultural trends through the quantitative analysis of digitized texts. 2 Related Works The analysis of language and words is an important aspect of culturomics, and large text corpora has been a necessity for language analysis for many of these studies. However, these researches differ in how they handle their data. For example, Michel et al. (2011) used a large corpus to investigate cultural trends between 1800 and 2000, such as insights into lexicography, the evolution of grammar, collective memory, the adoption of technology, the pursuit of fame, censorship, and historical epidemiology. Given that their scope is extremely broad, they focused more on the general state of things, and only go into specifics for important figures or events. For example, it was shown that Tiananmen (written in Chinese characters) was barely mentioned while the English term “Tiananmen” was mentioned increasingly in this period. This shows that the Chinese term of “Tiananmen” was not heavily used in China after the massacre due to censorship, while the English term of “Tiananmen” was mentioned frequently in Western media during the aftermath. This research will adopt the same focus and analyze the data based on news and historical events. According to Leetaru (2011), it is possible …

Content truncated to 500 words