Resource Description Framework Generation for Tropical Disease Using Web Scraping

Tropical diseases are diseases that commonly occur in tropical areas like Indonesia. The most frequent diseases plaguing Indonesia community consists of common diseases that are categorized as tropical diseases, e.g., malaria, leprosy, and lymphatic filariasis. As first aid, Indonesian people usually rely on a search engine to search for information about diseases and medication especially of general illnesses like coughs, colds, and fever. However, some result of conventional search engine still generates un-related articles. Another drawback of a conventional search engine, sometimes it cannot handle synonym meaning of disease in searching especially for the non-medical term in Bahasa Indonesia. For example, for lymphatic filariasis is better known as kaki gajah (In English: Elephant foot). One solution is to build a search engine based on semantic web technology. The semantic web can show the relationships between the data, including synonym of disease name in Bahasa Indonesia in Resource Description Framework (RDF) serialization form. Unfortunately, based on our initial survey, the websites that provide health information in Bahasa Indonesia do not provide RDF files. This research aims to generate an RDF serialization that describes the relationship ontology in tropical diseases and treatments terms. We implemented a web scraping method to harvesting and extracting vocabularies from some popular health websites, creating the relationship between terms with ontology and convert this information into an RDF serialization.

I. INTRODUCTION

Many people in Indonesia rely on the internet search engine to find out about a disease and its medicine as first aid. However, some result of this searching information still contains un-related articles, and the search engine sometimes cannot handle synonym meaning of disease in searching especially for the non-medical term in Bahasa Indonesia. For example, for lymphatic filariasis is better known as kaki gajah (In English: Elephant foot), Rubeola is better known as campak or kerumut. The drawbacks of this conventional search engine because it uses keywords to generate the result. This search engine generates information that depends on the keywords without knowing the meaning of the terms. Most of the information circulating on the internet are only provided for human consumable and not yet available for machine-readable. One solution is to build search engine based on semantic web technology, where semantic web can identify information or meaning of the data, such as disease, drug classes, the synonym name of the disease both in a medical term or in the term that better known in Bahasa Indonesia, e.g., "nyctalopia" or "rabun ayam" (in English: myopic chicken).