Desarrollo de un sistema de web scraping para recopilar información inmobiliaria del Marketplace de Mercado Libre en Ecuador

Abstract

The growth of digital real estate platforms has led to the publication of large volumes of dispersed and unstructured information, making its analysis and use for decision-making increasingly difficult. In this context, this technical project addresses the need to efficiently collect and organize real estate information published on Mercado Libre Ecuador. The main objective was to develop a web scraping system capable of extracting relevant data from real estate listings, storing it in a local database, and enabling its export to formats such as Excel and CSV. The applied methodology followed a sequential approach, using web scraping techniques implemented in Python and specialized libraries for data extraction and processing, along with an ETL-based architecture designed to transform the collected information into structured data. The web scraping system was developed in a local environment and subjected to testing and verification processes to evaluate its correct operation under common variations in the web environment. As a result, the system successfully automated the collection of real estate information from Mercado Libre Ecuador, organizing key data such as prices, location, and property characteristics in a structured manner. It is concluded that the developed solution improves access to relevant real estate information from this specific source and provides a useful basis for further analysis within the same context.