Google Trends data for 520 business-related search terms in Bahasa Indonesia

The dataset used in this research consists of weekly Google Trends search volume data for 520 business-related keywords in Bahasa Indonesia, collected from January 2020 to December 2024 across 34 Indonesian provinces. Each row in the dataset represents a specific search term, location, and week, with the main variable being the normalized search volume index (ranging from 0 to 100). Additional columns include the name of the province, the keyword or phrase, and its classification into one of eight business domains: retail and e-commerce, financial services, technology and digital services, healthcare and wellness, education and training, food and beverage, transportation and logistics, or entertainment and media. To enhance the predictive power of the model, the dataset includes various engineered features such as lagged search volumes (from 1 to 8 weeks), moving averages over 4, 8, and 12-week windows, seasonal decomposition components, and regional aggregation metrics that reflect the intensity of search activity within each province. A binary flag is also added to indicate whether a particular week contains a major holiday (e.g., Ramadan or Christmas). These features are designed to capture temporal dynamics, seasonal patterns, and geographic variations in search behavior, enabling the forecasting model to effectively predict future business trend movements.