Database of gastronomic establishments in Barcelona (clean)

Below is a dictionary that clearly distinguishes between facts (quantitative variables to be analyzed) and dimensions (variables that allow organizing, segmenting, or contextualizing the data). Facts price_min: Quantitative variable derived from the price_range variable of the original dataset. It indicates the approximate minimum price of the restaurant. price_max: Quantitative variable derived from the price_range variable of the original dataset. It indicates the approximate maximum price of the restaurant. score: Continuous quantitative variable representing the average rating obtained from consumer reviews, with values ranging from 0 to 10. opinions_count: Discrete quantitative variable indicating the total number of consumer reviews. days_open_count: Discrete quantitative variable indicating the number of days per week the restaurant is open. It is derived from the time_{day}_{period} variables in the original dataset. Dimensions price_category: Ordinal categorical variable calculated from the price_range variable in the original dataset. It indicates the price range based on: [1] Budget (€0–25) [2] Moderate (€25–35) [3] Expensive (€35–50) [4] Luxury (>€50) zone1, zone2, zone3: Categorical variables indicating the area or neighborhood where the restaurant is located, derived from the zone variable in the original dataset. After cleaning, the dataset includes a total of 22 different zones. It is split into multiple fields because a restaurant may belong to multiple adjacent zones. latitude and longitude: Continuous quantitative variables derived from address data, enabling precise geolocation of establishments. food: Categorical variable indicating the type of cuisine (homemade, pizzeria, rice dishes, etc.). It has high cardinality (99 categories after cleaning) and may require further transformation (e.g., one-hot encoding). ambient: Categorical variable describing the atmosphere or style of the venue (romantic, traditional, iconic, etc.), with a total of 17 categories. target: Categorical variable indicating the main target audience of the restaurant, with values such as families, young people, groups, or business. open_lunch_count: Binary categorical variable indicating whether the restaurant offers lunch service. open_dinner_count: Binary categorical variable indicating whether the restaurant offers dinner service. open_weekend_count: Binary categorical variable indicating whether the restaurant is open on weekends. Other supporting variables (high cardinality) These variables can be useful for displaying detailed or contextual information, but they are not particularly suitable as main metrics or global filters: name: Name of the establishment. description: Brief description of the restaurant. phone: Contact phone number. address, city, postal_code: Location information. food_sample: Categorical variable with examples of typical dishes from the restaurant. menu: Categorical variable with types of available menus (126 different values), which could be further leveraged after a cleaning process. keywords: Categorical variable with descriptive keywords (279 values), requiring transformation for analytical use. comments: Free-text user comments. It has high semantic value but requires more advanced techniques (e.g., text analysis) for exploitation. Finally, it is worth noting that the original dataset includes a detailed set of variables related to service hours by day of the week and time slot (time_{day}morning, time{day}_evening). Some of the variables presented have been derived from these, as they were considered the most relevant to answer the posed questions. However, additional variables could be derived, such as the total number of weekly opening hours, to further enrich the analysis.