ID-Aaker: a dataset for Indonesian Aaker Brand Personality
Description
this data article introduces ID-Aaker, an Indonesian social media dataset annotated with Aaker’s five brand personality dimensions. The dataset comprises 75,756 rows of Indonesian-language texts collected from X (formerly Twitter) using a custom Python-based web scraping script implemented with Selenium WebDriver. Target accounts were selected based on the alignment of their posting characteristics with one of the five Aaker brand personality dimensions, guided by expert recommendation from a communication researcher. Each account was assigned to a single personality dimension prior to data collection, and all posts from that account were labelled accordingly. The dataset is distributed across five personality classes: competence (19,883 instances), sophistication (17,473 instances), excitement (15,270 instances), sincerity (11,909 instances), and ruggedness (11,221 instances). All data are stored in a single machine-readable CSV file containing the pre-processed text, engagement metrics (favorites, retweets, replies, quotes), word count, language code, and the assigned brand personality label. To protect user privacy, all usernames mentioned within the text were replaced with randomly generated pseudonyms (e.g., @user_4921), and all URLs were replaced with a uniform dummy link format. No personally identifiable information is retained in the published dataset