Facebook posts for analyzing Content Strategies for Digital Consumer Engagement: a curated dataset

This database contains public data from publications made on Facebook by small and medium companies in the tourism sector operating in the Amazon region in  Brazil. The collection was carried out from January to June 2018 using the public API provided by the platform.

These data were processed and classified by different evaluators according to the content categories proposed by Gavilanes. Thus, there is a column in the dataset called "category", which contains the classification of each publication, with each number is associated with a category, as described below:

“0” - No category

“1” - New product announcement

“2” - Sweepstakes and contest

“3” - Sales

“4” - Consumer Feedback

“5” - Infotainment

“6” - Organization Branding

“N\A” - Non-agreement between evaluators

Therefore, the columns with data present in the CSV file are:

“status_id” - Identification of each publication in the social network. String Textual content of each publication;

“status_message” - The textual content of each publication;

“link_name” - Which part of the profile the post are related;

“status_type” - The type of the publication is a ’photo’, ’video’, ’status’ and/or ’link’;

“status_link” - URL to the publications on the platform;

“status_published” - Publication date in the social network;

“num_comments” - Numbers of comments made by users;

Reactions - Numbers of emoticons reactions from users on post – these numbers are  presents in the columns: "num_reactions", "num_shares", "num_likes", "num_loves", "num_wows", "num_hahas", "num_sads", "num_angrys", "num_special";

“category” - The category of content that we mentioned before.