Expert Systems (ES) are a core element of human decision making. Initially, in the 1970s and 1980s, ES were focused on extracting explicit knowledge from human experts. With the availability of big data, after the 2000s, ES incorporated data-driven models, thus being associated with business intelligence, big data, data science, and machine learning systems (Cortez & Santos, 2017). The importance of data-driven models in the ES area is confirmed by the recent Wiley’s Expert Systems (EXSY) literature survey that analyzed all journal research articles published from 2000 to 2016 (Cortez et al., 2018). The survey revealed data-driven as the most prevalent ES method type, corresponding to around 35% of all recently published EXSY papers. The first “Knowledge Discovery and Business Intelligence” (KDBI) track was held at the EPIA conference onArtificial Intelligence in 2009, with the goal of strengthening the interaction between Knowledge Discovery (KD) and Business Intelligence (BI). Both are important data related topics. KD is the subfield of Artificial Intelligence that is focused on the extraction of human interesting knowledge from raw data (Fayyad et al., 1996). BI is an umbrella term that encompasses methodologies and technologies (e.g., data warehousing and dashboards) to support managerial decision making (Delen et al., 2014). Following the success of its first 2009 edition, the KDBI workshop become a regular track of the biannual EPIA conference, with its fifth edition taking place at the 18th EPIA conference, held at Porto, Portugal, in September 2017. Since 2011, the track had a dedicated special EXSY issue (Cortez & Santos, 2013, Cortez & Santos, 2015, Cortez & Santos, 2017). This is the fourth special issue on Knowledge Discovery and Business Intelligence, and it includes extended versions of the best papers presented at the 5th KDBI thematic track of EPIA 2017, which received 20 paper submissions. The EXSY special issue included two rounds of reviews, which involved reviewers from the 5th KDBI track of EPIA2017 and also the EXSY journal. After the revision stage, a total of five papers were accepted to be published in this issue, which corresponds to an acceptance rate of 25%. While in the last decade there have been remarkable developments in the KD and BI areas, there are still challenges and opportunities. For instance, most KDBI research has been focused on processing structured data but little has been devoted to Mining Software Repositories (MSR), which is quite valuable for the software engineering industry. Also, feature selection is a challenging task in the context of large genomics datasets. Some domain applications, such as food truck recommendation, produce difficult multi-label classification tasks. Another challenge comes from the development of online recommendation systems, which are relevant for high velocity data and big data streams. Moreover, imbalanced domains are important in several real-world applications, such as medicine and finance, often prejudicing the predictive performance of the data-driven models. These challenges are approached in the five accepted papers published here. In the next section, we summarize the main contributions of these papers.