Editorial: Seventh special issue on Knowledge Discovery and Business Intelligence

• Main Author: Cortez, Paulo
• Publication Date: 2023
• Other Authors: Bifet, Albert
• Format: Other
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: https://hdl.handle.net/1822/87710

[Excerpt] 1 Introduction: Currently, there is a growing interest in the field of Artificial Intelligence (AI) [Russell and Norvig, 2022]. Indeed, there are several successful AI applications that are impacting in diverse real-world domains, such as Industry 4.0 [Silva et al. 2021], Recommendation Systems (e.g., Amazon, Netflix) [Deepjyoti and Mala, 2022] and Virtual Assistants (e.g., Siri, Alexa, ChatGPT) [Grudin 2023]. AI encompasses sev eral approaches to provide machine intelligence, including: Expert Systems (ES) and Decision Support Systems (DSS) [Artnott and Pervan, 2014]; Machine Learn ing (ML) and Deep Learning (DL) [Alpaydin21]; and Metaheuristics or Modern Optimization [Cortez, 2021]. ES were a popular AI tool in the 1970s and 1980s, assuming an explicit knowledge representation (e.g., via symbolic rules) that was extracted from decision-makers and then processed by inference systems in order to support real-world tasks [Cortez et al., 2018]. After the 1990s, following the growth of big data and computational power, there was a shift towards data-driven AI [Darwiche, 2018]. This shift gave the rise to several data processing terms that often overlap and that aim to extract value from raw data, such as: Knowledge Discovery (KD) and Data Mining (DM) [Fayyad et al., 1996]; Business Intelligence (BI) and Business Analytics (BA) [Artnott and Pervan, 2014]; and Big Data and Data Science [Provost and Fawcett, 2013]. [...]