Editorial: Seventh special issue on Knowledge Discovery and Business Intelligence

Paulo Cortez, Albert Bifet

TL;DR

This editorial introduces the seventh special issue on Knowledge Discovery and Business Intelligence (KDBI), highlighting recent advancements in AI and its applications. It showcases four selected papers that present novel methods and address challenging real-world problems in areas like predictive maintenance, environmental monitoring, data integration, and competitive intelligence.

What Problem Does This Paper Address?

The paper addresses the growing interest and rapid advancements in Artificial Intelligence (AI), particularly in areas that extract value from data. While AI has seen widespread adoption in various domains (e.g., Industry 4.0, recommendation systems, virtual assistants), there's a continuous need for novel methods and systems to tackle complex, real-world challenges. This special issue aims to bridge the gap between theoretical advancements in Knowledge Discovery (KD) and Business Intelligence (BI) and their practical applications, by presenting cutting-edge research that pushes the boundaries of these fields.

Methods

This is an editorial piece, so it doesn't present novel methods itself. Instead, it introduces and summarizes the methods presented in the four selected papers within the special issue. These include:

  • Machine Learning for Predictive Maintenance: An empirical study using multi-label classification ML algorithms, including Logistic Regression and Extreme Gradient Boosting (XGBoost), to identify spare parts for digital device repair.
  • Clustering for Environmental Monitoring: A three-stage clustering algorithm employing a least-squares criterion to identify seasonal coastal upwelling patterns from sea surface temperature data.
  • Service-Oriented ETL Architecture: A proposed service-oriented approach for designing and developing Extract-Transform-Load (ETL) systems, breaking down processes into self-contained components.
  • Visual Analytics for Competitive Intelligence: A platform called MapIntel designed for acquiring and visually exploring competitive intelligence from large text collections, capable of handling complex natural language queries.

Datasets

Not discussed in the paper. The editorial mentions that the papers in the special issue utilize various datasets, including "millions of data records" for the spare parts consumption study and "sea surface temperature data" for the upwelling patterns analysis. However, specific dataset names or sources are not provided.

Key Insights and Ideas

  • The enduring relevance of AI: Despite the hype, AI continues to be a critical field with diverse and impactful real-world applications.
  • The synergy between KD and BI: The KDBI track and its associated special issues highlight the ongoing importance of integrating knowledge discovery techniques with business intelligence strategies for practical problem-solving.
  • Addressing complex real-world challenges: The selected papers demonstrate that cutting-edge research in KDBI is actively tackling significant issues in areas like industrial maintenance, environmental science, data management, and business strategy.
  • The evolution of AI approaches: The editorial notes the shift from rule-based expert systems to data-driven AI, emphasizing the growing importance of machine learning and deep learning.

Key Results

Not discussed in the paper. As an editorial, this piece focuses on introducing the content of the special issue rather than presenting specific quantitative or qualitative results from the papers themselves. It highlights that the four featured papers address "novel KDBI methods, systems and challenging real-world applications."

Limitations and Future Work

Not discussed in the paper. The editorial focuses on introducing the special issue and its contributions. It does not delve into the limitations of the presented work or suggest specific future research directions beyond the general advancement of AI and KDBI fields.

Summary updated

BibTeX
@article{cortez2023editorial,
  author = {Paulo Cortez and Albert Bifet},
  title = {{Editorial: Seventh special issue on Knowledge Discovery and Business Intelligence}},
  journal = {Expert Syst. J. Knowl. Eng.},
  year = {2023},
  doi = {10.1111/exsy.13466},
}
Figure