Multi-Relational Data Mining Task

A Multi-Relational Data Mining Task is a data mining task whose input is a multi-relational database (that cannot make an iid assumption).

• AKA: Multi-Relational Mining.
• Context:
  • It can be solved by a Relational Mining System (that implements a relational mining algorithm such as inductive logic programming).
  • It can range from being an Explicit to being an Implicit.
• Example(s):
  • a Graph Mining Task.
  • a Social Network Analysis Task.
  • Graph Data Analysis Task/Graph Mining Task.
  • a Spatial Data Mining Task.
  • …
• Counter-Example(s):
  • IID Mining.
• See: Single-Table Learning Task, Complex Input Learning Task, Tabular Data Mining Task.

---

---

References

2011

• (Struyf & Blockeel, 2011) ⇒ Jan Struyf; Hendrik Blockeel. (2011). “Relational Learning.” In: (Sammut & Webb, 2011) p.851
  • QUOTE: Relational learning refers to learning in a context where there may be relationships between learning examples, or where these examples may have a complex internal structure (i.e., consist of multiple components and there may be relationships between these components). In other words, the “relational” may refer to both an internal or external relational structure describing the examples. In fact, there is no essential difference between these two cases, as it depends on the definition of an example whether relations are internal or external to it. Most methods, however, are clearly set in one of these two contexts. … Learning from Examples with External Relationships ...

2011

• (De Raedt, 2011c) ⇒ Luc De Raedt. (2011). “Multi-Relational Data Mining.” In: (Sammut & Webb, 2011) p.711
  • QUOTE: Multi-relational data mining is the subfield of knowledge discovery that is concerned with the mining of multiple tables or relations in a database. This allows it to cope with structured data in the form of complex data that cannot easily be represented using a single table, or an attribute as is common in machine learning.

    Relevant techniques of multi-relational data mining include those from relational learning, statistical relational learning, and inductive logic programming.

2008

• (Singh et al., 2008) ⇒ Ajit P. Singh, and Geoffrey J. Gordon. (2008). “Relational Learning via Collective Matrix Factorization.” In: Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2008). doi:10.1145/1401890.1401969
  • QUOTE: Relational learning is concerned with predicting unknown values of a relation, given a database of entities and observed relations among entities. An example of relational learning is movie rating prediction, where entities could include users, movies, genres, and actors.

2003

• (Džeroski, 2003) ⇒ Sašo Džeroski. (2003). “Multi-Relational Data Mining: an introduction.” In: ACM SIGKDD Explorations Newsletter, 5(1). doi:10.1145/959242.959245
  • ABSTRACT: Data mining algorithms look for patterns in data. While most existing data mining approaches look for patterns in a single data table, multi-relational data mining (MRDM) approaches look for patterns that involve multiple tables (relations) from a relational database. In recent years, the most common types of patterns and approaches considered in data mining have been extended to the multi-relational case and MRDM now encompasses multi-relational (MR) association rule discovery, MR decision trees and MR distance-based methods, among others. MRDM approaches have been successfully applied to a number of problems in a variety of areas, most notably in the area of bioinformatics. This article provides a brief introduction to MRDM, while the remainder of this special issue treats in detail advanced research topics at the frontiers of MRDM.

---