Constraint-Based Mining Task

A Constraint-Based Mining Task is a Pattern Mining Task that can find frequent patterns.

• AKA: Constraint-Based Pattern Recognition Task, Constraint-Based Pattern Learning Task.
• Context:
  • Input: a Dataset.
  • output: a Frequent Pattern.
  • Task Requirements:
    • a Constraint Model,
    • a Machine Learning Model.
  • It is a generalization of a Frequent Pattern Mining Task.
  • It can be by solve a Constraint-Based Mining System by implementing a Constraint-Based Mining Algorithm.
  • It can range from being a Monotonically Constraint-Based Mining Task to being a Anti-Monotonically Constraint-Based Mining Task.
  • It can range from being a Succinctly Constraint-Based Mining Task, to being a Convertible Constraint-Based Mining Task, to being a Boundable Constraint-Based Mining Task.
  • …
• Example(s):
  • a Breadth-First Constraint-Based Mining Task (De Raedt et al. 2002),
  • a Depth-First Constraint-Based Mining Task (Bucila et al. 2003; Kifer et al. 2003).
  • …
• Counter-Example(s):
  • an Emerging Pattern Recognition Task,
  • a Frequent Pattern Recognition Task,
  • a Relational Pattern Recognition Task.
• See: Pattern Recognition System, Knowledge Discovery, Correlation, Exploratory Data Analysis, Graph Pattern Mining Task, Structured Pattern Mining Task.

---

---

References

2017

• (Nijssen, 2017) ⇒ Siegfried Nijssen. (2017). "Constraint-Based Mining". In: (Sammut & Webb, 2017). DOI: 10.1007/978-1-4899-7687-1_164
  • QUOTE: Constraint-based mining is the research area studying the development of data mining algorithms that search through a pattern or model space restricted by constraints. The term is usually used to refer to algorithms that search for patterns only. The most well-known instance of constraint-based mining is the mining of frequent patterns. Constraints are needed in pattern mining algorithms to increase the efficiency of the search and to reduce the number of patterns that are presented to the user, thus making knowledge discovery more effective and useful.

---