A Survey of Utility-Oriented Pattern Mining

The main purpose of data mining and analytics is to find novel, potentially useful patterns that can be utilized in real-world applications derive beneficial knowledge. For identifying evaluating the usefulness different kinds patterns, many techniques constraints have been proposed, such as support, confidence, sequence order, utility parameters (e.g., weight, price, profit, quantity, satisfaction, etc.). In recent years, there has an increasing demand for utility-oriented pattern (UPM, or called mining). UPM a vital task, with numerous high-impact applications, including cross-marketing, e-commerce, finance, medical, biomedical applications. This survey aims provide general, comprehensive, structured overview state-of-the-art methods UPM. First, we introduce in-depth understanding UPM, concepts, examples, comparisons related concepts. A taxonomy most common approaches high-utility presented detail, Apriori-based, tree-based, projection-based, vertical-/horizontal-data-format-based, other hybrid approaches. comprehensive review advanced topics existing offered, discussion their pros cons. Finally, present several well-known open-source software packages We conclude our on open practical challenges this field.