A review of contrast pattern based data mining

Abstract

Contrast pattern based data mining is concerned with the mining of patterns and models that contrast two or more datasets. Contrast patterns can describe similarities or differences between the datasets. They represent strong contrast knowledge and have been shown to be very successful for constructing accurate and robust clusters and classifiers. The increasing use of contrast pattern data mining has initiated a great deal of research and development attempts in the field of data mining. A comprehensive revision on the existing contrast pattern based data mining research is given in this paper. They are generally categorized into background and representation, definitions and mining algorithms, contrast pattern based classification, clustering, and other applications, the research trends in future. The primary of this paper is to server as a glossary for interested researchers to have an overall picture on the current contrast based data mining development and identify their potential research direction to future investigation.

Emergent pattern detection algorithm for big data streams

“…Several approaches have been proposed to detect emerging patterns, such as approaches based on the "growth rate" measure, which is represented by the ratio of the support of the pattern at two different timestamps. 20,21 If this rate exceeds a predefined threshold, the pattern is considered as emergent, as follows: Definition 2.5. The growth rate of P between timestamps t and t is denoted: growthRate t ,t (P ) = supp t (P ) / supp t (P ).…”

Finding Contrast Patterns in Imbalanced Classification based on Sliding Window

“…Compared to the traditional classification, the classification accuracy of the contrast patterns (Shiwei Zhu et al, 2015) classifier is better than that of the traditional classifier because that the contrast pattern has a strong discriminative capability. At present, there are many contrast patterns mining algorithm.…”

Concepts encoding via knowledge-guided self-attention networks

“…Due to the increasing demand and applications in broad domains, such as automatic target detection 1 , burned area mapping 2 , recommendation system 3 and healthcare, data mining plays increasingly important roles, highlighted in various communities. Thus many deep learning models 4 have been proposed for mining the digital data. As the field of healthcare is closely related to people's life, numerous efforts have been made to predictive analytics based on the representation of medical concepts, especially for medication recommendation(MR) 5 which aims at predicting medication combinations for patients according to their visit history.…”