SLIQ: A Fast Scalable Classifier for Data Mining

By Manish Mehta, Rakesh Agrawal, Jorma Rissanen. 1996

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges (2021, methodology): “Full decision tree optimization is known to be an NP-complete problem (Laurent and Rivest, 1976), and heuristic greedy splitting and pruning procedures have been the major type of approach since the 1980s to grow decision trees (Breiman et al., 1984; Quinlan, 1993; Loh and Shih, 1997; Mehta et al., ”
A survey on concept drift adaptation (2014, methodology): “[Agrawal et al. 1992] used to be a popular data model for early work on scaling up decision tree learners [Agrawal et al. 1993; Mehta et al. 1996; Shafer et al. 1996; Gehrke et al. 1998].”
TEXT CLASSIFICATION AND CLASSIFIERS: A SURVEY (2012, methodology): “To handle this issue [24] presents method which can handle numeric and categorical data.”
Top-down induction of decision trees classifiers - a survey (2005, methodology): “[62] have proposed SLIQ an algorithm that does not require loading the entire dataset into the main memory, instead it uses secondary memory (disk) namely a certain instance is not necessarily resident in main memory all the time.”
Good Practice in Large-Scale Learning for Image Classification (2014, methodology): “To break the time-complexity of training to sublinear in the number of classes, various approaches have been proposed which employ tree structures [52], [4], [29], [14], [53], [30].”
Order preserving encryption for numeric data (2004, methodology): “(4)This procedure is reminiscent of the procedure for building decision tree classifiers, and in particular SLIQ [17], but the details are quite different.”
Pruning and summarizing the discovered associations (1999, methodology): “There are also a number of other methods in classification research for r ule pruning such as pessimistic error r ate [21] and minimum description length based pruning [ 15 ].”
Krimp: mining itemsets that compress (2011, methodology): “Mehta et al. (1996) use MDL to prune decision trees for classification.”
Mathematical Programming for Data Mining: Formulations and Challenges (1999, methodology): “In classi cation, examples include scalable decision tree algorithms [95, 33] and scalable approaches to computing classi cation surfaces [21, 102].”
Integrating Classification and Association Rule Mining (1998, background): “Standard classification systems need to load the entire database into the main memory (e.g., Quinlan 1992), although some work has been done on the scaling up of classification systems ( Mahta, Agrawal and Rissanen 1996 ).”
Data Mining for the Internet of Things: Literature Review and Challenges (2015, background): “SLIQ (Supervised Learning In Quest) is capable of handling large data sets with ease and lesser time complexity [19, 20], SPRINT (Scalable Parallelizable Induction of Decision Tree algorithm) is also fast and highly scalable, and there is no storage constraint on larger data sets in SPRINT [21].”
Building Decision Trees with Constraints (2003, methodology): “…a simple information theory driven scheme that selects a test that minimizes the information entropy of the partitions (we discuss entropy further in Section 3), while CART (Breiman et al., 1984), SLIQ (Mehta et al., 1996) and SPRINT (Shafer et al., 1996) select the test with the lowest GINI index.”
A Survey of Methods for Scaling Up Inductive Algorithms (1999, methodology): “Another notable optimization is the pre-sorting procedure used by the decision-tree learner SLIQ (Mehta et al., 1996).”