Introduction to Data Mining

By Chet Langin. 2019

6,233 citations recorded by the source corpus.

How researchers see this paper

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

In Information Systems

Citations in the corpus

Local Rule-Based Explanations of Black Box Decision Systems (2018, methodology): “.19 .965±.15 .956±.17 4.5±2.7 2.3±2.3 1.8±0.9 neuclid .966±.17 .967±.15 .963±.19 4.3±2.6 2.2±2.1 1.8±1.0 Table 1: Comparison of distance measures. We measure the first three of them by the f1-measure [35]. Aggregated values of f1 and hit/c-hit are reported by averaging them over the the set of test ”
A Survey of Methods for Explaining Black Box Models (2018, methodology): “ction work on tabular data. Explanation of Neural Networks. The following papers describe the implementation of functions fwhich are able to interpret a black box bconsisting in a Neural Network (NN) [95] with a comprehensible global predictor c g consisting in a decision tree. In these works, the N”
Causal Interpretability for Machine Learning - Problems, Methods and Evaluation (2020, methodology): “ally exclusive (i.e., two or more rules might trigger by the same record), not exhausted (i.e., a record may not trigger any rules) and ordered (i.e., the rule set is ordered based on their priority) [95]. Linear Regression. Another common method known to be interpretable is Linear Regression. Linea”
Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks (2017, methodology): “Similarity (or distance) based methods are widely used in data mining and machine learning (Tan et al., 2006).”
KEEL: a software tool to assess evolutionary algorithms for data mining problems (2008, neutral): “− Learning Variety is support over main areas of DM, such as predictive tasks (classification, regression, anomaly/deviation detection), and descriptive tasks (clustering, association rule discovery, sequential pattern discovery) (Tan et al. 2006).”
Understanding and Enhancement of Internal Clustering Validation Measures (2013, background): “As the goal of clustering is to make objects within the same cluster similar and objects in different clusters distinct, internal validation measures are often based on the following two criteria [8], [14], [15].”
Mastering the Information Age - Solving Problems with Visual Analytics (2010, positive): “As we have seen, the objective of knowledge discovery and data mining is to extract information from large datasets[55, 108].”
Discovering important people and objects for egocentric video summarization (2012, neutral): “Since events can overlap, we retain (almost) disjoint events by eliminating those with greater than θevent overlap with events with higher silhouette-coefficients [26] in a greedy manner.”
Top 10 algorithms in data mining (2007, positive): “A more sophisticated approach, k-nearest neighbor (kNN) classification [23,75], finds a group of k objects in the training set that are closest to the test object, and bases the assignment of a label on the predominance of a particular class in this neighborhood.”
Human Focused Video Description (2011, methodology): “It was a similarity measure related to the Jaccard index [19].”
ForEx++: A New Framework for Knowledge Discovery from Decision Forests (2017, background): “Interestingly, an ensemble of classifiers is found to be effective for unstable classifiers such as decision trees (Tan et al. (2005)).”
Spatio-Temporal Data Mining (2017, background): “A unique quality of ST data that differentiates it from other data studied in classical data mining literature (e.g., see [Tan et al. 2017]) is the presence of dependencies among measurements induced ∗Both authors contributed equally to the paper.”
Online Machine Learning in Big Data Streams (2018, methodology): “idering the conditional probability tables and the class distribution as priors, the predicted class of an instance will be the one that maximizes the posterior probability computed by the Bayes rule [187]. The simplest, Naive Bayes model makes the “naive” assumption that each input variable is cond”
Outlier Detection Techniques for Wireless Sensor Networks: A Survey (2008, neutral): “Noisy data as well as erroneous data should be eliminated or corrected if possible as noise is a random error without any real significance that dramatically affects the data analysis [10].”
Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning (2018, background): “It is said that LOOCV "[LOOCV has] high variance because the test set only contains one sample" [Tan et al., 2005] and "[LOOCV] is highly variable, since it is based upon a single observation (x1, y1)" [James et al., 2013].”
Learning the easy things first: Self-paced visual category discovery (2011, methodology): “Evaluation metrics: To quantify discovery accuracy, we use purity [21], which is the percentage of correctly labeled instances, where all instances in a cluster are assigned to its majority class’s true label.”
Factorizing YAGO: scalable machine learning for linked data (2012, methodology): “To evaluate the quality of our clustering, we followed the procedure suggested in [30] and assign that F-measure score to a particular concept that is the highest for this concept out of all clusters.”
Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining: Experimental analysis of power (2010, neutral): “This question has given rise to the growing interest in the analysis of experiments in the field of computational intelligence (CI) [15] or the field of data mining (DM) [24] [45].”