Feature Extraction, Construction and Selection: A Data Mining Perspective

By Huan Liu, Hiroshi Motoda. 1998

Citations in the corpus

• A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, neutral): “ut it is known that variables that separately do not provide useful information may do so when they are used together. For this reason other alternatives, primarily feature construction or extraction [18], emerged. In addition to these two denominations, feature selection and feature extraction, whe”
• Distributional Word Clusters vs. Words for Text Categorization (2003, unknown): “, Koller and Sahami, 1996). Many of the greedy methods are supervised and rank the features according to their contribution to separation between categories, and then extract the most “useful” features. Yang and Pedersen (1997) empirically compare five filter-based (supervised and unsupervised) gree”
• Binary Optimization Using Hybrid Grey Wolf Optimization for Feature Selection (2019, unknown): “Metaheuristics can be considered as ‘‘directed’’ random algorithms [8], [10]; they find an acceptable solution but does not guarantee determining the optimal solution in each run [15].”
• On Similarity Preserving Feature Selection (2013, unknown): “In addition, feature selection algorithms may return either a subset of features [17], [46] or the weights of all features measuring their utility [1], [35].”
• Many-Layered Learning (2002, unknown): “This is an important step toward deep networks of building blocks, and away from limitations imposed by employing only gradient-descent driven by output error.”
• A Survey of Modern Questions and Challenges in Feature Extraction (2015, unknown): “The creation of new features from input data is usually referred to as feature construction (Liu and Motoda, 1998; Guyon et al., 2008).”
• Hashing with binary autoencoders (2015, unknown): “This procedure can be seen as a “filter” approach (Kohavi and John, 1998) and is suboptimal: in the example, the thresholded PCA projection is not necessarily the best thresholded linear projection (i.e., the one that minimizes the objective function under all thresholded linear projections).”
• Machine Learning: Algorithms, Real-World Applications and Research Directions (2021, unknown): “Thus, “feature selection” [66] [99] is considered as one of”
• Hierarchical Text Categorization Using Neural Networks (2004, unknown): “Reduction of the set of features to train the neural networks is necessary because the performance of the network and the cost of classification are sensitive to the size and quality of the input features used to train the network (Yang and Honovar 1998).”
• Decision tree classifier for network intrusion detection with GA-based feature selection (2005, unknown): “Our proposed GA-based Feature Selection algorithm is based on the wrapper model [12][13] as discussed in section 2.”
• The Decoding Toolbox (TDT): a versatile software package for multivariate analyses of functional imaging data (2015, unknown): “The term feature transformation is not commonly used in the brain decoding community, but it is known in the machine learning community as a set of methods that change the meaning of each individual feature (Liu and Motoda, 1998).”
• A discrete particle swarm optimization method for feature selection in binary classification problems (2010, unknown): “Sequential Forward Selection, SFS (Whitney, 1971) and Sequential Backward Selection, SBS (Marill and Green, 1963) are the two most common greedy methods. To avoid the nesting effect, Stearns (1976) proposes ‘‘plus-ltake-away-r” method in which SFS is applied l times forward and then r back-tracking ”
• Feature Selection for Classification of Hyperspectral Data by SVM (2010, unknown): “useful information to separate the classes with highly correlated and redundant features excluded from the classification analysis [25].”
• Characterization of complex networks: A survey of measurements (2005, unknown): “[ 85 ]). The former family of approaches consider some measurement of the separability of the classes (e.g.”
• Consistency-based search in feature selection (2003, unknown): “Reviews and books on feature selection can be found in [11] [26] [27].”