Open AccessBook Chapter10.1007/978-3-319-50349-3_10

Feature Selection Using Tabu Search with Learning Memory: Learning Tabu Search

- 29 May 2016

- pp 141-156

TL;DR: This paper proposes to solve the feature selection problem in classification with a tabu search algorithm integrating a learning mechanism, originally designed for a railway network problem in which the evaluation of a solution is time-consuming.

Abstract: Feature selection in classification can be modeled as a combinatorial optimization problem. One of the main particularities of this problem is the large amount of time that may be needed to evaluate the quality of a subset of features. In this paper, we propose to solve this problem with a tabu search algorithm integrating a learning mechanism. To do so, we adapt to the feature selection problem, a learning tabu search algorithm originally designed for a railway network problem in which the evaluation of a solution is time-consuming. Experiments are conducted and show the benefit of using a learning mechanism to solve hard instances of the literature.

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What have the authors contributed in "Feature selection using tabu search with learning memory: learning tabu search" ?

In this paper, the authors propose to solve this problem with a tabu search algorithm integrating a learning mechanism.

2. What have the authors stated for future works in "Feature selection using tabu search with learning memory: learning tabu search" ?

This work is very encouraging and perspectives of future works are interesting.. As far as the Feature Selection problem is concerned, such perspectives may deal either with the extension of the learning mechanism ( other definition of the trail, for example ) for LTS, or with the integration of the proposed learning mechanism in other metaheuristics that may benefit from the estimation function.

Table 2. Instances description. The total number of features (# Features), the size of the training |T | and validation |V | sets (i.e., number of observations), the runtime (in seconds) needed by SVM to build and evaluate a model on each training set (without feature selection), and the runtime (in seconds) allocated to each optimization algorithm are given. Instances are divided into two groups according to the SVM runtime.

Table 3. Learning Tabu Search parameters. Gives each parameter together with its setting value.

Table 5. Average and standard deviation (in brackets) of Accuracy values obtained on both training and validation sets for HC, TS and LTS. Accuracy values in bold stand for algorithms outperforming the other one(s) according to the Wilcoxon test. The statistical comparison between algorithms is given. The double line shows the separation between low and high evaluation time cost.

Table 1. Metaheuristics for the feature selection problem. The bibliographic reference, the date and the resolution approach are also given.

Table 4. Average and standard deviation (in brackets) of Fitness, Accuracy and # S Features values obtained on training sets for HC, TS and LTS. For each algorithm, the fitness values have been computed from the Accuracy and # S Features, the number of selected features (see Section 3.1). Fitness values in bold stand for algorithms outperforming the other one(s) according to the Wilcoxon test. For each instance, the value of the accuracy obtained by SVM without any feature selection is pointed out in brackets. The statistical comparison between algorithms is given under the instance name.

Citations

Journal Article•10.1007/S10916-017-0703-X

Diagnosis of Chronic Kidney Disease Based on Support Vector Machine by Feature Selection Methods

Huseyin Polat, +2 more

- 01 Apr 2017

- Journal of Medical Systems

TL;DR: The results showed that the Support Vector Machine classifier by using filtered subset evaluator with the Best First search engine feature selection method has higher accuracy rate in the diagnosis of Chronic Kidney Disease compared to other selected methods.

...read moreread less

260

Journal Article•10.1016/J.ESWA.2020.113176

A novel wrapper feature selection algorithm based on iterated greedy metaheuristic for sentiment classification

Osman Gokalp, +2 more

- 15 May 2020

- Expert Systems With Applications

TL;DR: A novel wrapper feature selection algorithm based on Iterated Greedy metaheuristic for sentiment classification is proposed and a selection procedure that is based on pre-calculated filter scores for the greedy construction part of the IG algorithm is developed.

...read moreread less

113

Journal Article•10.1016/J.DAM.2018.03.019

Learning Variable Neighborhood Search for a scheduling problem with time windows and rejections

Simon Thevenin, +1 more

- 31 May 2019

- Discrete Applied Mathematics

TL;DR: This work enhances variable neighborhood search with a learning mechanism which helps to drive the search toward promising areas of the search space and significantly outperforms state-of-the-art algorithms for the considered problem.

...read moreread less

•Journal Article•10.1007/s10462-023-10472-w

A systematic review of social network sentiment analysis with comparative study of ensemble-based techniques

Dimple Tiwari, +4 more

- 12 Apr 2023

- Artificial Intelligence Review

TL;DR: In this paper , the authors present a systematic survey that organizes and describes the current scenario of sentiment analysis and provides a structured overview of proposed approaches from traditional to advance, and discuss the SA-related challenges, feature engineering techniques, benchmark datasets, popular publication platforms, and best algorithms to advance the automatic SA.

...read moreread less

Journal Article•10.1007/S12205-017-2000-0

Improved Fruit Fly Optimization Algorithm Incorporating Tabu Search for Optimizing the Selection of Elements in Trusses

Yancang Li, +1 more

- 29 Oct 2018

- Ksce Journal of Civil Engineering

TL;DR: Optimization results and comparison with other algorithms show that the stability of improved fruit fly Optimization Algorithm is apparently improved and the efficiency is obviously remarkable.

...read moreread less

References

•Journal Article•10.1016/S0004-3702(97)00043-X

Wrappers for feature subset selection

Ron Kohavi, +1 more

- 01 Dec 1997

- Artificial Intelligence

TL;DR: The wrapper method searches for an optimal feature subset tailored to a particular algorithm and a domain and compares the wrapper approach to induction without feature subset selection and to Relief, a filter approach tofeature subset selection.

...read moreread less

9.6K

Journal Article•10.1109/5254.708428

Support vector machines

Marti A. Hearst, +4 more

- 01 Jul 1998

- IEEE Intelligent Systems & Their Applica...

TL;DR: This issue's collection of essays should help familiarize readers with this interesting new racehorse in the Machine Learning stable, and give a practical guide and a new technique for implementing the algorithm efficiently.

...read moreread less

6.1K

•Book

Support Vector Machines

Ingo Steinwart, +1 more

- 12 Aug 2008

TL;DR: This book explains the principles that make support vector machines (SVMs) a successful modelling and prediction tool for a variety of applications and provides a unique in-depth treatment of both fundamental and recent material on SVMs that so far has been scattered in the literature.

...read moreread less

6.1K

Reference Book•10.1007/978-0-387-30164-8

Encyclopedia of Machine Learning

Claude Sammut, +1 more

- 28 Mar 2011

TL;DR: The style of the entries in the Encyclopedia of Machine Learning is expository and tutorial, making the book a practical resource for machine learning experts, as well as professionals in other fields who need to access this vital information but may not have the time to work their way through an entire text on their topic of interest.

...read moreread less

3.5K

•Book

Feature extraction : foundations and applications

Isabelle Guyon, +3 more

- 01 Jan 2006

TL;DR: This book discusses Feature Extraction for Classification of Proteomic Mass Spectra, Sequence Motifs: Highly Predictive Features of Protein Function, and Combining a Filter Method with SVMs.

...read moreread less

1.9K

...

Expand

Feature Selection Using Tabu Search with Learning Memory: Learning Tabu Search

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What have the authors contributed in "Feature selection using tabu search with learning memory: learning tabu search" ?

2. What have the authors stated for future works in "Feature selection using tabu search with learning memory: learning tabu search" ?

Figures

Citations

Diagnosis of Chronic Kidney Disease Based on Support Vector Machine by Feature Selection Methods

A novel wrapper feature selection algorithm based on iterated greedy metaheuristic for sentiment classification

Learning Variable Neighborhood Search for a scheduling problem with time windows and rejections

A systematic review of social network sentiment analysis with comparative study of ensemble-based techniques

Improved Fruit Fly Optimization Algorithm Incorporating Tabu Search for Optimizing the Selection of Elements in Trusses

References

Wrappers for feature subset selection

Support vector machines

Support Vector Machines

Encyclopedia of Machine Learning

Feature extraction : foundations and applications

Related Papers (5)

Adaptive tabu tenure computation in local search

Human-guided tabu search

A Tabu-search Based Bayesian Network Structure Learning Algorithm

Intelligent Neighborhood Exploration in Local Search Heuristics

Writing Local Search Algorithms Using Easylocal