FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2986157773> ?p ?o ?g. }

• W2986157773 abstract "Three major factors that determine the performance of a machine learning are the choice of arepresentative set of features, choosing a suitable machine learning algorithm and the rightselection of the training parameters for a specified machine learning algorithm. This thesistackles the problem of feature selection for supervised machine learning prediction tasksthrough dependency information. The feature evaluation strategy is formulated based onmutual information (MI) to handles both classification and regression supervised learningtasks and the search strategy is a modified greedy forward strategy designed to manageredundancy between features and avoiding features that are irrelevant to the predicting output.The problem with many existing feature selections that evaluate features based on mutualinformation is that they are designed to handles classification tasks only. And the few existingones that can work for regression tasks were recently found to underestimate mutualinformation between two strongly dependent variables. In addition to these problems, thesearch strategy which is usually a heuristic greedy method used with many existing featureselections, lacks scientifically sound stopping criterion and the forward greedy proceduredespite its advantages over the backward procedure is found to reveal suboptimal. Thus, thisthesis has developed and evaluated a filter based Information Theoretic-based FeatureSelection (IFS) for machine learning. Various experiments were carried out to assess and testcomponents of IFS algorithm. The first test was designed to evaluate the formulated IFSSelection Criterion Strategy (MI estimator) by comparing it with six different MI estimatorbenchmarks. The second test evaluates IFS in a controlled study using simulated datasets.Moreover, the third test used ten natural domain datasets obtained from UCI Repository, inabout fifteen different experiments, using three to four different Machine Learning Algorithms for performance evaluation. Also, additional experiments to compare the relativeperformance of the IFS with five related feature selection algorithms were carried out usingnatural domain datasets. Besides, this thesis developed a hybrid filter method to enhance theperformance of the IFS. IFS served as filter together with an Ant Colony OptimizationSystem (ACO) as a metaheuristic form the hybrid system. In these extended IFS method,feature selection method was defined and presented as a 0-1 Knapsack Problem (MKP). Thus,this thesis precisely developed and evaluated IFS_BACS (Binary Ant Colony System) hybridmethod. Further experiments were carried out using the natural domain datasets andcomparison were made between IFS and hybrid IFS_BACS methods. In most of the cases,experimental results of IFS and its extended IFS_BACS hybrid method significantly reducedfeatures and produce competitive performance accuracy when compared to the results of thefull feature set before applying the IFS or IFS_BACS method. And comparing the IFS with itsextended version, the extended version (IFS_BACS) seems to be more promising in selectingoptimal feature subset from large datasets." @default.
• W2986157773 created "2019-11-22" @default.
• W2986157773 creator A5091636164 @default.
• W2986157773 date "2018-01-01" @default.
• W2986157773 modified "2023-09-28" @default.
• W2986157773 title "Information Theoretic-based Feature Selection for Machine Learning" @default.
• W2986157773 hasPublicationYear "2018" @default.
• W2986157773 type Work @default.
• W2986157773 sameAs 2986157773 @default.
• W2986157773 citedByCount "0" @default.
• W2986157773 crossrefType "dissertation" @default.
• W2986157773 hasAuthorship W2986157773A5091636164 @default.
• W2986157773 hasConcept C105795698 @default.
• W2986157773 hasConcept C111919701 @default.
• W2986157773 hasConcept C11413529 @default.
• W2986157773 hasConcept C119857082 @default.
• W2986157773 hasConcept C124101348 @default.
• W2986157773 hasConcept C138885662 @default.
• W2986157773 hasConcept C148483581 @default.
• W2986157773 hasConcept C152124472 @default.
• W2986157773 hasConcept C152139883 @default.
• W2986157773 hasConcept C154945302 @default.
• W2986157773 hasConcept C173801870 @default.
• W2986157773 hasConcept C185429906 @default.
• W2986157773 hasConcept C2776401178 @default.
• W2986157773 hasConcept C33923547 @default.
• W2986157773 hasConcept C41008148 @default.
• W2986157773 hasConcept C41895202 @default.
• W2986157773 hasConcept C51823790 @default.
• W2986157773 hasConceptScore W2986157773C105795698 @default.
• W2986157773 hasConceptScore W2986157773C111919701 @default.
• W2986157773 hasConceptScore W2986157773C11413529 @default.
• W2986157773 hasConceptScore W2986157773C119857082 @default.
• W2986157773 hasConceptScore W2986157773C124101348 @default.
• W2986157773 hasConceptScore W2986157773C138885662 @default.
• W2986157773 hasConceptScore W2986157773C148483581 @default.
• W2986157773 hasConceptScore W2986157773C152124472 @default.
• W2986157773 hasConceptScore W2986157773C152139883 @default.
• W2986157773 hasConceptScore W2986157773C154945302 @default.
• W2986157773 hasConceptScore W2986157773C173801870 @default.
• W2986157773 hasConceptScore W2986157773C185429906 @default.
• W2986157773 hasConceptScore W2986157773C2776401178 @default.
• W2986157773 hasConceptScore W2986157773C33923547 @default.
• W2986157773 hasConceptScore W2986157773C41008148 @default.
• W2986157773 hasConceptScore W2986157773C41895202 @default.
• W2986157773 hasConceptScore W2986157773C51823790 @default.
• W2986157773 hasLocation W29861577731 @default.
• W2986157773 hasOpenAccess W2986157773 @default.
• W2986157773 hasPrimaryLocation W29861577731 @default.
• W2986157773 hasRelatedWork W144959541 @default.
• W2986157773 hasRelatedWork W1550193501 @default.
• W2986157773 hasRelatedWork W1554646759 @default.
• W2986157773 hasRelatedWork W1869713303 @default.
• W2986157773 hasRelatedWork W2002296895 @default.
• W2986157773 hasRelatedWork W2053836403 @default.
• W2986157773 hasRelatedWork W2055660620 @default.
• W2986157773 hasRelatedWork W2100728046 @default.
• W2986157773 hasRelatedWork W2174408904 @default.
• W2986157773 hasRelatedWork W2395046609 @default.
• W2986157773 hasRelatedWork W2535117420 @default.
• W2986157773 hasRelatedWork W2892867475 @default.
• W2986157773 hasRelatedWork W2910328857 @default.
• W2986157773 hasRelatedWork W3041906399 @default.
• W2986157773 hasRelatedWork W3048669090 @default.
• W2986157773 hasRelatedWork W307298448 @default.
• W2986157773 hasRelatedWork W3139252922 @default.
• W2986157773 hasRelatedWork W3170968284 @default.
• W2986157773 hasRelatedWork W3200073826 @default.
• W2986157773 hasRelatedWork W997093450 @default.
• W2986157773 isParatext "false" @default.
• W2986157773 isRetracted "false" @default.
• W2986157773 magId "2986157773" @default.
• W2986157773 workType "dissertation" @default.