31,557 Hits in 6.7 sec

Algorithms for Active Classifier Selection

Paul N. Bennett, David M. Chickering, Christopher Meek, Xiaojin Zhu
2017 Proceedings of the Tenth ACM International Conference on Web Search and Data Mining - WSDM '17  
In this paper, we consider model-selection algorithms for these precision-constrained scenarios.  ...  We develop adaptive model-selection algorithms to identify, using as few samples as possible, the best classifier from among a set of (precision) qualifying classifiers.  ...  Among precision-acceptable classifiers, we want to select the one with the largest recall.  ... 
doi:10.1145/3018661.3018730 dblp:conf/wsdm/BennettCMZ17 fatcat:ged2kupzc5bhhdmyqezeh5s6ii

Approximate Selection with Guarantees using Proxies [article]

Daniel Kang, Edward Gan, Peter Bailis, Tatsunori Hashimoto, Matei Zaharia
2022 arXiv   pre-print
In this work, we introduce novel algorithms for approximate selection queries with statistical accuracy guarantees.  ...  We show that our algorithms can improve query result quality by up to 30x for both the precision and recall targets in both real and synthetic datasets.  ...  We further thank Tadashi Fukami, Trevor Hebert, and Isaac Westlund for their helpful discussions.  ... 
arXiv:2004.00827v4 fatcat:5mktihaghrgc3ajx6xxhfotvpi

Tuning the ensemble selection process of schema matchers

Avigdor Gal, Tomer Sagi
2010 Information Systems  
To the best of our knowledge, none of the existing algorithmic solutions offer such a selection feature. In this paper we provide a thorough investigation of this research topic.  ...  Schema matching research has been going on for more than 25 years now.  ...  To understand this phenomenon, recall that the error measure that was defined for SMB balances Precision with Recall. SMB chose matchers that improve Precision.  ... 
doi:10.1016/ fatcat:nketdrw5czavvcihc327tnmlqe

Active seed selection for constrained clustering

Viet-Vu Vu, Nicolas Labroche
2017 Intelligent Data Analysis  
Active learning for semi-supervised clustering allows algorithms to solicit a domain expert to provide side information as instances constraints, for example a set of labeled instances called seeds.  ...  In this paper, we propose a new active seed selection algorithm that relies on a k-nearest neighbors structure to locate dense potential clusters and efficiently query and propagate expert information.  ...  active constraint selection algorithms [43, 44, 45] .  ... 
doi:10.3233/ida-150499 fatcat:vdnckohisfaqjanu6le7alh7de

Prototype-Based Sample Selection for Active Hashing

Cheong Hee Park
2015 Journal of Computer Science  
For expert labeling, we select prototypes from clusters which do not contain any data points with labeled information so that all areas can be covered effectively.  ...  In this study, we present an active hashing method by prototype-based sample selection.  ...  Precision and recall are computed for each query point and then averaged for all the query points.  ... 
doi:10.3844/jcssp.2015.839.844 fatcat:qltdpb3kmfhgbbmg7mhdt7qt4u

Core-set Selection Using Metrics-based Explanations (CSUME) for multiclass ECG [article]

Sagnik Dakshit, Barbara Mukami Maweu, Sristi Dakshit, Balakrishnan Prabhakaran
2022 arXiv   pre-print
Our experimental results show a 9.67% and 8.69% precision and recall improvement with a significant training data volume reduction of 50%.  ...  This also provides an understanding (for algorithm developers) as to why a sample was selected as more informative over others for the improvement of deep learning model performance.  ...  These three metrics allow us to select the samples from the incoming dataset that maximize testing model performance in terms of accuracy, precision, and recall.  ... 
arXiv:2205.14508v1 fatcat:yw5imemrgjeqvpirneyuweezdu

Comparative Analysis of Selected Heterogeneous Classifiers for Software Defects Prediction Using Filter-Based Feature Selection Methods

Abimbola G Akintola, Abdullateef Balogun, Fatimah B Lafenwa-Balogun, Hameed A Mojeed
2018 FUOYE Journal of Engineering and Technology  
The datasets were classified by the selected classifiers which were carefully selected based on heterogeneity.  ...  with FilterSubsetEval had the best accuracy.  ...  It is used for classifying data into different classes according to some constraints.  ... 
doi:10.46792/fuoyejet.v3i1.178 fatcat:m4cjd63o4jhtjhsbhxd57b5bge

Software Module Fault Prediction using Convolutional Neural Network with Feature Selection

Rupali Sharma, Parveen Kakkar
2016 International Journal of Software Engineering and Its Applications  
A sequence of rigorous activities under certain constraints is followed to come up with reliable software.  ...  The comparative analysis is performed on the basis of accuracy, precision, recall and F1-measure. The results clearly show better performance of the proposed CNN based technique than HySOM.  ...  A sequence of rigorous activities under certain constraints is followed to come up with reliable software.  ... 
doi:10.14257/ijseia.2016.10.12.27 fatcat:br6q52awq5ewrm2aurkj26qmiu

Harmonious Semantic Line Detection via Maximal Weight Clique Selection [article]

Dongkwon Jin, Wonhui Park, Seong-Gyun Jeong, Chang-Su Kim
2021 arXiv   pre-print
A novel algorithm to detect an optimal set of semantic lines is proposed in this work. We develop two networks: selection network (S-Net) and harmonization network (H-Net).  ...  Finally, we determine a maximal weight clique representing an optimal set of semantic lines. Moreover, to assess the overall harmony of detected lines, we propose a novel metric, called HIoU.  ...  The proposed algorithm provides a poorer recall but a better precision than the conventional algorithms. F-measure is the harmonic mean of recall and precision.  ... 
arXiv:2104.06903v1 fatcat:3dukcxfhm5dupkck6ywvydb3cm

Feature Selection for Effective Text Classification using Semantic Information

Rajul Jain, Nitin Pise
2015 International Journal of Computer Applications  
Different datasets are constructed with each different collection of features to gain an understanding about what is the best representation for text data depending on different types of classifiers.  ...  to incorporate the context information with the text in machine learning for better classification accuracy.  ...  Its score is maximized when the values of recall and precision are equal or close; otherwise, the smaller of recall and precision dominates the value of Fl.  ... 
doi:10.5120/19861-1818 fatcat:s5rozhv4krg4hofinpx6lhdrtq

Constraint Selection in Metric Learning [article]

Hoel Le Capitaine
2016 arXiv   pre-print
The proposed approach relies on a loss-dependent weighted selection of constraints that are used for learning the metric.  ...  This paper presents a simple way of improving accuracy and scalability of any iterative metric learning algorithm, where constraints are obtained prior to the algorithm.  ...  (e.g. precision, recall, F-measure).  ... 
arXiv:1612.04853v1 fatcat:yp7uudeqxrgszop2nclhdu7a5u

Feature Selection using Stochastic Gates [article]

Yutaro Yamada and Ofir Lindenbaum and Sahand Negahban and Yuval Kluger
2020 arXiv   pre-print
Feature selection problems have been extensively studied for linear estimation, for instance, Lasso, but less emphasis has been placed on feature selection for non-linear functions.  ...  In this study, we propose a method for feature selection in high-dimensional non-linear function estimation problems.  ...  Acknowledgements The authors thank Nicolas Casey and the anonymous reviewers for their helpful feedback.  ... 
arXiv:1810.04247v7 fatcat:towtuaxgtva7ff2x42lpyzb63y

Ontology-based Feature Selection: A Survey [article]

Konstantinos Sikelis, George E Tsekouras, Konstantinos I Kotis
2021 arXiv   pre-print
First, some of the most common classification and feature selection algorithms are briefly presented.  ...  selection.  ...  For the evaluation, they compared precision (ratio of correct feature to retrieved features) and recall (ratio of correct features to ideal features) against manually selected features from hu-man experts  ... 
arXiv:2104.07720v2 fatcat:zxd5milohbfj3ewtaw3siwhi4m

Feature and Region Selection for Visual Learning

Ji Zhao, Liantao Wang, Ricardo Cabral, Fernando De la Torre
2016 IEEE Transactions on Image Processing  
To answer these questions, this paper presents a method for feature selection and region selection in the visual BoW model.  ...  The main idea is to assign latent weights to the features or regions, and jointly optimize these latent variables with the parameters of a classifier (e.g., support vector machine).  ...  To provide a quantitative measure for the localization performance, we compared all methods using precision-recall curves, as shown in Fig. 5 .  ... 
doi:10.1109/tip.2016.2514503 pmid:26742135 fatcat:6v7zgdldzveeblnd5zs6bgsupm

On optimal service selection

P. A. Bonatti, P. Festa
2005 Proceedings of the 14th international conference on World Wide Web - WWW '05  
We designed and implemented both exact and heuristic (suboptimal) algorithms for the hard case, and carried out a preliminary experimental evaluation with interesting results.  ...  In this paper we formalize three kinds of optimal service selection problems, based on different criteria. Then we study their complexity and implement solutions.  ...  Currently, it seems that the algorithm with a guaranteed 0.52 bound on relative error is too slow for real-time service selection over large workflows and offer sets.  ... 
doi:10.1145/1060745.1060823 dblp:conf/www/BonattiF05 fatcat:423zxbujfff3hjc5wcr6sejfq4
« Previous Showing results 1 — 15 out of 31,557 results