Filters








15,313 Hits in 5.5 sec

Issues in the Mining of Heart Failure Datasets

Nongnuch Poolsawad, Lisa Moore, Chandrasekhar Kambhampati, John G. F. Cleland
2014 International Journal of Automation and Computing  
models. 2) Supervised learning has proven to be more suitable for mining clinical data than unsupervised methods.  ...  This paper investigates the characteristics of a clinical dataset using a combination of feature selection and classification methods to handle missing values and understand the underlying statistical  ...  Feature selection approaches with missing values handling for data mining -A case study of heart failure dataset (2011) Scopus  ... 
doi:10.1007/s11633-014-0778-5 fatcat:22e64ezxgfghpfad2pvudxs5hq

Benchmark and Survey of Automated Machine Learning Frameworks [article]

Marc-André Zöller, Marco F. Huber
2021 arXiv   pre-print
This paper is a combination of a survey on current AutoML methods and a benchmark of popular AutoML frameworks on real data sets.  ...  Driven by the selected frameworks for evaluation, we summarize and review important AutoML techniques and methods concerning every step in building an ML pipeline.  ...  A better understanding of the relation between data set meta-features and AutoML algorithms may enable AutoML for the failing data sets and boost meta-learning.  ... 
arXiv:1904.12054v5 fatcat:3gbpofwnl5a3zduqr5vportaly

Benchmark and Survey of Automated Machine Learning Frameworks

Marc-André Zöller, Marco F. Huber
2021 The Journal of Artificial Intelligence Research  
This paper is a combination of a survey on current AutoML methods and a benchmark of popular AutoML frameworks on real data sets.  ...  Driven by the selected frameworks for evaluation, we summarize and review important AutoML techniques and methods concerning every step in building an ML pipeline.  ...  A better understanding of the relation between data set meta-features and AutoML algorithms may enable AutoML for the failing data sets and boost meta-learning.  ... 
doi:10.1613/jair.1.11854 fatcat:whi5mdcidrfffmid6mabprzug4

Incremental Search Space Construction for Machine Learning Pipeline Synthesis [article]

Marc-André Zöller, Tien-Dung Nguyen, Marco F. Huber
2021 arXiv   pre-print
Many studies have investigated efficient methods for algorithm selection and hyperparameter optimization.  ...  In this paper, we propose a data-centric approach based on meta-features for pipeline construction and hyperparameter optimization inspired by human behavior.  ...  At least if supported by the frameworks. For example, TPOT can only handle discretized continuous hyperparameters.  ... 
arXiv:2101.10951v1 fatcat:2jtf7ljkgfcfhojqzwmrnrplty

Accessing Imbalance Learning Using Dynamic Selection Approach in Water Quality Anomaly Detection

Eustace M. Dogo, Nnamdi I. Nwulu, Bhekisipho Twala, Clinton Aigbavboa
2021 Symmetry  
SMOTE+Tomek Links), and one missing data method (missForest) are proposed and evaluated.  ...  (KNORA-U) and Meta-Learning for Dynamic Ensemble Selection (META-DES) in combination with homogeneous and heterogeneous ensemble models and three SMOTE-based resampling algorithms (SMOTE, SMOTE+ENN and  ...  Acknowledgments: We want to thank the University of Johannesburg and Durban University of Technology South Africa for making the resources available to complete this work.  ... 
doi:10.3390/sym13050818 fatcat:lpuhnsdvtrbvlp2mpikicywpuy

What you see is what you can change: Human-centered machine learning by interactive visualization

Dominik Sacha, Michael Sedlmair, Leishi Zhang, John A. Lee, Jaakko Peltonen, Daniel Weiskopf, Stephen C. North, Daniel A. Keim
2017 Neurocomputing  
Visual analytics (VA) systems help data analysts solve complex problems interactively, by integrating automated data analysis and mining, such as machine learning (ML) based methods, with interactive visualizations  ...  We propose a conceptual framework that models human interactions with ML components in the VA process, and that puts the central relationship between automated algorithms and interactive visualizations  ...  Hence data editing might be seen as a kind of "meta" cross-validation, requiring proper quality assessment for the user's task.  ... 
doi:10.1016/j.neucom.2017.01.105 fatcat:nv7dh4uhprd43npfhhy7objowm

Dynamic integration of multiple data mining techniques in a knowledge discovery management system

Seppo J. Puuronen, Vagan Terziyan, Artyom Katasonov, Alexey Tsymbal, Belur V. Dasarathy
1999 Data Mining and Knowledge Discovery: Theory, Tools, and Technology  
Method was evaluated on three data sets taken from the UCI machine learning repository, with which well-known classifier integration methods are proven to perform badly.  ...  Thus first we have developed a learning technique to derive the "competence areas" for all classifiers in our space of cases.  ...  There are no missing values in the data set.  ... 
doi:10.1117/12.339975 dblp:conf/dmkdttt/PuuronenTKT99 fatcat:tycon2hhpjeclh2xw3llx6rrti

Detecting Credit Card Fraud using Data Mining Techniques - Meta-Learning

T. Abdul Razak, G. Najeeb Ahmed
2015 Indian Journal of Science and Technology  
Meta-learning techniques extend this concept by providing methods for knowledge discovery process automatization.  ...  Meta-learning introduces various interesting concepts, including data meta-features, meta-knowledge, algorithm recommendation systems, autonomous process builders, etc.  ...  to induce a meta model for algorithm selection.  ... 
doi:10.17485/ijst/2015/v8i28/83326 fatcat:bj6rpfck3vfwpo6d5sm5i2ezk4

Feedback-driven interactive exploration of large multidimensional data supported by visual classifier

Michael Behrisch, Fatih Korkmaz, Lin Shao, Tobias Schreck
2014 2014 IEEE Conference on Visual Analytics Science and Technology (VAST)  
We introduce a framework for a feedback-driven view exploration, inspired by relevance feedback approaches used in Information Retrieval.  ...  Focus+Context or Semantic Zoom Interfaces can help to some extent to efficiently search for interesting views or data segments, yet they show scalability problems for very large data sets.  ...  Interest-Driven Data Filtering for Visual Analysis Methods for visual data analysis need to handle increasingly large data sets.  ... 
doi:10.1109/vast.2014.7042480 dblp:conf/ieeevast/BehrischKSS14 fatcat:wbwwazsjzrbyva4guov3gxjjji

Graph Prototypical Networks for Few-shot Learning on Attributed Networks [article]

Kaize Ding, Jianling Wang, Jundong Li, Kai Shu, Chenghao Liu, Huan Liu
2020 arXiv   pre-print
To answer these questions, in this paper, we propose a graph meta-learning framework -- Graph Prototypical Networks (GPN).  ...  model for handling the target classification task.  ...  Specifically, PN and MAML are two representative few-shot learning models for i.i.d. data, while Meta-GNN is able to handle graph-structured data by integrating graph neural networks with meta-learning  ... 
arXiv:2006.12739v3 fatcat:4m6v3zp7ybbdpagccimrg2l6te

One-shot Learning for Temporal Knowledge Graphs [article]

Mehrnoosh Mirtaheri, Mohammad Rostami, Xiang Ren, Fred Morstatter, Aram Galstyan
2020 arXiv   pre-print
We address this shortcoming by proposing a one-shot learning framework for link prediction in temporal knowledge graphs.  ...  This observation has given rise to recent interest in low-shot learning methods that are able to generalize from only a few examples.  ...  Meta-Graph (Bose et al. 2019 ) is a gradient-based meta learning approach for graphs.  ... 
arXiv:2010.12144v1 fatcat:oef5cubo6fdhdhy6rl2z3ectjm

Heterogeneous Data and Big Data Analytics

Lidong Wang
2017 Automatic Control and Information Sciences  
This paper introduces data processing methods for heterogeneous data and Big Data analytics, Big Data tools, some traditional data mining (DM) and machine learning (ML) methods.  ...  Deep learning and its potential in Big Data analytics are analysed.  ...  Whenever people handle a dataset with missing values, they can follow several strategies.  ... 
doi:10.12691/acis-3-1-3 fatcat:t3yzrk4r2bfornki34khobe4su

Enhanced spatiotemporal relational probability trees and forests

Amy McGovern, Nathaniel Troutman, Rodger A. Brown, John K. Williams, Jennifer Abernethy
2012 Data mining and knowledge discovery  
increase their ability to learn using spatiotemporal data.  ...  Keywords Spatiotemporal relational learning · Statistical relational learning · Hazardous weather A. McGovern et al.  ...  A common difficulty when dealing with real-world data is handling missing values (Liu et al. 1997 ). In the spatiotemporal framework, there are three cases of missing values.  ... 
doi:10.1007/s10618-012-0261-2 fatcat:gcgu6nskzzh7xhcjlaavqrnomq

Practical and sample efficient zero-shot HPO [article]

Fela Winkelmolen, Nikita Ivkin, H. Furkan Bozkurt, Zohar Karnin
2020 arXiv   pre-print
Zero-shot hyperparameter optimization (HPO) is a simple yet effective use of transfer learning for constructing a small list of hyperparameter (HP) configurations that complement each other.  ...  The second, for settings where finding, tuning and testing a surrogate model is problematic, is a multi-fidelity technique combining HyperBand with submodular optimization.  ...  We design an adaptive method for selecting i, j pairs to query while learning this surrogate function.  ... 
arXiv:2007.13382v1 fatcat:cmi6dtoj7zghhfp4tyhfgmegnq

Empirical study of seven data mining algorithms on different characteristics of datasets for biomedical classification applications

Yiyan Zhang, Yi Xin, Qin Li, Jianshe Ma, Shuai Li, Xiaodan Lv, Weiqi Lv
2017 BioMedical Engineering OnLine  
Hence, finding a suitable algorithm for a dataset is becoming an important emphasis for biomedical researchers to solve practical problems promptly.  ...  The applicability of the seven data mining algorithms on the datasets with different characteristics was summarized to provide a reference for biomedical researchers or beginners in different fields.  ...  Consent for publication Not applicable. Ethics approval and consent to participate Not applicable.  ... 
doi:10.1186/s12938-017-0416-x pmid:29096638 pmcid:PMC5668968 fatcat:chqj7xyifvctzpbz643mgoipwa
« Previous Showing results 1 — 15 out of 15,313 results