Visible to the public An ensemble model with hierarchical decomposition and aggregation for highly scalable and robust classification

TitleAn ensemble model with hierarchical decomposition and aggregation for highly scalable and robust classification
Publication TypeConference Paper
Year of Publication2017
AuthorsVu, Q. H., Ruta, D., Cen, L.
Conference Name2017 Federated Conference on Computer Science and Information Systems (FedCSIS)
Date Publishedsep
Keywordsbinary classification problem, compositionality, Computational modeling, computer game, computer games, Data models, Decision trees, decomposition, Deep Learning, ensemble model, extreme gradient boosted decision trees, feature extraction, flexible robust scheme, game state information, Hearthstone, highly scalable classification, learning (artificial intelligence), logistic regression, Logistics, machine learning, Metrics, model decomposition, pattern classification, Predictive models, pubcrawl, regression analysis, sub-model integration, sub-model training, Training, Xgboost
Abstract

This paper introduces an ensemble model that solves the binary classification problem by incorporating the basic Logistic Regression with the two recent advanced paradigms: extreme gradient boosted decision trees (xgboost) and deep learning. To obtain the best result when integrating sub-models, we introduce a solution to split and select sets of features for the sub-model training. In addition to the ensemble model, we propose a flexible robust and highly scalable new scheme for building a composite classifier that tries to simultaneously implement multiple layers of model decomposition and outputs aggregation to maximally reduce both bias and variance (spread) components of classification errors. We demonstrate the power of our ensemble model to solve the problem of predicting the outcome of Hearthstone, a turn-based computer game, based on game state information. Excellent predictive performance of our model has been acknowledged by the second place scored in the final ranking among 188 competing teams.

URLhttps://annals-csis.org/proceedings/2017/drp/564.html
DOI10.15439/2017F564
Citation Keyvu_ensemble_2017