209 Hits in 6.7 sec

Tuned data mining

Wolfgang Konen, Patrick Koch, Oliver Flasch, Thomas Bartz-Beielstein, Martina Friese, Boris Naujoks
2011 Proceedings of the 13th annual conference on Genetic and evolutionary computation - GECCO '11  
The framework TDM (Tuned Data Mining) was developed to facilitate the search for good parameters and the comparison of different tuners.  ...  In benchmark tasks like the Data Mining Cup (DMC) tuned models achieve remarkably better ranks than their untuned counterparts.  ...  RESULTS ON BENCHMARK TASKS The benchmark tasks studied in this paper are briefly summarized in Tab. 3.  ... 
doi:10.1145/2001576.2001844 dblp:conf/gecco/KonenKFBFN11 fatcat:gxbk3ne24nfwhb7gtr4rfiagcu

To Tune or Not to Tune?

Ayat Fekry, Lucian Carata, Thomas Pasquier, Andrew Rice, Andy Hopper
2020 Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining  
This experimental study presents a number of issues that pose a challenge for practical configuration tuning and its deployment in data analytics frameworks.  ...  We adapt different ML techniques in order to obtain efficient incremental tuning in our problem domain, and propose Tuneful, a configuration tuning framework.  ...  ACKNOWLEDGEMENTS We thank Google and Amazon for generously supporting us with Google Cloud and AWS research credits to perform this study.  ... 
doi:10.1145/3394486.3403299 fatcat:agiqj22vpba2jdwajdfueyakxu

ICICLES: Self-Tuning Samples for Approximate Query Answering

Venkatesh Ganti, Mong-Li Lee, Raghu Ramakrishnan
2000 Very Large Data Bases Conference  
In this paper, we introduce icicles, a new class of samples that tune themselves to a dynamic workload.  ...  In a detailed experimental study, we examine the validity and performance of icicles. * Supported by a Microsoft Graduate Fellowship.  ...  Wei-Yin Loh for discussions on statistical estimators.  ... 
dblp:conf/vldb/GantiLR00 fatcat:gnjw7qjjxjfmpknyfjotaicaqq

Investigating the parameter space of evolutionary algorithms

Moshe Sipper, Weixuan Fu, Karuna Ahuja, Jason H. Moore
2018 BioData Mining  
What probabilities should one assign to crossover and mutation?  ...  The practice of evolutionary algorithms involves the tuning of many parameters. How big should the population be? How many generations should the algorithm run?  ...  Availability of data and materials The datasets supporting the conclusions of this article are available in the PMLB repository, EpistasisLab/penn-ml-benchmarks (see also [30] ).  ... 
doi:10.1186/s13040-018-0164-x pmid:29467825 pmcid:PMC5816380 fatcat:4isbh6l47zaafmk5salg33blyy

Product Feature Mining: Semantic Clues versus Syntactic Constituents

Liheng Xu, Kang Liu, Siwei Lai, Jun Zhao
2014 Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
Product feature mining is a key subtask in fine-grained opinion mining. Previous works often use syntax constituents in this task.  ...  This paper proposes a novel product feature mining method which leverages lexical and contextual semantic clues.  ...  To alleviate the data sparsity problem, EB is first trained on a very large corpus 3 (denoted by C), and then fine-tuned on the target review corpus R.  ... 
doi:10.3115/v1/p14-1032 dblp:conf/acl/XuLLZ14 fatcat:hvnlc5ytcvg2lgkugkuzitobhe

Analyzing textual databases using data mining to enable fast product development processes

Rakesh Menon, Loh Han Tong, S. Sathiyakeerthi
2005 Reliability Engineering & System Safety  
Further, an analysis tool, data mining that could be used to analyse these textual databases so that we could extract information from them quickly and hence be able to access them at the right time when  ...  Due to the fact that, unanticipated information plays a more and more dominant role, especially in highly innovative business processes, we focus our attention on textual databases since textual databases  ...  Data Mining Operations Depending on the objectives of an analysis, different types of data mining operations could be used.  ... 
doi:10.1016/j.ress.2004.07.007 fatcat:cpjqun33angybdocpe5ejk6swq

Data-driven search-based software engineering

Vivek Nair, Amritanshu Agrawal, Jianfeng Chen, Wei Fu, George Mathew, Tim Menzies, Leandro Minku, Markus Wagner, Zhe Yu
2018 Proceedings of the 15th International Conference on Mining Software Repositories - MSR '18  
The paper briefly sets out to act as a practical guide to develop new DSE techniques and also to serve as a teaching resource. This paper also presents a resource ( for exploring DSE.  ...  This paper introduces Data-Driven Search-based Software Engineering (DSE), which combines insights from Mining Software Repositories (MSR) and Search-based Software Engineering (SBSE).  ...  One word of warning: as algorithm tuning is already computationally expensive, the tuning of algorithm tuners is even more so. While Dang et al.  ... 
doi:10.1145/3196398.3196442 dblp:conf/msr/NairACFMMM0Y08 fatcat:2hyij3e5v5fz5ct4xo73z3heym

Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts

Jiaqi Ma, Zhe Zhao, Xinyang Yi, Jilin Chen, Lichan Hong, Ed H. Chi
2018 Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD '18  
Furthermore, we demonstrate the performance improvements by MMoE on real tasks including a binary classi cation benchmark, and a large-scale content recommendation system at Google.  ...  To validate our approach on data with di erent levels of task relatedness, we rst apply it to a synthetic dataset where we control the task relatedness.  ...  Finally, we conduct experiments on real benchmark data and a large-scale production recommendation system with hundreds of millions of users and items.  ... 
doi:10.1145/3219819.3220007 dblp:conf/kdd/MaZYCHC18 fatcat:islsbwhuhjfdnavw5n77n7zw7q

Tuning and evolution of support vector kernels

Patrick Koch, Bernd Bischl, Oliver Flasch, Thomas Bartz-Beielstein, Claus Weihs, Wolfgang Konen
2012 Evolutionary Intelligence  
To circumvent these obstacles for the 'non-expert' data mining user, which may hinder the wider use of SVM in data mining, we present two solutions for optimizing kernel functions: (a) automated hyperparameter  ...  A tuned kernel can improve the trained model, if standard kernels are insufficient for achieving a good transformation. We apply tuning to SVM kernels for both regression and classification.  ...  We give a detailed experimental study on tuned SVM kernels in Sec. 3 . Here, we use GP-evolved and SPO-tuned kernels to build optimized models for data mining problems.  ... 
doi:10.1007/s12065-012-0073-8 fatcat:mkvbwojnffbhtnaosgfht752fy

ArabicDialects: An Efficient Framework for Arabic Dialects Opinion Mining on Twitter using Optimized Deep Neural Networks

Diaa Salama AbdElminaam, Nabil Neggaz, Ibrahim Abd Elatif Gomaa, Fatma Helmy Ismail, Ahmed Elsawy
2021 IEEE Access  
Many researchers focused on collecting Arabic tweets to establish a benchmark data set as ASTD introduced in [23] .  ...  Several studies have been realized for opinion mining based on different language as English, French , Spanish and Italien however a few works are interested to Arabic language.  ... 
doi:10.1109/access.2021.3094173 fatcat:3hmw674qt5fp3jk4m46j2lr4z4

Evolutionary Algorithm Parameters and Methods to Tune Them [chapter]

A. E. Eiben, S. K. Smit
2011 Autonomous Search  
I n this chapter we discuss the notion of Evolutionary Algorithm (EA) parameters and propose a distinction between EAs and EA instances, based on the type of parameters used to specify their details.  ...  Furthermore, we consider the most important aspects of the parameter tuning problem and give an overview of existing parameter tuning methods.  ...  In the context of statistics and data mining one distinguishes two types of variables (rather then parameters) depending on the presence of an ordered structure, but a universal terminology is lacking  ... 
doi:10.1007/978-3-642-21434-9_2 fatcat:zur55ehoere7zlfsmv57uf3feq

Multi-Intention-Aware Configuration Selection for Performance Tuning

Haochen He, Zhouyang Jia, Shanshan Li, Yue Yu, Chenglong Zhou, Qing Liao, Ji Wang, Xiangke Liao
2022 International Conference on Software Engineering  
Automatic configuration tuning helps users who intend to improve software performance. However, the auto-tuners are limited by the huge configuration search space.  ...  To reduce the search space, researchers mainly focus on pre-selecting performance-related parameters which requires a heavy stage of dynamically running under different configurations to build performance  ...  To illustrate the effectiveness of SafeTune, we conduct a user study on one of the auto-tuners and manually validate the result.  ... 
doi:10.1145/3510003.3510094 dblp:conf/icse/HeJLYZ0WL22 fatcat:jeiuaonq5bbzhnz5pmyxnphf44

AutoTunium: An Evolutionary Tuner for General-Purpose Multicore Applications

Andreas Zwinkau, Victor Pankratius
2012 2012 IEEE 18th International Conference on Parallel and Distributed Systems  
To obtain good performance, programmers typically try out different software tuning parameter configurations on each platform.  ...  We quantify the effectiveness of various tuning strategies on a diverse set of real applications and multicore platforms.  ...  The AutoTunium system demonstrates the applicability of automatic tuning on a wide set of different programs including video encoding, image processing, ray tracing, clustering, data mining, simulations  ... 
doi:10.1109/icpads.2012.61 dblp:conf/icpads/ZwinkauP12 fatcat:epbezr6f6fg5lhouqqhknihvru

Parameter tuning for configuring and analyzing evolutionary algorithms

A.E. Eiben, S.K. Smit
2011 Swarm and Evolutionary Computation  
The framework is based on a three-tier hierarchy of a problem, an evolutionary algorithm (EA), and a tuner.  ...  For the survey part we establish different taxonomies to categorize tuning methods and review existing work.  ...  In the context of statistics and data mining one distinguishes two types of variables (rather than parameters) depending on the presence of an ordered structure, but a universal terminology is lacking  ... 
doi:10.1016/j.swevo.2011.02.001 fatcat:iqwlokswenbsvky3vk43rxokwm


Denis Baylor, Levent Koc, Chiu Yuen Koo, Lukasz Lew, Clemens Mewald, Akshay Naresh Modi, Neoklis Polyzotis, Sukriti Ramesh, Sudip Roy, Steven Euijong Whang, Martin Wicke, Eric Breck (+10 others)
2017 Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17  
We present the case study of one deployment of TFX in the Google Play app store, where the machine learning models are refreshed continuously as new data arrive.  ...  Creating and maintaining a platform for reliably producing and deploying machine learning models requires careful orchestration of many components-a learner for generating models based on training data  ...  A tuner integrated with the trainer can also automatically optimize the hyperparameters based on users' objectives and data.  ... 
doi:10.1145/3097983.3098021 dblp:conf/kdd/BaylorBCFFHHIJK17 fatcat:vwiikv3q7fcvhlpe2aakgruweu
« Previous Showing results 1 — 15 out of 209 results