Filters








139,404 Hits in 3.4 sec

Non-exhaustive, Overlapping Clustering via Low-Rank Semidefinite Programming

Yangyang Hou, Joyce Jiyoung Whang, David F. Gleich, Inderjit S. Dhillon
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
Non-exhaustive, overlapping clustering A data point is allowed to be outside of any cluster. Clusters are allowed to overlap with each other.  ...  Overview Goal: more accurate and more reliable solutions than the iterative NEO-K-Means algorithm by paying additional computational cost  ...  Non-exhaustive, overlapping k-means. SDM, 2015.  ... 
doi:10.1145/2783258.2783398 dblp:conf/kdd/HouWGD15 fatcat:fmfaqd5hlnh65jpxjqxykcenaq

Non-exhaustive, Overlappingk-means [chapter]

Joyce Jiyoung Whang, Inderjit S. Dhillon, David F. Gleich
2015 Proceedings of the 2015 SIAM International Conference on Data Mining  
By studying the objective, we are able to obtain a simple iterative algorithm which we call NEO-K-Means (Non-Exhaustive Overlapping K-Means).  ...  Furthermore, by considering an extension to weighted kernel k-means, we can tackle the case of non-exhaustive and overlapping graph clustering.  ...  To optimize the new objective, we present a simple iterative algorithm called non-exhaustive, overlapping k-means, or NEO-K-Means in short.  ... 
doi:10.1137/1.9781611974010.105 dblp:conf/sdm/WhangDG15 fatcat:jzfql4xoerc3hhuqgx4e6lu2pe

Clustering patents using non-exhaustive overlaps

Charles V. Trappey, Amy J.C. Trappey, Chun-Yi Wu
2010 Journal of Systems Science and Systems Engineering  
A clustering algorithm with non-exhaustive overlaps is proposed to overcome deficiencies with exhaustive clustering methods used in patent mining and technology discovery.  ...  The non-exhaustive clustering approach allows for the clustering of patent documents with overlapping technical findings and claims, a feature that enables the grouping of patents that define related key  ...  The characteristic comparison between non-exhaustive overlapping and K-means is summarized in Table 9 .  ... 
doi:10.1007/s11518-010-5134-x fatcat:wlqumj7bfnhpndgpwykvwbmd6q

Fast Multiplier Methods to Optimize Non-exhaustive, Overlapping Clustering

Yangyang Hou, Joyce Jiyoung Whang, David F. Gleich, Inderjit S. Dhillon
2016 Proceedings of the 2016 SIAM International Conference on Data Mining  
We recently proposed the NEO-K-Means (Non-Exhaustive, Overlapping K-Means) objective as a way to address both issues in an integrated fashion.  ...  Traditional clustering algorithms, such as K-means, assign every data point to exactly one cluster. However, in real-world datasets, the clusters may overlap with each other.  ...  We recently proposed the NEO-K-Means (Non-Exhaustive, Overlapping K-Means) objective as a generalization of the k-means clustering objective that allows us to simultaneously identify overlapping clusters  ... 
doi:10.1137/1.9781611974348.34 dblp:conf/sdm/HouWGD16 fatcat:72issb6yzzd3liylse2droux5y

Fast Multiplier Methods to Optimize Non-exhaustive, Overlapping Clustering [article]

Yangyang Hou, Joyce Jiyoung Whang, David F. Gleich, Inderjit S. Dhillon
2016 arXiv   pre-print
We recently proposed the NEO-K-Means (Non-Exhaustive, Overlapping K-Means) objective as a way to address both issues in an integrated fashion.  ...  Traditional clustering algorithms, such as K-means, assign every data point to exactly one cluster. However, in real-world datasets, the clusters may overlap with each other.  ...  We recently proposed the NEO-K-Means (Non-Exhaustive, Overlapping K-Means) objective as a generalization of the k-means clustering objective that allows us to simultaneously identify overlapping clusters  ... 
arXiv:1602.01910v1 fatcat:mqu6q7jnb5cbphv7tqshwy6544

Towards a Unified Taxonomy of Biclustering Methods [article]

Dmitry I. Ignatov, Bruce W. Watson
2017 arXiv   pre-print
So, in this paper we propose to use concept lattices as a tool for taxonomy building (in the biclustering domain) and attribute exploration as means for cross-domain taxonomy completion.  ...  Exclusive means whether a gene (condition) has to belong no more than one bicluster; e.g., in non-exclusive case overlapping is allowed.  ...  biclusters with checkerboard structure. d) Exclusive-rows biclusters. e) Exclusive-columns biclusters. f) Non-Overlapping biclusters with tree structure. g) Non-Overlapping non-exclusive biclusters. h  ... 
arXiv:1702.05376v1 fatcat:lzso53vqgvbphkjnmuk7uk6yb4

Infinite Plaid Models for Infinite Bi-Clustering

Katsuhiko Ishiguro, Issei Sato, Masahiro Nakano, Akisato Kimura, Naonori Ueda
2016 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
We propose a probabilistic model for non-exhaustive and overlapping (NEO) bi-clustering.  ...  More specifically we study the non-exhaustive and overlapping (NEO) bi-clustering. Here we distinguish between the exhaustive and non-exhaustive bi-clustering. Exhaustive bi-clustering (e.g.  ...  However model selection inevitably requires Non-exhaustive and Overlapping (NEO) Bi-clustering Reduce man-hour cost! only a limited number of sub-matrices!  ... 
doi:10.1609/aaai.v30i1.10192 fatcat:a2nwfa7gbnh7hfwxtvdukuicfy

On subgroup discovery in numerical domains

Henrik Grosskreutz, Stefan Rüping
2009 Data mining and knowledge discovery  
In particular, we have considered frequency and entropy discretization, combined with both overlapping and non-overlapping intervals, as well as exhaustive search.  ...  The figure shows the result for different strategies: non-overlapping frequency discretization with five splits ('n/o freq.'), non-overlapping entropy discretization ('n/o entr.'), overlapping frequency  ... 
doi:10.1007/s10618-009-0136-3 fatcat:boxdnjif3janbgh4mecw32wnsu

On Subgroup Discovery in Numerical Domains [chapter]

Henrik Grosskreutz, Stefan Rüping
2009 Lecture Notes in Computer Science  
In particular, we have considered frequency and entropy discretization, combined with both overlapping and non-overlapping intervals, as well as exhaustive search.  ...  The figure shows the result for different strategies: non-overlapping frequency discretization with five splits ('n/o freq.'), non-overlapping entropy discretization ('n/o entr.'), overlapping frequency  ... 
doi:10.1007/978-3-642-04180-8_15 fatcat:qgbc2xrxpjfatffuswppwtq3pu

Feasibility of optimally assigning channels by exhaustive search in commercial multi-radio wireless mesh networks

Wei Xie, Ying Jun (Angela) Zhang, Mihail L. Sichitiu, Liqun Fu, Yan Yao
2009 Telecommunications Systems  
., exhaustive search), as most of existing commercial WMNs are of small/medium scale. In this paper, we attempt to answer this practical question.  ...  Furthermore, we estimate the theoretical runtime of determining the optimal channel assignment by exhaustive search and validate it through experiments.  ...  In CA k , the link set L is partitioned into k non-overlapping subsets as follows. (9) where L k corresponds to the distinct channel CH k that has not been used by former k − 1 link subsets.  ... 
doi:10.1007/s11235-009-9216-3 fatcat:7p5ffcooo5ckpmrl7eab7rgvke

Air Mass Flow Analysis for SI Engine: EGR and Scavenging

Antonio Palma, Angelo Palladino, Giovanni Fiengo, Ferdinando De Cristofaro, Fabio Garofalo, Luigi Glielmo
2008 IFAC Proceedings Volumes  
Similarly, the Scavenging phenomenon is the air mass flowing from intake manifold directly to exhaust manifold, due to an overlap of intake and exhaust valves, without participating at the combustion.  ...  In this paper, the authors present a mean value engine model, aimed at the challenging purpose of the analysis of EGR and Scavenging.  ...  It means that fresh air flow toward exhaust manifold can occur only at the beginning of the intake phase if the exhaust valves are still opened.  ... 
doi:10.3182/20080706-5-kr-1001.01436 fatcat:tmtgvg5c3bfldi4m7uo67fzxmy

Bayesian Nonparametrics for Non-exhaustive Learning [article]

Yicheng Cheng, Bartek Rajwa, Murat Dundar
2019 arXiv   pre-print
Non-exhaustive learning (NEL) is an emerging machine-learning paradigm designed to confront the challenge of non-stationary environments characterized by anon-exhaustive training sets lacking full information  ...  about the available classes.Unlike traditional supervised learning that relies on fixed models, NEL utilizes self-adjusting machine learning to better accommodate the non-stationary nature of the real-world  ...  All cases in between are considered to be non-exhaustive learning. All features in each dataset are normalized to have zero mean and unit variance.  ... 
arXiv:1908.09736v1 fatcat:kvdp2uhnknhgvj2co4qpvg6g4q

Document allocation policies for selective searching of distributed indexes

Anagha Kulkarni, Jamie Callan
2010 Proceedings of the 19th ACM international conference on Information and knowledge management - CIKM '10  
An overlap-based evaluation with an additional 1000 queries for each dataset tests and confirms the conclusions drawn using the smaller TREC query sets.  ...  For organizations with modest computational resources the high query processing cost incurred in this exhaustive search setup can be a deterrent to working with large collections.  ...  However, our results consistently demonstrated both K-means and LDA to be equally effective in defining topical shards. We chose K-means over LDA due to its lower computational cost.  ... 
doi:10.1145/1871437.1871497 dblp:conf/cikm/KulkarniC10 fatcat:mvf4qve6frbjnd2yzmuvw3okvq

A multi-regional generalized RAS updating technique

Umed Temursho, Jan Oosterhaven, M. Alejandro Cardenete
2020 figshare.com  
The technique, which we refer to as MR-GRAS, also handles non-exhaustive constraints, in which case the missing values are endogenously generated in the updating process.  ...  The framework is applicable to updating/regionalizing/balancing any partitioned matrix that needs to conform to new row sums, column sums and additional non-overlapping aggregation constraints.  ...  the previous settings due to the use of exhaustive disaggregate or/and (non-overlapping) aggregation constraints.  ... 
doi:10.6084/m9.figshare.13181858.v1 fatcat:2g5fabqtlvdn3cyrojqzjwrgf4

Numerical model to study the valve overlap period in the Wärtsilä 6L 46 four-stroke marine engine

M. Lamas, C. Rodríguez, J. Rebollido
2012 Polish Maritime Research  
The aim is to analyze the scavenging of gases, especially during the valve overlap period. Particularly, the pressure, velocity and mass fraction fields were numerically obtained.  ...  Besides, to compute the quantity of burnt gases which remain inside the cylinder, fresh charge which is expelled through the exhaust valves and study the influence of parameters such as the exhaust and  ...  Computational mesh The principle of operation of CFD codes is subdividing the domain into a number of smaller, non-overlapping subdomains. The result is a grid (or mesh) of cells (or elements).  ... 
doi:10.2478/v10012-012-0004-8 fatcat:v7heowtntjg2fhyedv247amq4q
« Previous Showing results 1 — 15 out of 139,404 results