398 Hits in 3.1 sec

Interacting with Large Distributed Datasets Using Sketch [article]

Mihai Budiu, Rebecca Isaacs, Derek Murray, Gordon Plotkin, Paul Barham, Samer Al-Kiswany, Yazan Boshmaf, Qingzhou Luo, Alexandr Andoni
2016 Eurographics Symposium on Parallel Graphics and Visualization  
Sketch applications allow interactive and responsive exploration of complex distributed datasets, scaling effectively to use large computational resources.  ...  We present Sketch, a library and a distributed runtime for building interactive tools for exploring large datasets, distributed across multiple machines.  ...  MapReduce [DG04] is adapted for complex big-data rendering in [VBS * 11], foregoing interactive response.  ... 
doi:10.2312/pgv.20161180 dblp:conf/egpgv/BudiuIMPBABLA16 fatcat:agzzn3jvlreujmdj42oye77ooe

OSoMe: the IUNI observatory on social media

Clayton A. Davis, Giovanni Luca Ciampaglia, Luca Maria Aiello, Keychul Chung, Michael D. Conover, Emilio Ferrara, Alessandro Flammini, Geoffrey C. Fox, Xiaoming Gao, Bruno Gonçalves, Przemyslaw A. Grabowicz, Kibeom Hong (+17 others)
2016 PeerJ Computer Science  
The study of social phenomena is becoming increasingly reliant on big data from online social networks.  ...  We illustrate a number of interactive open-source tools to retrieve, visualize, and analyze derived data from this collection.  ...  McDonnell Foundation (grant 220020274), the Swiss National Science Foundation (fellowship PBTIP2_142353), the Lilly Endowment, the Center for Complex Networks and Systems Research (CNetS), the Digital  ... 
doi:10.7717/peerj-cs.87 fatcat:pnvw3l4jpba4jcpobftugkjv7m

The Matthew effect in empirical data

M. Perc
2014 Journal of the Royal Society Interface  
for future research.  ...  It is closely related to the concept of preferential attachment in network science, where the more connected nodes are destined to acquire many more links in the future than the auxiliary nodes.  ...  Special thanks go to Esteban Moro for his insightful tutorial that enabled the creation of figure 2 and the pertaining videos. Funding statement.  ... 
doi:10.1098/rsif.2014.0378 pmid:24990288 pmcid:PMC4233686 fatcat:tqi2cpk45ngvnpxa23ghsiigtu

2-D Embedding of Large and High-dimensional Data with Minimal Memory and Computational Time Requirements [article]

Witold Dzwinel, Rafal Wcislo, Stan Matwin
2019 arXiv   pre-print
In the advent of big data era, interactive visualization of large data sets consisting of M*10^5+ high-dimensional feature vectors of length N (N ~ 10^3+), is an indispensable tool for data exploratory  ...  The state-of-the-art data embedding (DE) methods of N-D data into 2-D (3-D) visually perceptible space (e.g., based on t-SNE concept) are too demanding computationally to be efficiently employed for interactive  ...  i.e, nn = 3, rn = 1 and c = 0.1 (or c = 0.01), worked very well for most of data (and networks) visualized by the Authors of this paper.  ... 
arXiv:1902.01108v1 fatcat:grfv7iihvnayrdiqx7dx57tpou

Harvesting the Aggregate Computing Power of Commodity Computers for Supercomputing Applications

Dereje Regassa, Heonyoung Yeom, Yongseok Son
2022 Applied Sciences  
Most of the parallel computing researchers focused on harnessing the power of commodity processors and even internet computers to aggregate their computation powers to solve computationally complex problems  ...  Using flexible commodity cluster computers for supercomputing workloads over a dedicated supercomputer and expensive high-performance computing (HPC) infrastructure is cost-effective.  ...  Major applications of HPC are in data storage and analysis, data mining, simulation and modeling, scientific calculations, bioinformatics, big data challenges, and complex visualizations.  ... 
doi:10.3390/app12105113 fatcat:42km36vbeve23plh5e4seiw2tq

Dendritic Signal Integration [chapter]

N. Spruston
2009 Encyclopedia of Neuroscience  
A good starting point for understanding synaptic integration is the cerebellar granule cell (Figure 1(e) ). These cells are the most numerous types of neurons in Dendritic Signal Integration 445  ...  For example, a 7 mV EPSP would reduce the driving force by 10%. This is the basis for sublinear EPSP summation in a passive system.  ...  This sublinear summation is caused by the effect of membrane potential on current flow at the synapse, which is governed by the driving force for synaptic current.  ... 
doi:10.1016/b978-008045046-9.01648-x fatcat:2x7ggu4cvjhxlou4a5bn5orp3e

Elites in social networks: An axiomatic approach to power balance and Price's square root law

Chen Avin, Zvi Lotker, David Peleg, Yvonne-Anne Pignolet, Itzik Turkel, Takashi Nishikawa
2018 PLoS ONE  
In particular, we show that in social networks that respect our axioms, the size of a compact elite is sublinear in the network size.  ...  A common two-tier structure for social networks is based on partitioning society into two parts, referred to as the elite and the periphery, where the "elite" is the relatively small but well-connected  ...  by most complex networks.  ... 
doi:10.1371/journal.pone.0205820 fatcat:fywtr6wreva3lnsg2mofkxkxzu

Scalable and Sustainable Deep Learning via Randomized Hashing [article]

Ryan Spring, Anshumali Shrivastava
2016 arXiv   pre-print
We present a novel hashing based technique to drastically reduce the amount of computation needed to train and test deep networks.  ...  The matrix operations, associated with both training and testing of deep networks, are very expensive from a computational and energy standpoint.  ...  networks computationally cheap and parallelizable.  ... 
arXiv:1602.08194v2 fatcat:fo2pjpzsivgzxnmxurmlalfvda

Searching the World Wide Web: Challenges and Partial Solutions [chapter]

Ricardo A. Baeza-Yates
1998 Lecture Notes in Computer Science  
In this article we analyze the problem of searching the WWW, giving some insight and models to understand its complexity. Then we survey the two main current techniques used to search the WWW.  ...  These computers are interconnected with no predefined topology and with very different bandwiths. ¡ High percentage of volatile data: due to Internet dynamics, new computers and data can be added or removed  ...  In fact, in [11] we show that for searching words with errors we can have sublinear space and search time simultaneously.  ... 
doi:10.1007/3-540-49795-1_4 fatcat:dft4bf3lxvfvhhl7suqetnkuve

Semi-supervised hashing for scalable image retrieval

Jun Wang, Sanjiv Kumar, Shih-Fu Chang
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Zhenguo Li for his valuable comments. J. Wang was supported in part by Google Intern Scholarship. S.-F. Chang is supported in part by National Science Foundation Award CNS-07-51078.  ...  In terms of the query time, RBMs also need about 10 times more time to compute the binary codes through the trained neural network.  ...  For large-scale applications, to avoid excessive computational and memory costs, one would like to instead do an Approximate Nearest Neighbor (ANN) search with sublinear query complexity [12] .  ... 
doi:10.1109/cvpr.2010.5539994 dblp:conf/cvpr/WangKC10 fatcat:moyggvoinncffgd5be7ylfi4dm

Intractability and approximation of optimization theories of cognition

Iris van Rooij, Todd Wareham
2012 Journal of Mathematical Psychology  
Many computational-or rational-level theories of human cognition suffer from computational intractability: the postulated optimization functions are impossible to compute in a reasonable time by a finite  ...  We illustrate the use of our framework for a particular intractable cognitive theory: i.e., Thagard and Verbeurgt's (1998) Coherence model, known to be equivalent to harmony maximization in Hopfield networks  ...  Antonina Kolokolova for bringing Kumar and Sivakumar (1999) and Feige et al. (2000) to our attention.  ... 
doi:10.1016/ fatcat:j24n5mmz2nc4xcp66iuo4vvo44

A Survey on Big IoT Data Indexing: Potential Solutions, Recent Advancements, and Open Issues

Zineddine Kouahla, Ala-Eddine Benrazek, Mohamed Amine Ferrag, Brahim Farou, Hamid Seridi, Muhammet Kurulay, Adeel Anjum, Alia Asheralieva
2021 Future Internet  
The real-world applications of the existing indexing techniques in different areas, such as health, business, scientific experiments, and social networks, are presented.  ...  However, efficient retrieval and management of such information in terms of index size and search time require optimization of indexing schemes which is rather difficult to implement.  ...  Thus, several challenging areas of research can serve as a basis for possible future research directions for the indexing of large IoT data.  ... 
doi:10.3390/fi14010019 fatcat:xnlzg7cs2fb3lgng65ha5ucf5m

Factors of Transferability for a Generic ConvNet Representation [article]

Hossein Azizpour, Ali Sharif Razavian, Josephine Sullivan, Atsuto Maki, Stefan Carlsson
2015 arXiv   pre-print
Evidence is mounting that Convolutional Networks (ConvNets) are the most effective representation learning method for visual recognition tasks.  ...  Recent studies have shown this form of representation transfer to be suitable for a wide range of target visual recognition tasks.  ...  Since the Tiny network has 10 times fewer parameters than OverFeat while preserving most of the performance, it could be useful for scenarios where real-time computation is an issue. with 1000 semantic  ... 
arXiv:1406.5774v3 fatcat:he252tis7faqjbxkg7y5gqicmy

Foundations of Data Management (Dagstuhl Perspectives Workshop 16151)

Marcelo Arenas, Richard Hull, Wim Marten, Tova Milo, Thomas Schwentick, Marc Herbstritt
2016 Dagstuhl Reports  
The workshop brought together specialists from the existing database theory community, and from adjoining areas, particularly from various subdisciplines within the Big Data community, to understand the  ...  Big Data is not the only force that is driving expansion and transformation for the Foundations of Data Management.  ...  Sublinear-time algorithms have been extensively studied in the algorithms community, but not fully exploited yet in DB.  ... 
doi:10.4230/dagrep.6.4.39 dblp:journals/dagstuhl-reports/ArenasHMMS16 fatcat:gpdqvqbhr5futlluo3qqdvg5hu

Drawbacks and Proposed Solutions for Real-time Processing on Existing State-of-the-art Locality Sensitive Hashing Techniques [article]

Omid Jafari, Khandker Mushfiqul Islam, Parth Nagarkar
2019 arXiv   pre-print
Locality Sensitive Hashing (LSH) is a very popular and efficient approximate nearest neighbor technique that is known for its sublinear query processing complexity and theoretical guarantees.  ...  In this paper, we discuss the challenges and drawbacks of existing LSH techniques for processing real-time high-dimensional image data.  ...  For the Sift dataset, which is one of the common datasets in Big Data researches, the query time of QALSH is about 50% times more than the query time in C2LSH.  ... 
arXiv:1912.07091v1 fatcat:56i35cxw45ccxmgm64jpdxwiru
« Previous Showing results 1 — 15 out of 398 results