Filters








5,074 Hits in 5.9 sec

Variance maximization via noise injection for active sampling in learning to rank

Wenbin Cai, Ya Zhang
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
Experimental results on both the LETOR 4.0 dataset and a real-world Web search ranking dataset have demonstrated the effectiveness of the proposed active learning approach.  ...  In this paper, we propose a general active learning for ranking strategy called Variance Maximization (VM).  ...  We propose a general active learning strategy for Web search ranking called Variance Maximization (VM).  ... 
doi:10.1145/2396761.2398522 dblp:conf/cikm/CaiZ12 fatcat:pkjuojyzezfbdfiglefhnuvfwy

Deep Learning Models of the Retinal Response to Natural Scenes

Lane T McIntosh, Niru Maheswaranathan, Aran Nayebi, Surya Ganguli, Stephen A Baccus
2016 Advances in Neural Information Processing Systems  
An examination of the learned CNNs reveals several properties. First, a richer set of feature maps is necessary for predicting the responses to natural scenes compared to white noise.  ...  Third, the injection of latent noise sources in intermediate layers enables our model to capture the sub-Poisson spiking variability observed in retinal ganglion cells.  ...  Acknowledgments The authors would like to thank Ben Poole and EJ Chichilnisky for helpful discussions related to this work.  ... 
pmid:28729779 pmcid:PMC5515384 fatcat:6rdirhmmgjbf7isreamk523jae

Plausible Deniability in Web Search -- From Detection to Assessment [article]

Pol Mac Aonghusa, Douglas J. Leith
2017 arXiv   pre-print
We show this remains the case when attempting to disrupt search engine learning through noise query injection and click obfuscation.  ...  We ask how to defend user ability to plausibly deny their interest in topics deemed sensitive in the face of search engine learning.  ...  Plausible deniability as a privacy defence for web search is addressed in the literature.  ... 
arXiv:1703.03471v2 fatcat:qif5xpp2sfgtjmoeaqdm26ads4

Text selections as implicit relevance feedback

Ryen W. White, Georg Buscher
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
Users' search activity has been used as implicit feedback to model search interests and improve the performance of search systems.  ...  However, richer data on how people engage with search results can now be captured at scale, creating new opportunities to enhance search.  ...  INTRODUCTION Searchers express their needs to search engines via query statements.  ... 
doi:10.1145/2348283.2348514 dblp:conf/sigir/WhiteB12 fatcat:6e3lopz5szbyvhxbtzutr4rwaq

Adaptive educational hypermedia on the web

Paul De Bra
2002 Communications of the ACM  
On this basis, matching the learner's learning style with the web page features and mining specific attributes is more desirable.  ...  The primary contribution of this research work is to introduce a feature-based recommender system that delivers educational web pages according to the user's individual learning style.  ...  This learning style model is used by another web site that we developed to search against a user's query, for OCW pages, matching the student's style.  ... 
doi:10.1145/506218.506247 fatcat:twxejwgbt5fqxikhzbtgnxdwbi

LIGO Detector Characterization in the Second and Third Observing Runs [article]

D. Davis, J. S. Areeda, B. K. Berger, R. Bruntz, A. Effler, R. C. Essick, R. P. Fisher, P. Godwin, E. Goetz, A. F. Helmling-Cornell, B. Hughey, E. Katsavounidis (+275 others)
2021 arXiv   pre-print
The characterization of the Advanced LIGO detectors in the second and third observing runs has increased the sensitivity of the instruments, allowing for a higher number of detectable gravitational-wave  ...  Finally, we explain anticipated changes in the role of detector characterization and current efforts to prepare for the high rate of gravitational-wave alerts and events in future observing runs.  ...  Additional support for Advanced LIGO was provided by the Australian Research Council.  ... 
arXiv:2101.11673v2 fatcat:mbfs5idgyvfijnxokodv4cybb4

Surfer100: Generating Surveys From Web Resources, Wikipedia-style [article]

Irene Li, Alexander Fabbri, Rina Kawamura, Yixin Liu, Xiangru Tang, Jaesung Tae, Chang Shen, Sally Ma, Tomoe Mizutani, Dragomir Radev
2022 arXiv   pre-print
This is the first study on utilizing web resources for long Wikipedia-style summaries to the best of our knowledge.  ...  We show that recent advances in pretrained language modeling can be combined for a two-stage extractive and abstractive approach for Wikipedia lead paragraph generation.  ...  They proposed to inject noise into each layer of the Network before calculating the subsequent layer.  ... 
arXiv:2112.06377v3 fatcat:tbmlxs2d4jb25plfjslelby5oq

Searching for software learning resources using application context

Michael Ekstrand, Wei Li, Tovi Grossman, Justin Matejka, George Fitzmaurice
2011 Proceedings of the 24th annual ACM symposium on User interface software and technology - UIST '11  
web search system, and describe a prototype implementation and initial user study of this system.  ...  We conclude with a discussion of open issues and an agenda for further research.  ...  Users typically access computerized help resources via text queries issued web search engines or the search facility of a program's help browser [12] , or via in-program help inter-faces and agents [  ... 
doi:10.1145/2047196.2047220 dblp:conf/uist/EkstrandLGMF11 fatcat:ygddaz2uufd75cijs6ggyxivnu

Why label when you can search?

Josh Attenberg, Foster Provost
2010 Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '10  
, such as active learning, can be ineffective.  ...  We show that under extreme skew, even basic techniques for guided learning completely dominate smart (active) strategies for applying human resources to select cases for labeling.  ...  learning; they are supported by systems for web-page labeling and for web search.  ... 
doi:10.1145/1835804.1835859 dblp:conf/kdd/AttenbergP10 fatcat:3v6v2ojhozagrgh7hhjqvix23q

Learning Parameters in Entity Relationship Graphs from Ranking Preferences [chapter]

Soumen Chakrabarti, Alekh Agarwal
2006 Lecture Notes in Computer Science  
Extending from vector-space information retrieval, we wish to automatically learn ranking function for searching such typed graphs.  ...  We present a unified model for ranking in ER graphs, and propose an algorithm to learn the parameters of the model.  ...  In Section 2 we will review a number of efforts to learn some of the parameters of the Markov walk system, most typically via heuristic search [17] , quadratic programs [20] or local hill-climbing  ... 
doi:10.1007/11871637_13 fatcat:f6a37epesng7pa6fsyuwtus54i

PIXS

Pili Hu, Junbo Li, Wing Cheong Lau
2013 Proceedings of the 5th ACM workshop on HotPlanet - HotPlanet '13  
different platforms via a Rank Preserving Regression (RPR) framework.  ...  As a proof-of-concept to demonstrate the flexibility and effectiveness of PIXS, we have developed for it a Python-based semi-supervised learning application which can prioritize incoming messages from  ...  learning via the Stochastic Gradient Descent (SGD) approach.  ... 
doi:10.1145/2491159.2491167 dblp:conf/sigcomm/HuLL13 fatcat:x2nsfz5s7rexxmkr5tu3ea7kuu

An Insight on Social Media Stream Mining

Rojina Deuja, Krishna Bikram Shah
2019 SCITECH Nepal  
While social media streaming has received a lot of attention over the past decade, the ever-expanding streams of data presents huge challenges for learning and maintaining control.  ...  Here, a tool called the Noise Web Data Learning (NWDL) -capable of learning noisy web data in user profi les was presented.  ...  A comparison of users' search behaviours between Web search and Twitter search was done in [10] while the difference between traditional search engines and social media search associated with health  ... 
doi:10.3126/scitech.v14i1.25532 fatcat:i2focs3ijjamxbhu6zyfgupl7i

Draco: Statistical diagnosis of chronic problems in large distributed systems

Soila P. Kavulya, Scott Daniels, Kaustubh Joshi, Matti Hiltunen, Rajeev Gandhi, Priya Narasimhan
2012 IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2012)  
days or weeks, and coexist with other problems active at the same time.  ...  We have deployed Draco in production for the VoIP operations of a major ISP.  ...  They are likely to be applicable to many other large platforms (e.g., e-commerce, web-search, social networks) that serve users via independent interactions such as web requests. 1 Draco is a genus of  ... 
doi:10.1109/dsn.2012.6263927 dblp:conf/dsn/KavulyaDJHGN12 fatcat:apesv3nkofcpxltytouzdt3l4m

Implicit Predictive Indicators: Mouse Activity and Dwell Time [chapter]

Stephen Akuma, Chrisina Jayne, Rahat Iqbal, Faiyaz Doctor
2014 IFIP Advances in Information and Communication Technology  
Humans vary in their learning behaviour. It is difficult to predict the actual needs of learners through their search activity.  ...  This research is a preliminary study to examine the predictive strength of some implicit indicators on web documents.  ...  The software was injected in 15 web documents to record users' mouse activity, dwell time and explicit rating.  ... 
doi:10.1007/978-3-662-44654-6_16 fatcat:ovznjwthh5dofkletqot4ekgsi

Social Search [chapter]

Peter Brusilovsky, Barry Smyth, Bracha Shapira
2018 Lecture Notes in Computer Science  
Modern web search engines have evolved from their roots in information retrieval to developing new ways to cope with the unique nature of web search.  ...  In this chapter, we review recent research that aims to make search a more social activity by combining readily available social signals with various strategies for using these signals to influence or  ...  Once learned from data, click models can produce a data-informed ranking of web pages for a given query. The idea of learning a model of user search behavior was introduced by Agichtein et al. [6] .  ... 
doi:10.1007/978-3-319-90092-6_7 fatcat:ro6hjgwhgvgwrncakrpfv5brye
« Previous Showing results 1 — 15 out of 5,074 results