Filters








4,718 Hits in 9.5 sec

Explanation vs Performance in Data Mining: A Case Study with Predicting Runaway Projects

Tim MENZIES, Osamu MIZUNO, Yasunari TAKAGI, Tohru KIKUNO
2009 Journal of Software Engineering and Applications  
In the case of predicting runaway software projects, we show that the twin goals of high performance and good explanatory power are achievable after applying a variety of data mining techniques (discrimination  ...  Often, the explanatory power of a learned model must be traded off against model performance.  ...  Learning Latent Features Numerous data mining methods check if the available features can be combined in useful ways. In this way, latent features within a data set can be discovered.  ... 
doi:10.4236/jsea.2009.24030 fatcat:qrafpsm7p5cldkwdgpp4pzwvsm

Language Models as Knowledge Bases? [article]

Fabio Petroni, Tim Rocktäschel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander H. Miller, Sebastian Riedel
2019 arXiv   pre-print
Language models have many advantages over structured knowledge bases: they require no schema engineering, allow practitioners to query about an open class of relations, are easy to extend to more data,  ...  and require no human supervision to train.  ...  Acknowledgments We would like to thank the reviewers for their thoughtful comments and efforts towards improving our manuscript.  ... 
arXiv:1909.01066v2 fatcat:vxq4o7qx5fgt3jha4njhpjrwqu

Language Models as Knowledge Bases?

Fabio Petroni, Tim Rocktäschel, Sebastian Riedel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander Miller
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
Language models have many advantages over structured knowledge bases: they require no schema engineering, allow practitioners to query about an open class of relations, are easy to extend to more data,  ...  and require no human supervision to train.  ...  Acknowledgments We would like to thank the reviewers for their thoughtful comments and efforts towards improving our manuscript.  ... 
doi:10.18653/v1/d19-1250 dblp:conf/emnlp/PetroniRRLBWM19 fatcat:aalqzrmjf5gmjg6l2imffkbgky

Sentiment Analysis of Political Tweets From the 2019 Spanish Elections

Margarita Rodriguez-Ibanez, Francisco-Javier Gimeno-Blanes, Pedro Manuel Cuenca-Jimenez, Cristina Soguero-Ruiz, Jose Luis Rojo-Alvarez
2021 IEEE Access  
In this paper, we apply a set of basic methods to analyze the statistical and temporal dynamics of sentiment analysis on political campaigns and assess their scope and limitations.  ...  We then followed a twofold analysis strategy: (1) statistical characterization using indices derived from well-known temporal and information metrics and methods -including entropy, mutual information,  ...  of the embeddings to obtain a more detailed and complete description perspective on the data.  ... 
doi:10.1109/access.2021.3097492 fatcat:z6nfftzvibh5lgkma2mrtpsueq

Lossless Compression with Latent Variable Models [article]

James Townsend
2021 arXiv   pre-print
We then make use of a novel empirical insight, that fully convolutional generative models, trained on small images, are able to generalize to images of arbitrary size, and extend BB-ANS to hierarchical  ...  We develop a simple and elegant method for lossless compression using latent variable models, which we call 'bits back with asymmetric numeral systems' (BB-ANS).  ...  Applications to lossless compression were less well covered in works published prior to Townsend et al. (2019) , upon which Chapter 3 is based.  ... 
arXiv:2104.10544v2 fatcat:ndur24ecsbfxxjholb6aiakko4

A Review of the Asymmetric Numeral System and Its Applications to Digital Images

Ping Ang Hsieh, Ja-Ling Wu
2022 Entropy  
Therefore, we think a thorough overview of ANS is beneficial, and this idea brings our contributions to the first part of this work.  ...  Combining these two characteristics helps process digital images, e.g., art collection images and medical images, to achieve compression and encryption simultaneously.  ...  Similarly, it is easy to find that the encoding functions C(a, x) = x p a = 8 5 x, C(b, x) = x p b = 8 2 x, and C(c, x) = x p c = 8x do not work well.  ... 
doi:10.3390/e24030375 pmid:35327886 pmcid:PMC8946946 fatcat:jjshclxz55ganivuqufu2m65oa

Risk Assessment of Public Safety and Security Mobile Service

Matti J. Peltola, Pekka Kekolahti
2015 2015 10th International Conference on Availability, Reliability and Security  
A latent (Naïve) variable model (a), Extension to BN: Hierarchical latent (Naïve) variable model (b), Two-slice temporal BN (2-TBN) (c), Influence diagram, where X5 is a decision and X10 a utility node  ...  In the simplest form the PSEM is a hierarchical latent Naïve variable model (Figure 3b ) but also other than Naïve structures can be used between latent variables.  ... 
doi:10.1109/ares.2015.65 dblp:conf/IEEEares/PeltolaK15 fatcat:iqzzjbyqrzbmfanhry2rj742pi

Completion Reasoning Emulation for the Description Logic EL+ [article]

Aaron Eberhart, Monireh Ebrahimi, Lu Zhou, Cogan Shimizu, Pascal Hitzler
2019 arXiv   pre-print
We present a new approach to integrating deep learning with knowledge-based systems that we believe shows promise.  ...  We also show that this trained system is resistant to noise by corrupting a percentage of the test data and comparing the reasoner's and LSTM's predictions on corrupt data with correct answers.  ...  Testing Environment All testing was done on a computer running Ubuntu 19.10 64-bit with an Intel Core i7-9700K CPU@3.60GHz x 8, 47.1 GiB DDR4, and a GeForce GTX 1060 6GB/PCIe/SSE2.  ... 
arXiv:1912.05063v1 fatcat:bcddshwbqbc2xl3g3a3wxfnwgq

Great Expectations: Unsupervised Inference of Suspense, Surprise and Salience in Storytelling [article]

David Wilmot
2022 arXiv   pre-print
Narrative theory methods (rules and procedures) are applied to the knowledge built into deep learning models to directly infer salience, surprise, and salience in stories.  ...  Extensions add memory and external knowledge from story plots and from Wikipedia to infer salience on novels such as Great Expectations and plays such as Macbeth.  ...  Knowledge salience is the difference between an expert-informed reader versus a naive one by taking the difference between the average log-likelihood of a base LM and an LM enriched with memory and a KB  ... 
arXiv:2206.09708v1 fatcat:k4oefywyxvgn5gdtedyvr5mbpi

Labels or attributes?

Luke K. McDowell, David W. Aha
2013 Proceedings of the 22nd ACM international conference on Conference on information & knowledge management - CIKM '13  
Recent work has focused on option (a), because early work showed it was more accurate and because option (b) fit poorly with discriminative classifiers.  ...  to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of  ...  To our knowledge, no prior work has reported this effect. Why does it occur?  ... 
doi:10.1145/2505515.2505628 dblp:conf/cikm/McDowellA13 fatcat:gh63u64cczbyxfdpibfyz5tuhq

The resource-based view and marketing: The role of market-based assets in gaining competitive advantage

R Srivastava
2001 Journal of Management  
Finally, the article posits a set of research directions designed to enable scholars to further advance the integration of RBV and marketing from both theory-driven practice management as well as a problem-driven  ...  This article posits a framework that shows how market-based assets and capabilities are leveraged via market-facing or core business processes to deliver superior customer value and competitive advantages  ...  too naïve, simplistic and static to handle the exigencies of an unfolding future.  ... 
doi:10.1016/s0149-2063(01)00123-4 fatcat:tifvvaabjbduvnvxs24bvwgybm

The resource-based view and marketing: The role of market-based assets in gaining competitive advantage

Rajendra K. Srivastava, Liam Fahey, H. Kurt Christensen
2001 Journal of Management  
Finally, the article posits a set of research directions designed to enable scholars to further advance the integration of RBV and marketing from both theory-driven practice management as well as a problem-driven  ...  This article posits a framework that shows how market-based assets and capabilities are leveraged via market-facing or core business processes to deliver superior customer value and competitive advantages  ...  too naïve, simplistic and static to handle the exigencies of an unfolding future.  ... 
doi:10.1177/014920630102700610 fatcat:b3hriojc2nev5f3mnhj6wieeay

The planner's subjective destitution: towards a hysterical-analytical triad of planning theory-research-practice

Ignacio Castillo Ulloa
2019 Raumforschung und Raumordnung | Spatial Research and Planning  
and "why I am doing this or that?"  ...  Hence, the attempt here is to look in more depth at the 'ambivalent' role of the planner as well as to bring in 'planning research', as a key, somewhat occluded, element within the discussion on bridging  ...  We acknowledge support by the German Research Foundation and the Open Access Publication Fund of TU Berlin.  ... 
doi:10.2478/rara-2019-0009 fatcat:ojr5kfhigba67joutsnr36y3za

Social Link Prediction in Online Social Tagging Systems

Charalampos Chelmis, Viktor K. Prasanna
2013 ACM Transactions on Information Systems  
In this article, we propose latent topic models as a principled way of reducing the dimensionality of such data and capturing the dynamics of collaborative annotation process.  ...  Social networks have become a popular medium for people to communicate and distribute ideas, content, news and advertisements.  ...  ACKNOWLEDGMENT This work is supported by Chevron Corp. under the joint project, Center for Interactive Smart Oilfield Technologies (CiSoft), at the University of Southern California.  ... 
doi:10.1145/2516891 fatcat:mticr2ax4bdffp6hpt6ty632ri

A survey on data‐efficient algorithms in big data era

Amina Adadi
2021 Journal of Big Data  
from rich-data domains into poor-data domains, or by (iv) altering data-hungry algorithms to reduce their dependency upon the amount of samples, in a way they can perform well in small samples regime.  ...  Unfortunately, many application domains do not have access to big data because acquiring data involves a process that is expensive or time-consuming.  ...  [90] provided a complete overview of approaches for making graph-based methods more scalable.  ... 
doi:10.1186/s40537-021-00419-9 fatcat:v4uahsvhlzdldlxqf24bshmja4
« Previous Showing results 1 — 15 out of 4,718 results