1,404,315 Hits in 8.5 sec

A two-level approach to making class predictions

A. Costea, T. Eklund
2003 36th Annual Hawaii International Conference on System Sciences, 2003. Proceedings of the  
In this paper we propose a new two-level methodology for assessing countries'/companies' economic/financial performance.  ...  Furthermore, we focus our efforts on understanding the decision process corresponding to the two predictive models.  ...  Acknowledgements The authors would like to thank Professor Barbro Back for her constructive comments on the article.  ... 
doi:10.1109/hicss.2003.1174207 dblp:conf/hicss/CosteaE03 fatcat:cxwa6zx4cfex5kbjikro3fajky

Revisiting the CompCars Dataset for Hierarchical Car Classification: New Annotations, Experiments, and Results

Marco Buzzelli, Luca Segantin
2021 Sensors  
To evaluate this revisited dataset, we design and implement three different approaches to car classification, two of which exploit the hierarchical nature of car annotations.  ...  We address the task of classifying car images at multiple levels of detail, ranging from the top-level car type, down to the specific car make, model, and year.  ...  -all approach directly predicts the finest-level class detail, the two-step cascade introduces a first high-level classifier to narrow the finest-level search domain, and the hierarchical multilabel performs  ... 
doi:10.3390/s21020596 pmid:33467700 pmcid:PMC7830427 fatcat:pca6u6t4qnf77n5j2nn3z4ahbe

Integrating Propositional and Relational Label Side Information for Hierarchical Zero-Shot Image Classification [article]

Colin Samplawski, Heesung Kwon, Erik Learned-Miller, Benjamin M. Marlin
2019 arXiv   pre-print
We propose benchmark tasks for this framework that focus on making predictions across a range of semantic levels.  ...  It enables predicting that images belong to classes for which no labeled training instances are available.  ...  Note that the lifted zeroshot models make their initial prediction f (x) over all the classes in H which fall under a higher-level class.  ... 
arXiv:1902.05492v1 fatcat:q2e2bushafdxpnjgsn2xqlq7da

Confirmation Rule Sets [chapter]

Dragan Gamberger, Nada Lavrač
2000 Lecture Notes in Computer Science  
The concept of confirmation rule sets represents a framework for reliable decision making that combines two principles that are effective for increasing the predictive accuracy: consensus in an ensemble  ...  The confirmation rules concept uses a separate classifier set for every class of the domain.  ...  supported in part by Croatian Ministry of Science and Technology, Slovenian Ministry of Science and Technology, and the EU funded project Data Mining and Decision Support for Business Competitiveness: A  ... 
doi:10.1007/3-540-45372-5_4 fatcat:rr2ri4m4zvc7jevkxd6k3ylwou

IMPEX: An Approach to Analyze Source Code Changes on Software Run Behavior

David Nemer
2013 Journal of Software Engineering and Applications  
The approach taken in this paper was able to precisely predict what would be impacted on the software execution when a change in the source code was made in 70% of the cases.  ...  Developers may have a hard time knowing what could happen to the software when making changes.  ...  So the prediction model is produced to make predictions in three levels: method, class and package.  ... 
doi:10.4236/jsea.2013.64020 fatcat:v5kg57h2pvfx7kwgx73zxscsae

A Global-Model Naive Bayes Approach to the Hierarchical Prediction of Protein Functions

Carlos N. Silla Jr., Alex A. Freitas
2009 2009 Ninth IEEE International Conference on Data Mining  
In this paper we propose a new global-model approach for hierarchical classification, where a single global classification model is built by considering all the classes in the hierarchy -rather than building  ...  The achieved results are positive and show that the proposed global model is better than using a local model approach.  ...  ACKNOWLEDGMENT We want to thank Dr. Nick Holden for kindly providing us with the datasets used in this experiments.  ... 
doi:10.1109/icdm.2009.85 dblp:conf/icdm/SillaF09 fatcat:qn3kwx77rvbxdnbdz5l2q6evgu

A General Bayesian Network-Assisted Ensemble System for Context Prediction: An Emphasis on Location Prediction [chapter]

Kun Chang Lee, Heeryon Cho
2010 Lecture Notes in Computer Science  
To leverage the power of both the GBN and the ensemble system, we propose a GBN-assisted ensemble system for location prediction.  ...  The proposed system was applied to a real-world location prediction dataset, and promising results were obtained. Practical implications are discussed.  ...  When a new instance is tested, each base classifier makes a prediction, and the final prediction is derived from the base classifiers that are predicted to be correct by the metaclassification schemes.  ... 
doi:10.1007/978-3-642-17569-5_29 fatcat:emedwq2xxfexpfowtmns5wx4ke

Prediction-Based Audiovisual Fusion for Classification of Non-Linguistic Vocalisations

2016 IEEE Transactions on Affective Computing  
In virtually all cases prediction-based audiovisual fusion consistently outperforms the two most commonly used fusion approaches, decision-level and feature-level fusion. ! 1.  ...  Prediction plays a key role in recent computational models of the brain and it has been suggested that the brain constantly makes multisensory spatiotemporal predictions.  ...  This example corresponds to a two-class problem and that is why there are two error curves in each module. level fusion.  ... 
doi:10.1109/taffc.2015.2446462 fatcat:2xtesk55lfewnn66jzo6fu45q4

Improving Peer Feedback Prediction: The Sentence Level is Right

Huy Nguyen, Diane Litman
2014 Proceedings of the Ninth Workshop on Innovative Use of NLP for Building Educational Applications  
We propose to perform prediction at the sentence level, even though the educational task is to label feedback at a multi-sentential comment level.  ...  We first introduce a corpus annotated at a sentence level granularity, then build comment prediction models using this corpus.  ...  We are grateful to our colleagues for sharing the data.  ... 
doi:10.3115/v1/w14-1812 dblp:conf/bea/NguyenL14 fatcat:sl22hyjeizewdg66hwnux6aw4y

Replaying IDE interactions to evaluate and improve change prediction approaches

Romain Robbes, Damien Pollet, Michele Lanza
2010 2010 7th IEEE Working Conference on Mining Software Repositories (MSR 2010)  
They are thus not a valid basis for evaluation in the case of developmentstyle prediction, where the order of the predictions has to match the order of the changes a developer makes.  ...  Moreover, the change prediction approaches themselves can use the more accurate data to fine-tune their prediction.  ...  It is applicable both at the class level (predicting changes to classes) as a comparison point with the previous evaluations, and at the method level.  ... 
doi:10.1109/msr.2010.5463278 dblp:conf/msr/RobbesPL10 fatcat:c7f5b2d34fh7jf7shqy5x7w4wm

A Change Impact Analysis Tool for Software Development Phase

Sufyan Basri, Nazri Kama, Roslina Ibrahim, Saiful Adli Ismail
2015 International Journal of Software Engineering and Its Applications  
This is due to some classes in the class artifact are still under development or partially developed.  ...  Software project management might use a reliable estimation on potential impacted artifacts to decide whether to accept or reject the changes.  ...  Acknowledgements We would like to thank Universiti Teknologi Malaysia (UTM) and Ministry of Higher Education (MoHE) Malaysia for their financial support under the Vote No 4L067.  ... 
doi:10.14257/ijseia.2015.9.9.21 fatcat:57tlnuz2ojaknozu6ug5ba7pgy

Leveraging Declarative Knowledge in Text and First-Order Logic for Fine-Grained Propaganda Detection [article]

Ruize Wang, Duyu Tang, Nan Duan, Wanjun Zhong, Zhongyu Wei, Xuanjing Huang, Daxin Jiang, Ming Zhou
2020 arXiv   pre-print
Experiments show that our method achieves superior performance, demonstrating that leveraging declarative knowledge can help the model to make more accurate predictions.  ...  The latter refers to the literal definition of each propaganda technique, which is utilized to get class representations for regularizing the model parameters.  ...  A BERT-based multi-task learning approach is adopted to make predictions for 18 propaganda techniques at both sentence level and token level.  ... 
arXiv:2004.14201v2 fatcat:vfscs2s2vzb35o3gwlkqq6kwfu

Explainable AI: A Hybrid Approach to Generate Human-Interpretable Explanation for Deep Learning Prediction

Tanusree De, Prasenjit Giri, Ahmeduvesh Mevawala, Ramyasri Nemani, Arati Deo
2020 Procedia Computer Science  
However, to trust an AI model prediction or to take downstream action based on a prediction outcome, one needs to understand the reasons for the prediction.  ...  We propose a hybrid of two prior approaches, integrating clustering of the network's hidden layer representation [2] with TREPAN decision tree [1], both of which uniquely deconstruct a neural network.  ...  To improvise the approach, we present here a new integrated approach to generate explanations for a predicted outcome at an instance level.  ... 
doi:10.1016/j.procs.2020.02.255 fatcat:tbt6gorfcrcwlfx5rhdyufz4je

Deepenz: prediction of enzyme classification by deep learning

Hamza Chehili, Salah Eddine Aliouane, Abdelhafedh Bendahmane, Mohamed Abdelhafid Hamidechi
2021 Indonesian Journal of Electrical Engineering and Computer Science  
The combination of these two tools give a model with a great capacity to extract knowledge from enzyme data to predict and classify them.  ...  Their goal is to increase the speed of processing and to improve the accuracy of predictions. The Purpose of this work is to develop an approach that predicts the enzymes' classification.  ...  The result obtained in level 1, that is common in the two approaches, shows that DeepEnz gives a prediction accuracy of 97.69% (≈ 98 %) while UDSMprot obtains 97% (Table 3) .  ... 
doi:10.11591/ijeecs.v22.i2.pp1108-1115 fatcat:rtuz3plmanckth72fcsuec7nkq

Predicting with confidence: Using conformal prediction in drug discovery

Jonathan Alvarsson, Staffan Arvidsson McShane, Ulf Norinder, Ola Spjuth
2020 Journal of Pharmaceutical Sciences  
For regression, a prediction interval consists of an upper and a lower bound. For classification, a prediction interval is a set that contains none, one, or many of the potential classes.  ...  The size of the prediction interval is affected by a user-specified confidence/significance level, and by the nonconformity of the predicted object; i.e., the strangeness as defined by a nonconformity  ...  Approaches to Conformal Prediction The two most widely used approaches for conformal prediction are inductive conformal prediction (ICP) and transductive conformal prediction (TCP).  ... 
doi:10.1016/j.xphs.2020.09.055 pmid:33075380 fatcat:o5gigc6plffm7n7nxxvnsirwri
« Previous Showing results 1 — 15 out of 1,404,315 results