Context-aware feature query to improve the prediction performance

Kachuee, Mohammad; Hosseini, Anahita; Moatamed, Babak; Darabi, Sajad; Sarrafzadeh, Majid

doi:10.1109/globalsip.2017.8309078

Cited by 4 publications

(16 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the devised method uses backpropagation of gradients and binary representation layers in neural networks to address the computational load as well as scalability concerns. In an earlier work [22], we introduced the idea of sensitivity analysis as a method for dynamic feature selection. However, in this paper, we extend the idea by considering feature acquisition costs, introducing improvements such as feature encoding, and conducting more detailed experiments and analysis.…”

Section: Related Workmentioning

confidence: 99%

“…Additionally, in this table, the area under the accuracy-cost curves (AUACC) of the proposed feature query method as well as the AUACC of randomly asking for the unknown features are presented. We have also included AUACC results of a cost-aware version of the method suggested in [22] in which sensitivities are normalized by feature costs (see the DPFQ column). It is worth mentioning that AUACC values are calculated as the normalized area under the accuracy versus the acquisition cost curve from a cost of zero to the cost at which the accuracy converges to the maximum accuracy.…”

Section: B Performance Evaluationmentioning

confidence: 99%

“…2 AUACC is defined as the area under the accuracy and the feature acquisition cost curve. 3 Cost-aware version of the method suggested in [22].…”

Section: B Performance Evaluationmentioning

confidence: 99%

See 2 more Smart Citations

Dynamic Feature Acquisition Using Denoising Autoencoders

Kachuee

Darabi

Moatamed

et al. 2019

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

In real-world scenarios, different features have different acquisition costs at test-time which necessitates cost-aware methods to optimize the cost and performance trade-off. This paper introduces a novel and scalable approach for cost-aware feature acquisition at test-time. The method incrementally asks for features based on the available context that are known feature values. The proposed method is based on sensitivity analysis in neural networks and density estimation using denoising autoencoders with binary representation layers. In the proposed architecture, a denoising autoencoder is used to handle unknown features (i.e., features that are yet to be acquired), and the sensitivity of predictions with respect to each unknown feature is used as a context-dependent measure of informativeness. We evaluated the proposed method on eight different real-world datasets as well as one synthesized dataset and compared its performance with several other approaches in the literature. According to the results, the suggested method is capable of efficiently acquiring features at test-time in a cost-and contextaware fashion.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: B Performance Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Feature Acquisition Using Denoising Autoencoders

Kachuee

Darabi

Moatamed

et al. 2019

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…As an alternative approach, sensitivity analysis of trained predictors is suggested to measure the importance of each feature given a context (Early et al, 2016a;Kachuee et al, 2017;. These approaches either require an exhaustive measurement of sensitivities or rely on approximations of sensitivity.…”

Section: Introductionmentioning

confidence: 99%

Opportunistic Learning: Budgeted Cost-Sensitive Learning from Data Streams

Kachuee,

Goldstein,

Karkkainen

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

In many real-world learning scenarios, features are only acquirable at a cost constrained under a budget. In this paper, we propose a novel approach for costsensitive feature acquisition at the prediction-time. The suggested method acquires features incrementally based on a context-aware feature-value function. We formulate the problem in the reinforcement learning paradigm, and introduce a reward function based on the utility of each feature. Specifically, MC dropout sampling is used to measure expected variations of the model uncertainty which is used as a feature-value function. Furthermore, we suggest sharing representations between the class predictor and value function estimator networks. The suggested approach is completely online and is readily applicable to stream learning setups. The solution is evaluated on three different datasets including the well-known MNIST dataset as a benchmark as well as two cost-sensitive datasets: Yahoo Learning to Rank and a dataset in the medical domain for diabetes classification. According to the results, the proposed method is able to efficiently acquire features and make accurate predictions.

show abstract

“…With the result, the hidden layer learns to build a smaller representation of the input. Examples of IoT use-cases that utilized AEs include human activity recognition [113], privacy preservation in sensor data analytics [114], prediction performance improvement in sensor and wearable systems [115], botnet traffic detection [116], fault diagnosis [117], etc.…”

mentioning

confidence: 99%

Data Analytics in the Internet of Things: A Survey

Saleem

Chishti

2019

SCPE

View full text Add to dashboard Cite

The plethora of sensors deployed in Internet of Things (IoT) environments generate unprecedented volumes of data, thereby creating a data deluge. Data collected from these sensors can be used to comprehend, examine and control intricate environments around us, facilitating greater intelligence, smarter decision-making, and better performance. The key challenge here is how to mine out proﬁcient information from such immense data. Copious solutions have been put forth to obtain valuable inferences and insights, however, these solutions are still in their developing stages. Moreover, conventional procedures do not address the surging analytical demands of IoT systems. Motivated to resolve this concern, this work investigates the key enablers for performing desired data analytics in IoT applications. A comprehensive survey on the identiﬁed key enablers including their role in IoT data analytics, use cases in which they have been applied and the corresponding IoT applications for the use cases is presented. Furthermore, open research challenges and future research opportunities are also discussed. This article can be used as a basis to foster advanced research in the arena of IoT data analytics.

show abstract

Context-aware feature query to improve the prediction performance

Cited by 4 publications

References 9 publications

Dynamic Feature Acquisition Using Denoising Autoencoders

Dynamic Feature Acquisition Using Denoising Autoencoders

Opportunistic Learning: Budgeted Cost-Sensitive Learning from Data Streams

Data Analytics in the Internet of Things: A Survey

Contact Info

Product

Resources

About