English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Due to their inherent variabilities,nanomaterial-based sensors arechallenging to translate into real-world applications,wherereliability/reproducibility is key.Recently we showed Bayesian inference can beemployed on engineered variability in layered nanomaterial-based opticaltransmission filters to determine optical wavelengths with highaccuracy/precision.In many practical applications the sensing cost/speed andlong-term reliability can be equal or more important considerations.Thoughvarious machine learning tools are frequently used on sensor/detector networksto address these,nonetheless their effectiveness on nanomaterial-based sensorshas not been explored.Here we show the best choice of ML algorithm in acyber-nanomaterial detector is mainly determined by specific useconsiderations,e.g.,accuracy, computational cost,speed, and resilience againstdrifts/ageing effects.When sufficient data/computing resources areprovided,highest sensing accuracy can be achieved by the kNN and Bayesianinference algorithms,but but can be computationally expensive for real-timeapplications.In contrast,artificial neural networks are computationallyexpensive to train,but provide the fastest result under testing conditions andremain reasonably accurate.When data is limited,SVMs perform well even withsmall training sets,while other algorithms show considerable reduction inaccuracy if data is scarce,hence,setting a lower limit on the size of requiredtraining data.We show by tracking/modeling the long-term drifts of the detectorperformance over large (1year) period,it is possible to improve the predictiveaccuracy with no need for recalibration.Our research shows for the first timeif the ML algorithm is chosen specific to use-case,low-cost solution-processedcyber-nanomaterial detectors can be practically implemented under diverseoperational requirements,despite their inherent variabilities.

Development of Use-specific High Performance Cyber-Nanomaterial Optical Detectors by Effective Choice of Machine Learning Algorithms