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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Creating resilient machine learning (ML) systems has become necessary toensure production-ready ML systems that acquire user confidence seamlessly. Thequality of the input data and the model highly influence the successfulend-to-end testing in data-sensitive systems. However, the testing approachesof input data are not as systematic and are few compared to model testing. Toaddress this gap, this paper presents the Fault Injection for UndesirableLearning in input Data (FIUL-Data) testing framework that tests the resilienceof ML models to multiple intentionally-triggered data faults. Data mutatorsexplore vulnerabilities of ML systems against the effects of different faultinjections. The proposed framework is designed based on three main ideas: Themutators are not random; one data mutator is applied at an instance of time,and the selected ML models are optimized beforehand. This paper evaluates theFIUL-Data framework using data from analytical chemistry, comprising retentiontime measurements of anti-sense oligonucleotide. Empirical evaluation iscarried out in a two-step process in which the responses of selected ML modelsto data mutation are analyzed individually and then compared with each other.The results show that the FIUL-Data framework allows the evaluation of theresilience of ML models. In most experiments cases, ML models show higherresilience at larger training datasets, where gradient boost performed betterthan support vector regression in smaller training sets. Overall, the meansquared error metric is useful in evaluating the resilience of models due toits higher sensitivity to data mutation.

Machine Learning Data Suitability and Performance Testing Using Fault Injection Testing Framework