Computer Science
8 years ago

Missing data limits software prediction accuracy, 40% threshold discovered.

Impact Analysis of Missing Values on the Prediction Accuracy of Analogy-based Software Effort Estimation Method AQUA
Jingzhou Li, Ahmed Al‐Emran, Guenther Ruhe

Paper Summary

Related papers

Theme background image
Computer Science
8 years ago

Analogy Choices Unlock Transformative Insights Across Mathematics and Society

Fuzzy Logic Application
1 view
Theme background image
Arts and Humanities
8 years ago

Confucius' Relational Analogies Unveil Profound Insights into Human Connections

Literary Theory
1 view
Theme background image
Neuroscience
8 years ago

New model reveals how analogies transform knowledge for better decision-making.

Cognitive Science

Effort estimation software can handle missing data, but too many missing values can reduce its accuracy. A study on AQUA method found that the more missing values, the less accurate the predictions. The accuracy decreases quadratically with the percentage of missing values. The method can handle up to 40% missing values for reliable predictions. Further research is needed to explore other ways of using the method.