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

Predictive analytics aims to build machine learning models to predictbehavior patterns and use predictions to guide decision-making. Predictiveanalytics is human involved, thus the machine learning model is preferred to beinterpretable. In literature, Generalized Additive Model (GAM) is a standardfor interpretability. However, due to the one-to-many and many-to-one phenomenawhich appear commonly in real-world scenarios, existing GAMs have limitationsto serve predictive analytics in terms of both accuracy and trainingefficiency. In this paper, we propose FXAM (Fast and eXplainable AdditiveModel), a unified and fast interpretable model for predictive analytics. FXAMextends GAM's modeling capability with a unified additive model for numerical,categorical, and temporal features. FXAM conducts a novel training procedurecalled Three-Stage Iteration (TSI). TSI corresponds to learning over numerical,categorical, and temporal features respectively. Each stage learns a localoptimum by fixing the parameters of other stages. We design joint learning overcategorical features and partial learning over temporal features to achievehigh accuracy and training efficiency. We prove that TSI is guaranteed toconverge to the global optimum. We further propose a set of optimizationtechniques to speed up FXAM's training algorithm to meet the needs ofinteractive analysis. Thorough evaluations conducted on diverse data setsverify that FXAM significantly outperforms existing GAMs in terms of trainingspeed, and modeling categorical and temporal features. In terms ofinterpretability, we compare FXAM with the typical post-hoc approachXGBoost+SHAP on two real-world scenarios, which shows the superiority of FXAM'sinherent interpretability for predictive analytics.

A Unified and Fast Interpretable Model for Predictive Analytics