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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We study the interpretability of predictive systems that use high-dimensonalbehavioral and textual data. Examples include predicting product interest basedon online browsing data and detecting spam emails or objectionable web content.Recently, counterfactual explanations have been proposed for generating insightinto model predictions, which focus on what is relevant to a particularinstance. Conducting a complete search to compute counterfactuals is verytime-consuming because of the huge dimensionality. To our knowledge, forbehavioral and text data, only one model-agnostic heuristic algorithm (SEDC)for finding counterfactual explanations has been proposed in the literature.However, there may be better algorithms for finding counterfactuals quickly.This study aligns the recently proposed Linear Interpretable Model-agnosticExplainer (LIME) and Shapley Additive Explanations (SHAP) with the notion ofcounterfactual explanations, and empirically benchmarks their effectiveness andefficiency against SEDC using a collection of 13 data sets. Results show thatLIME-Counterfactual (LIME-C) and SHAP-Counterfactual (SHAP-C) have low andstable computation times, but mostly, they are less efficient than SEDC.However, for certain instances on certain data sets, SEDC's run time iscomparably large. With regard to effectiveness, LIME-C and SHAP-C findreasonable, if not always optimal, counterfactual explanations. SHAP-C,however, seems to have difficulties with highly unbalanced data. Because of itsgood overall performance, LIME-C seems to be a favorable alternative to SEDC,which failed for some nonlinear models to find counterfactuals because of theparticular heuristic search algorithm it uses. A main upshot of this paper isthat there is a good deal of room for further research. For example, we proposealgorithmic adjustments that are direct upshots of the paper's findings.

Counterfactual Explanation Algorithms for Behavioral and Textual Data