Business agility requires support from recommendation systems, but explaining recommendations may yield information disclosure. We analyze how to provide explanations in the scenario of Multi-Stakeholder Recommendation where the sensible information of one stakeholder should not be disclosed in the explanation to another stakeholder. Among the several types of explanations analyzed, counterfactual explanations come off best as they allow the system to preserve each stakeholder’s privacy and sensitive information in terms of preferences.
Explanation in Multi-Stakeholder Recommendation for Enterprise Decision Support Systems / Cornacchia, G.; Donini, F. M.; Narducci, F.; Pomo, C.; Ragone, A.. - 423:(2021), pp. 39-47. (Intervento presentato al convegno International Workshops associated with 33rd International Conference on Advanced Information Systems Engineering, CAiSE 2021 nel 2021) [10.1007/978-3-030-79022-6_4].
Explanation in Multi-Stakeholder Recommendation for Enterprise Decision Support Systems
Cornacchia G.;Donini F. M.;Narducci F.;Pomo C.;
2021-01-01
Abstract
Business agility requires support from recommendation systems, but explaining recommendations may yield information disclosure. We analyze how to provide explanations in the scenario of Multi-Stakeholder Recommendation where the sensible information of one stakeholder should not be disclosed in the explanation to another stakeholder. Among the several types of explanations analyzed, counterfactual explanations come off best as they allow the system to preserve each stakeholder’s privacy and sensitive information in terms of preferences.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
