Table 4 XAI objectives Objectives/goal... | Emerald Publishing

Table 4

XAI objectives

Objectives/goal	Definition (adopted from the given sources)	Sources
Understandability Other included codes: Intelligibility, comprehensibility, interpretability	The degree to which end users are able to form an accurate mental model regarding how the AI system works	Chazette and Schneider (2020), Cheng et al. (2019), Cirqueira et al. (2020), Cramer et al. (2008), Ehsan et al. (2019), Eiband et al. (2018), Eslami et al. (2018), Lim et al. (2009), Lim and Dey (2009), Oh et al. (2018), Putnam and Conati (2019), van der Waa et al. (2020), Weitz et al. (2019a), Wang et al. (2019), Xie et al. (2019)
Trustworthiness Other included codes: Trust	Refers to end users' perception about the truthfulness and honesty of the system, as well as beliefs that the system works as intended	Brennen (2020), Bussone et al. (2015), Cheng et al. (2019), Ehsan et al. (2019), Schrills and Franke (2020), Wang et al. (2019), Weitz et al. (2019a), Weitz et al. (2019b), Xie et al. (2019), Yin et al. (2019)
Transparency	The degree of information that is disclosed about the AI system. For example, high transparency systems disclose (almost) fully the system functioning from data to algorithms and parameters	Brennen (2020), Cai et al. (2019), Chazette and Schneider (2020), Cramer et al. (2008), Eiband et al. (2018), Ngo et al. (2020), Schrills and Franke (2020)
Controllability	End users' subjective sense of control over the AI system	Ngo et al. (2020), Oh et al. (2018), Wang et al. (2019)
Fairness Other included codes: Justice	Refers to the subjective perception of whether the decisions or recommendations made by the AI system feel right and just	Binns et al. (2018), Dodge et al. (2019)

Objectives/goal	Definition (adopted from the given sources)	Sources
UnderstandabilityOther included codes: Intelligibility, comprehensibility, interpretability	The degree to which end users are able to form an accurate mental model regarding how the AI system works	Chazette and Schneider (2020), Cheng et al. (2019), Cirqueira et al. (2020), Cramer et al. (2008), Ehsan et al. (2019), Eiband et al. (2018), Eslami et al. (2018), Lim et al. (2009), Lim and Dey (2009), Oh et al. (2018), Putnam and Conati (2019), van der Waa et al. (2020), Weitz et al. (2019a), Wang et al. (2019), Xie et al. (2019)
TrustworthinessOther included codes: Trust	Refers to end users' perception about the truthfulness and honesty of the system, as well as beliefs that the system works as intended	Brennen (2020), Bussone et al. (2015), Cheng et al. (2019), Ehsan et al. (2019), Schrills and Franke (2020), Wang et al. (2019), Weitz et al. (2019a), Weitz et al. (2019b), Xie et al. (2019), Yin et al. (2019)
Transparency	The degree of information that is disclosed about the AI system. For example, high transparency systems disclose (almost) fully the system functioning from data to algorithms and parameters	Brennen (2020), Cai et al. (2019), Chazette and Schneider (2020), Cramer et al. (2008), Eiband et al. (2018), Ngo et al. (2020), Schrills and Franke (2020)
Controllability	End users' subjective sense of control over the AI system	Ngo et al. (2020), Oh et al. (2018), Wang et al. (2019)
FairnessOther included codes: Justice	Refers to the subjective perception of whether the decisions or recommendations made by the AI system feel right and just	Binns et al. (2018), Dodge et al. (2019)