2D Transparency Space—Bring Domain Users and Machine Learning Experts Together

Zhou, Jianlong; Chen, Fang

doi:10.1007/978-3-319-90403-0_1

Cited by 14 publications

(9 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, much of the existing research focused on explaining the underlying algorithm [69]; and designers may assume that users have advanced knowledge of machine learning [69], or users have to understand the decision process (e.g., [59,67]). Besides machine learning experts, many other end users can also benefit from machine learning [41,129], and sometimes human-machine collaborations exhibit better performance than the human or the machine alone [4,30,115]. Yet, end users could be domain experts who may not have the background to understand how the algorithm operates.…”

Section: Introductionmentioning

confidence: 99%

How do visual explanations foster end users' appropriate trust in machine learning?

Yang

Huang

Scholtz

et al. 2020

Proceedings of the 25th International Conference on Intelligent User Interfaces

129

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

How do visual explanations foster end users' appropriate trust in machine learning?

Yang

Huang

Scholtz

et al. 2020

Proceedings of the 25th International Conference on Intelligent User Interfaces

129

View full text Add to dashboard Cite

show abstract

“…Visibility is operationalized, for example as interacting with AI/ML processes (via a user interface), providing visual analytics, and displaying (e.g., mapping) probabilities or clusters (Sacha et al, 2017). Explainability refers to the interpretability of AI/ML to developers and end users, for instance highlighting which parts of a feature influence its classification or why inferences are made, based, for example, on keywords (Zhou & Chen, 2018b). In a human-centred deployment of AI/ML, we must explain all steps even when some have obvious reasoning and show high interpretability (e.g., translating French to English).…”

Section: Employing Nlp and H CML To Re S E Arch The S Mart Cit Ymentioning

confidence: 99%

“…aspx?id=32592). Transparency refers to explicating how algorithms act on the data, generate data, how algorithms deal with uncertainty, and how to connect the end users with the thought processes of the algorithm developers (Zhou & Chen, 2018b).…”

Section: Employing Nlp and H CML To Re S E Arch The S Mart Cit Ymentioning

confidence: 99%

“…As much as terms like "trustworthy" evoke a kind of qualitative human engagement in the process of AI/ML, HCML still tends to be highly computational. Methods like explainable AI tend to be optimized in terms of comprehension for AI/ML experts and not domain (e.g., social science) users (Zhou & Chen, 2018b). It also could be in the conceptualization of human-centredness.…”

Section: Employing Nlp and H CML To Re S E Arch The S Mart Cit Ymentioning

confidence: 99%

See 1 more Smart Citation

Putting humans back in the loop of machine learning in Canadian smart cities

Zheng

Sieber

2021

Transactions in GIS

View full text Add to dashboard Cite

Even as researchers recognize smart cities as sociotechnological assemblages, the attractiveness of artificial intelligence and machine learning (AI/ML) continues to drive cities towards automating urban process and analysis. Rather than arguing whether researchers should automate their analysis, we are interested in identifying where non-automated, manual judgement calls are made and the analysis is more subjective than presented. We examine topic modelling, an ML method, in an analysis of applications to a pan-Canadian smart cities grant competition. We document 11 steps in topic modelling from data collection to interpretation. At each step, including the choice of the topic modelling method, some degree of human intervention is required. We draw on human-centred ML research to argue for a greater recognition of the role of humans-as-researchers to preclude further uncritical adoption of AI/ML to research smart cities. | INTRODUC TI ONSmart cities, through their sensors and software dashboards, produce enormous volumes, velocities and varieties of real-time streaming data. Such data come in large part from volunteered geographic information (VGI), whether that is from mobile devices as urban residents traverse space-time, or are made available from private services like Uber. We have created new methods (e.g., predictive analytics) and new sciences (e.g., urban science, data science) to take advantage of these data as a window into the city. While artificial intelligence and machine learning (AI/ ML) algorithms can analyse the vast amounts of data in the smart city, the methods also reinforce approaches that reify automation, not only in the management but also in the understanding of the city. This aligns with the long-standing association of "smartness" with the assumption that technology can improve urban functioning, despite the gradual transition from a technologically deterministic paradigm towards urban sustainability and well-being (Albino, Berardi, & Dangelico, 2015;Batty, 2013; Goodspeed, 2015). Municipalities continue to credit technologies with the ability to transform cities into places of high efficiency and innovation (Marshall, 2021). Kandt and Batty (2021) address the epistemological problems of urban data analytics and, by extension smart cities, both of which "can be automated with little human intervention" (Kandt & Batty, 2021, p. 1). They argue that it is doubtful whether big data can provide us with a general and in-depth understanding of urban problems. Big data solve problems on the surface; not why they happen. Urban data analytics offers a simple solution to urban problems, which have long been called wicked problems (Rittel & Webber, 1973). Despite a desire by some researchers and practitioners for straightforward answers, a wicked problem cannot be reduced to a single cause; the problem is incomplete and conflicting in understanding, and lacking a straightforward presentation. Instead of treating immediate symptoms of city problems, we should try to qualitatively understand why cer...

show abstract

“…As ML models can provide incorrect predictions [CLG * 15], ML experts have to check whether the model functions properly [TKC17]. Also, domain experts in critical fields need to understand how a specific prediction has been reached in order to trust in ML [ZC18]. For example, in medicine, a physician might not rely on a model without explanations of how and why it forms a prediction, since patient lives are at risk [RSG16,HTF01,LBL16].…”

Section: Introductionmentioning

confidence: 99%

VisRuler: Visual Analytics for Extracting Decision Rules from Bagged and Boosted Decision Trees

Chatzimparmpas¹,

Martins²,

Kerren³

2021

Preprint

View full text Add to dashboard Cite

Figure 1: Extracting decision rules for manual evaluation with VISRULER: (a) the panel with various visual metaphors for selecting performant and diverse models; (b) the box plot for feature selection according to per algorithmic importance; (c) the visual embedding of computed decisions that training instances fall in due to their values; (d) the vertical parallel coordinates plot that summarizes the rules with value ranges for each feature and highlights the current test instance; and (e) the horizontal stacked bar chart for revealing the class agreement of each model against the manual decisions, together with the parallel coordinates plots for tuning hyperparameters and training new models.

show abstract

2D Transparency Space—Bring Domain Users and Machine Learning Experts Together

Cited by 14 publications

References 28 publications

How do visual explanations foster end users' appropriate trust in machine learning?

How do visual explanations foster end users' appropriate trust in machine learning?

Putting humans back in the loop of machine learning in Canadian smart cities

VisRuler: Visual Analytics for Extracting Decision Rules from Bagged and Boosted Decision Trees

Contact Info

Product

Resources

About