We describe and validate a metric for estimating multi-class classifier performance based on cross-validation and adapted for improvement of small, unbalanced natural-language datasets used in chatbot design. Our experiences draw upon building recruitment chatbots that mediate communication between job-seekers and recruiters by exposing the ML/NLP dataset to the recruiting team. Evaluation approaches must be understandable to various stakeholders, and useful for improving chatbot performance. The metric, nex-cv, uses negative examples in the evaluation of text classification, and fulfils three requirements. First, it is actionable: it can be used by non-developer staff. Second, it is not overly optimistic compared to human ratings, making it a fast method for comparing classifiers. Third, it allows model-agnostic comparison, making it useful for comparing systems despite implementation differences. We validate the metric based on seven recruitmentdomain datasets in English and German over the course of one year.
In this essay, we draw on multidisciplinary scholarship and artistic interventions that we consider to be instances of anti-systematic practice, wherein artists' use of their own biological data and matter supports material research, and catalyzes alternative, embodied knowledge production of the self. While an anti-systematic practice builds on systematizing biomedical tools and practices, its primary aim is to integrate that general knowledge with complex, contextualized experiences for deeper collective self-understanding. The datafication of the self through wearable self-tracking technologies expands our capability to build bodily knowledge, but simultaneously entails pervasive (micro)biosurveillance and reproduces an internalized, isolating neoliberal ethos. Technological advances can constitute forms of (micro)biopower that dominate, control, classify, and govern our life on a molecular level. We consider theory and art practice that domesticates and demystifies biotechnology and citatresits or subverts forms of (micro)biopower, while engaging with medical knowledge and biotechnological capability for bodily observation. The works we describe invite participation in collective body projects and empower self-understanding that arises from collaborative conceptualization of alternate futures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.