There has been little research into how end users might be able to communicate advice to machine learning systems. If this resource—the users themselves—could somehow work hand-in-hand with machine learning systems, the accuracy of learning systems could be improved and the users ’ understanding and trust of the system could improve as well. We conducted a think-aloud study to see how willing users were to provide feedback and to understand what kinds of feedback users could give. Users were shown explanations of machine learning predictions and asked to provide feedback to improve the predictions. We found that users had no difficulty providing generous amounts of feedback. The kinds of feedback ranged from suggestions for reweighting of features to proposals for new features, feature combinations, relational features, and wholesale changes to the learning algorithm. The results show that user feedback has the potential to significantly improve machine learning systems, but that learning algorithms need to be extended in several ways to be able to assimilate this feedback. ACM Classification: H.5.2 [Information interfaces and presentation (e.g., HCI)] User Interfaces: Theory and methods, Evaluation/methodology. H.1.2 [Models and Principles]: User/Machine Systems: Human information processing,

Although machine learning is becoming commonly used in today’s software, there has been little research into how end users might interact with machine learning systems, beyond communicating simple "right/wrong " judgments. If the users themselves could work hand-in-hand with machine learning systems, the users ’ understanding and trust of the system could improve and the accuracy of learning systems could be improved as well. We conducted three experiments to understand the potential for rich interactions between users and machine learning systems. The first experiment was a think-aloud study that investigated users ’ willingness to interact with machine learning reasoning, and what kinds of feedback users might give to machine learning systems. We then investigated the viability of introducing such feedback into machine learning systems, specifically, how to incorporate some of these types of user feedback into machine learning systems, and what their impact was on the accuracy of the system. Taken together, the results of our experiments show that supporting rich interactions between users and machine learning systems is feasible for both user and machine. This shows the potential of rich humancomputer collaboration via on-the-spot interactions as a promising direction for machine learning systems and users to collaboratively share intelligence.

The potential for machine learning systems to improve via a mutually beneficial exchange of information with users has yet to be explored in much detail. Previously, we found that users were willing to provide a generous amount of rich feedback to machine learning systems, and that the types of some of this rich feedback seem promising for assimilation by machine learning algorithms. Following up on those findings, we ran an experiment to assess the viability of incorporating real-time keyword-based feedback in initial training phases when data is limited. We found that rich feedback improved accuracy but an initial unstable period often caused large fluctuations in classifier behavior. Participants were able to give feedback by relying heavily on system communication in order to respond to changes. The results show that in order to benefit from the user’s knowledge, machine learning systems must be able to absorb keyword-based rich feedback in a graceful manner and provide clear explanations of their predictions. Author Keywords Machine learning, user feedback. ACM CLASSIFICATION KEYWORDS H.5.2 [Information interfaces and presentation (e.g., HCI)] User Interfaces: Theory and methods. H.1.2 [Models and Principles]: User/Machine Systems: Human information processing, Human factors.

As I watched my 12-year-old nephew play Halo 2 online while strategizing with his friends over his microphone-enabled headset, I realized that his play environment might well be the next distance learning platform. In cooperative online video games like Halo 2, players win by working together to understand and overcome the obstacles set forth in the storyline. Essentially, cooperative learning is occurring in these games. The virtual worlds in which today's video games take place can be reshaped as real-time synchronous virtual classrooms; the advance of technology and increasing accessibility of that technology mean that virtual reality is a viable distance education option. As Net Generation students, already the leading population in online gaming, bring their well-documented learning styles and demands for flexibility and adaptability into higher education venues, the three-dimensional gaming environment remade as a virtual classroom could become the natural next step in online learning (Oblinger 2006). The educational analogue to the three-dimensional gaming world is the virtual learning environment (VLE). The VLE is an online space where learners represent themselves through images called avatars, graphical personifications that represent the learner’s identity, presence, location, and interaction within the VLE. Within this environment, students, represented by their individual avatars, can interact in real time with each other and with computer-based agents, digital artifacts, and virtual contexts. These three-dimensional worlds

Although computer games as a way to improve students ’ learning have received attention by many educational researchers, no consensus has been reached on the effects of computer games on student achievement. Moreover, there is lack of empirical research on differential effects of computer games on diverse learners. In response, this study empirically examined the effects of playing computer games on math achievement of 4th graders, with special focus on gender and language minority groups. The study used the 2005 National Assessment of Educational Progress (NAEP), a nationally representative database of the USA. The study performed regression analyses using more than 170,000 U.S. 4th-grade students by applying a proper weight and considering design effects to have high generalizability. The study specified three models for analyses: ELL Model, Gender Model, and Interaction Model. The results showed that English-speaking students who played computer math games in school every day displayed significantly lower math achievement than those who never played. Contrastingly, positive effects of daily computer use were noted among male students whose first language was other than English. Male language minority students who daily played computer games in math demonstrated higher math performance scores compared with their male English-speaking counterparts who never played.

Join our mailing list!

Abstract not found

The views and conclusions contained in this document are those of the authors and should not be interpreted as

This report is made possible by support from Uruguay’s Administración Nacional de Educación Pública and