Users, especially the non-expert users, commonly experience problems when connecting multiple devices with interoperability. While studies on multiple device connections are mostly concentrated on spontaneous device a...Users, especially the non-expert users, commonly experience problems when connecting multiple devices with interoperability. While studies on multiple device connections are mostly concentrated on spontaneous device association techniques with a focus on security aspects, the research on user interaction for device connection is still limited. More research into understanding people is needed for designers to devise usable techniques. This research applies the Research-through-Design method and studies the non-expert users＇ interactions in establishing wireless connections between devices. The ＂Learning from Examples＂ concept is adopted to develop a study focus line by learning from the expert users＇ interaction with devices. This focus line is then used for guiding researchers to explore the non-expert users＇ difficulties at each stage of the focus line. Finally, the Research-through-Design approach is used to understand the users＇ difficulties, gain insights to design problems and suggest usable solutions. When connecting a device, the user is required to manage not only the device＇s functionality but also the interaction between devices. Based on learning from failures, an important insight is found that the existing design approach to improve single-device interaction issues, such as improvements to graphical user interfaces or computer guidance, cannot help users to handle problems between multiple devices. This study finally proposes a desirable user-device interaction in which images of two devices function together with a system image to provide the user with feedback on the status of the connection, which allows them to infer any required actions.展开更多
In applications of learning from examples to real-world tasks, feature subset selection is important to speed up training and to improve generalization performance. ideally, an inductive algorithm should use subset of...In applications of learning from examples to real-world tasks, feature subset selection is important to speed up training and to improve generalization performance. ideally, an inductive algorithm should use subset of features as small as possible. In this paper however, the authors show that the problem of selecting the minimum subset of features is NP-hard. The paper then presents a greedy algorithm for feature subset selection. The result of running the greedy algorithm on hand-written numeral recognition problem is also given.展开更多
文摘Users, especially the non-expert users, commonly experience problems when connecting multiple devices with interoperability. While studies on multiple device connections are mostly concentrated on spontaneous device association techniques with a focus on security aspects, the research on user interaction for device connection is still limited. More research into understanding people is needed for designers to devise usable techniques. This research applies the Research-through-Design method and studies the non-expert users＇ interactions in establishing wireless connections between devices. The ＂Learning from Examples＂ concept is adopted to develop a study focus line by learning from the expert users＇ interaction with devices. This focus line is then used for guiding researchers to explore the non-expert users＇ difficulties at each stage of the focus line. Finally, the Research-through-Design approach is used to understand the users＇ difficulties, gain insights to design problems and suggest usable solutions. When connecting a device, the user is required to manage not only the device＇s functionality but also the interaction between devices. Based on learning from failures, an important insight is found that the existing design approach to improve single-device interaction issues, such as improvements to graphical user interfaces or computer guidance, cannot help users to handle problems between multiple devices. This study finally proposes a desirable user-device interaction in which images of two devices function together with a system image to provide the user with feedback on the status of the connection, which allows them to infer any required actions.
文摘In applications of learning from examples to real-world tasks, feature subset selection is important to speed up training and to improve generalization performance. ideally, an inductive algorithm should use subset of features as small as possible. In this paper however, the authors show that the problem of selecting the minimum subset of features is NP-hard. The paper then presents a greedy algorithm for feature subset selection. The result of running the greedy algorithm on hand-written numeral recognition problem is also given.
