Multidimensional pareto optimization of touchscreen keyboards for speed, familiarity and improved spell checking

Dunlop, Mark; Levine, John

doi:10.1145/2207676.2208659

Cited by 82 publications

(78 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One approach is to partially or completely redesign the standard QWERTY touchscreen typing experience. Examples include gesture keyboards [13], optimized keyboards such as ATOMIK [23], interlaced QWERTY [22], the quasi-Qwerty optimized keyboard [2], multidimensional Pareto keyboard optimization [5], multilingual keyboard optimization [3], KALQ [17], and systems that adapt to user input or sensor data (e.g. [4,6]).…”

Section: Introductionmentioning

confidence: 99%

VelociTap

Vertanen

Memmi

Emge

et al. 2015

Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

We present VelociTap: a state-of-the-art touchscreen keyboard decoder that supports a sentence-based text entry approach. VelociTap enables users to seamlessly choose from three word-delimiter actions: pushing a space key, swiping to the right, or simply omitting the space key and letting the decoder infer spaces automatically. We demonstrate that VelociTap has a significantly lower error rate than Google's keyboard while retaining the same entry rate. We show that intermediate visual feedback does not significantly affect entry or error rates and we find that using the space key results in the most accurate results. We also demonstrate that enabling flexible word-delimiter options does not incur an error rate penalty. Finally, we investigate how small we can make the keyboard when using VelociTap. We show that novice users can reach a mean entry rate of 41 wpm on a 40 mm wide smartwatch-sized keyboard at a 3% character error rate.

show abstract

Section: Introductionmentioning

confidence: 99%

VelociTap

Vertanen

Memmi

Emge

et al. 2015

Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

show abstract

“…Naturally, these tasks can vary a great deal: research relating to typing for example will often involve participants copying tens or hundreds of sentences on the whichever device is being studied (Dunlop, Komninos, & Durga, 2014;Dunlop & Levine, 2012;MacKenzie & Soukoreff, 2003;Salthouse, 1986) whereas research aiming to understand how users multitask when using various technology might ask participants to follow a set of instructions in order to interact with a specially designed interface (Brumby, Cox, Back, & Gould, 2013). To understand the effect these variations have upon the interaction, a number of metrics are recorded during these tasks, from the quantitative speed and accuracy readings, to more qualitative user feedback and ratings.…”

Section: Lab Experiments In Hcimentioning

confidence: 99%

Exploring the effects of non-monetary reimbursement for participants in HCI research

Wiseman

Cox

Gould

et al. 2017

HCj

View full text Add to dashboard Cite

When running experiments within the field of Human Computer Interaction (HCI) it is common practice to ask participants to come to a specified lab location, and reimburse them monetarily for their time and travel costs. This, however, is not the only means by which to encourage participation in scientific study. Citizen science projects, which encourage the public to become involved in scientific research, have had great success in getting people to act as sensors to collect data or to volunteer their idling computer or brain power to classify large data sets across a broad range of fields including biology, cosmology and physical and environmental science. This is often done without the expectation of payment. Additionally, data collection need not be done on behalf of an external researcher; the Quantified Self (QS) movement allows people to reflect on data they have collected about themselves. This too, then, is a form of non-reimbursed data collection. Here we investigate whether citizen HCI scientists and those interested in personal data produce reliable results compared to participants in more traditional lab-based studies. Through six studies, we explore how participation rates and data quality are affected by recruiting participants without monetary reimbursement: either by providing participants with data about themselves as reward (a QS approach), or by simply requesting help with no extrinsic reward (as in citizen science projects). We show that people are indeed willing to take part in online HCI research in the absence of extrinsic monetary reward, and that the data generated by participants who take part for selfless reasons, rather than for monetary reward, can be as high quality as data gathered in the lab and in addition may be of higher quality than data generated by participants given monetary reimbursement online. This suggests that large HCI experiments could be run online in the future, without having to incur the equally large reimbursement costs alongside the possibility of running experiments in environments outside of the lab.

show abstract

“…Numerous innovations in text entry have been studied, including gesture keyboards [39,4,23,29,1], key-target resizing [30,15], alternative layouts [9,38,6], and sensor-based adaptation [3,13]. Speed and accuracy gains have been reported in systems where the user presents a complete utterance to the system, which the system can then process as a whole: speech recognition with editing [35], or typing with feedback only at the end of a sentence [37].…”

Section: Related Workmentioning

confidence: 99%

On Suggesting Phrases vs. Predicting Words for Mobile Text Composition

Arnold

Gajos

Kalai

2016

Proceedings of the 29th Annual Symposium on User Interface Software and Technology

View full text Add to dashboard Cite

A system capable of suggesting multi-word phrases while someone is writing could supply ideas about content and phrasing and allow those ideas to be inserted efficiently. Meanwhile, statistical language modeling has provided various approaches to predicting phrases that users type. We introduce a simple extension to the familiar mobile keyboard suggestion interface that presents phrase suggestions that can be accepted by a repeated-tap gesture. In an extended composition task, we found that phrases were interpreted as suggestions that affected the content of what participants wrote more than conventional single-word suggestions, which were interpreted as predictions. We highlight a design challenge: how can a phrase suggestion system make valuable suggestions rather than just accurate predictions?

show abstract

Multidimensional pareto optimization of touchscreen keyboards for speed, familiarity and improved spell checking

Cited by 82 publications

References 27 publications

VelociTap

VelociTap

Exploring the effects of non-monetary reimbursement for participants in HCI research

On Suggesting Phrases vs. Predicting Words for Mobile Text Composition

Contact Info

Product

Resources

About