Predicting Chinese text entry speeds on mobile phones

Liu, Ying; Räihä, Kari-Jouko

doi:10.1145/1753326.1753657

Cited by 12 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A large number of the studies concerned with extending the KLM to smartphone interactions consider its adaptation to text entry methods and predictive text (e.g., [51][52][53][54][55]). One of the earliest KLM extensions for text entry involved revising the original model for three text entry methods on a smartphone's 9-key keyboard [51].…”

Section: Related Workmentioning

confidence: 99%

“…The revised model was used to predict typing speeds considering various methods of input. Another modification to the KLM utilised the adapted operations, Fitts' law, and a Chinese language model to predict user performance with two types of input on a smartphone [52]. The KLM was also extended to measure multi-finger touchscreen keystrokes for the purpose of representing 1Line, a text entry method for Chinese text input [53].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A Predictive Fingerstroke-Level Model for Smartwatch Interaction

Al-Megren

2018

MTI

View full text Add to dashboard Cite

The keystroke-level model (KLM) is commonly used to predict the time it will take an expert user to accomplish a task without errors when using an interactive system. The KLM was initially intended to predict interactions in conventional set-ups, i.e., mouse and keyboard interactions. However, it has since been adapted to predict interactions with smartphones, in-vehicle information systems, and natural user interfaces. The simplicity of the KLM and its extensions, along with their resource-and time-saving capabilities, has driven their adoption. In recent years, the popularity of smartwatches has grown, introducing new design challenges due to the small touch screens and bimanual interactions involved, which make current extensions to the KLM unsuitable for modelling smartwatches. Therefore, it is necessary to study these interfaces and interactions. This paper reports on three studies performed to modify the original KLM and its extensions for smartwatch interaction. First, an observational study was conducted to characterise smartwatch interactions. Second, the unit times for the observed interactions were derived through another study, in which the times required to perform the relevant physical actions were measured. Finally, a third study was carried out to validate the model for interactions with the Apple Watch and Samsung Gear S3. The results show that the new model can accurately predict the performance of smartwatch users with a percentage error of 12.07%; a value that falls below the acceptable percentage dictated by the original KLM~21%.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A Predictive Fingerstroke-Level Model for Smartwatch Interaction

Al-Megren

2018

MTI

View full text Add to dashboard Cite

show abstract

“…The model was later revised to include NFC interactions [10]. An extended model utilised KLM, Fitts' law, and a language model to predict user performance with two types of Chinese input methods on smartphones [17]. For the purpose of presenting a new keyboard, 1Line, KLM was extended to measure multi-finger touchscreen keystrokes [15].…”

Section: Klm Extensionsmentioning

confidence: 99%

“…Smartphone extensions to KLM review the model's decomposition by modifying the original operators, introducing new actions, and revising execution times. These new interactions and extended models include predictive text entry [17], voice recognition [6], Near Field Communication (NFC) technology [10], touch input [20], and touch-less interaction [8].…”

Section: Introductionmentioning

confidence: 99%

Blind FLM: An Enhanced Keystroke-Level Model for Visually Impaired Smartphone Interaction

Al-Megren

Altamimi

Al‐Khalifa

2017

Human-Computer Interaction - INTERACT 2017

View full text Add to dashboard Cite

The Keystroke-Level Model (KLM) is a predictive model used to numerically predict how long it takes an expert user to accomplish a task. KLM has been successfully used to model conventional interactions, however, it does not thoroughly render smartphone touch interactions or accessible interfaces (e.g. screen readers). On the other hand, the Fingerstroke-level Model (FLM) extends KLM to describe and assess mobile-based game applications, which marks it as a candidate model for predicting smartphone touch interactions. This paper aims to further extend FLM for visually impaired smartphone users. An initial user study identified basic elements of blind users' interactions that were used to extend FLM; the new model is called "Blind FLM'". Then an additional user study was conducted to determine the applicability of the new model for describing blind users' touch interactions with a smartphone, and to compute the accuracy of the new model. Blind FLM evaluation showed that it can predict blind users' performance with an average error of 2.36%.

show abstract

“…Consequently, the existing prediction technique is not culturespecific, as it can even be used with non-Latin-based alphabets. Proof of concepts can be found in works related to predictive evaluation of Korean text entry [17] and Chinese text entry [18] on a multitap-based mobile phone keypad with 12 buttons.…”

Section: Modeling Upper-bound Text Entry Speedsmentioning

confidence: 99%

Predicting upper-bound text entry speeds for discrete-tilt-based input on smartphones

2014

View full text Add to dashboard Cite

Motion sensors integrated into contemporary smartphones allow the introduction of new mobile interaction paradigms, here including tilt-based input control in the mobile context. Namely, as opposed to existing implementations that typically apply continuous feedback on tilting, we define Pitch and Roll movement sequences that change the orientation of the mobile device as discrete-tilt input primitives. The respective commands are then used to manage text entry within three discrete-tilt-based methods thus introduced: keyboard bisection, single cursor, and quad cursor. Each method is based on the use of a particular QWERTY-based keyboard layout with related strategy for character input. We model upper-bound text entry speeds for the input methods, taking into account both movement aspects and language context. The movement model corresponds to both the tilt-based shortest path between two consequent characters, which is theoretically defined, and the time of discrete-tilt execution, which is obtained from user testing experiment we conducted. The linguistic model, comprising digraph statistics, is constructed basing on available English corpora. This modeling approach provides discrete-tilt-based text entry speed predictions representing efficiency rates for expert behavior, i.e. for optimal performance. The results obtained enable the evaluation of the proposed designs without need to test with real users, and can furthermore serve as a baseline for efficiency of text entry implementations that rely on discrete tilt.

show abstract

Predicting Chinese text entry speeds on mobile phones

Cited by 12 publications

References 27 publications

A Predictive Fingerstroke-Level Model for Smartwatch Interaction

A Predictive Fingerstroke-Level Model for Smartwatch Interaction

Blind FLM: An Enhanced Keystroke-Level Model for Visually Impaired Smartphone Interaction

Predicting upper-bound text entry speeds for discrete-tilt-based input on smartphones

Contact Info

Product

Resources

About