On Suggesting Phrases vs. Predicting Words for Mobile Text Composition

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'

[1]  Mark D. Dunlop,et al.  Multidimensional pareto optimization of touchscreen keyboards for speed, familiarity and improved spell checking , 2012, CHI.

[2]  Peter Haider,et al.  Learning to Complete Sentences , 2005, ECML.

[3]  Kathleen F. McCoy,et al.  User Interaction with Word Prediction: The Effects of Prediction Quality , 2009, TACC.

[4]  Jeffrey Heer,et al.  Predictive translation memory: a mixed-initiative system for human language translation , 2014, UIST.

[5]  Setsuo Ohsuga,et al.  INTERNATIONAL CONFERENCE ON VERY LARGE DATA BASES , 1977 .

[6]  Krzysztof Z. Gajos,et al.  Toward Collaborative Ideation at Scale: Leveraging Ideas from Others to Generate More Creative and Diverse Ideas , 2015, CSCW.

[7]  Tim Paek,et al.  Text Text Revolution: A Game That Improves Text Entry on Mobile Touchscreen Keyboards , 2011, Pervasive.

[8]  Philipp Koehn,et al.  Scalable Modified Kneser-Ney Language Model Estimation , 2013, ACL.

[9]  Shwetak N. Patel,et al.  ContextType: using hand posture information to improve mobile touch screen text entry , 2013, CHI.

[10]  Per Ola Kristensson,et al.  The inviscid text entry rate and its application as a grand goal for mobile text entry , 2014, MobileHCI '14.

[11]  Shumin Zhai,et al.  The word-gesture keyboard: reimagining keyboard interaction , 2012, CACM.

[12]  Shumin Zhai,et al.  Bimanual gesture keyboard , 2012, UIST.

[13]  Ian McGraw,et al.  On the compression of recurrent neural networks with an application to LVCSR acoustic modeling for embedded speech recognition , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[14]  Shumin Zhai,et al.  Interlaced QWERTY: accommodating ease of visual search and input flexibility in shape writing , 2008, CHI.

[15]  Tim Paek,et al.  Usability guided key-target resizing for soft keyboards , 2010, IUI '10.

[16]  Shumin Zhai,et al.  Touch behavior with different postures on soft smartphone keyboards , 2012, Mobile HCI.

[17]  H. V. Jagadish,et al.  Effective Phrase Prediction , 2007, VLDB.

[18]  Kathleen F. McCoy,et al.  Evaluating Word Prediction: Framing Keystroke Savings , 2008, ACL.

[19]  Nicholas W. Kohn,et al.  Collaborative fixation: effects of others' ideas on brainstorming , 2011 .

[20]  Peter Young,et al.  Smart Reply: Automated Response Suggestion for Email , 2016, KDD.

[21]  Shumin Zhai,et al.  Effects of Language Modeling and its Personalization on Touchscreen Typing Performance , 2015, CHI.

[22]  Kasper Hornbæk,et al.  Vulture: a mid-air word-gesture keyboard , 2014, CHI.

[23]  Hermann Ney,et al.  Improved backing-off for M-gram language modeling , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[24]  Shumin Zhai,et al.  Long short term memory neural network for keyboard gesture decoding , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[25]  R. Marsh,et al.  How examples may (and may not) constrain creativity , 1996, Memory & cognition.

[26]  Per Ola Kristensson,et al.  Intelligently Aiding Human-Guided Correction of Speech Recognition , 2010, AAAI.

[27]  Sharad Vikram,et al.  Capturing Meaning in Product Reviews with Character-Level Generative Text Models , 2015, ArXiv.

[28]  Per Ola Kristensson,et al.  Complementing text entry evaluations with a composition task , 2014, TCHI.

[29]  Larry P. Heck,et al.  Contextual LSTM (CLSTM) models for Large scale NLP tasks , 2016, ArXiv.

[30]  Geehyuk Lee,et al.  Evaluation of a Smart-Restorable Backspace Technique to Facilitate Text Entry Error Correction , 2016, CHI.

[31]  Matthew R. Walter,et al.  What to talk about and how? Selective Generation using LSTMs with Coarse-to-Fine Alignment , 2015, NAACL.

[32]  Arpit Agarwal,et al.  AutoComPaste: auto-completing text as an alternative to copy-paste , 2012, AVI.

[33]  Derek G. Bridge,et al.  GhostWriter-2.0: Product Reviews with Case-Based Support , 2010, SGAI Conf..

[34]  Per Ola Kristensson,et al.  VelociTap: Investigating Fast Mobile Text Entry using Sentence-Based Decoding of Touchscreen Keyboard Input , 2015, CHI.

[35]  Matthew Richardson,et al.  Blending LSTMs into CNNs , 2015, ICLR 2016.

[36]  Shumin Zhai,et al.  Quasi-qwerty soft keyboard optimization , 2010, CHI.

[37]  Shumin Zhai,et al.  Both complete and correct?: multi-objective optimization of touchscreen keyboard , 2014, CHI.

[38]  Alexander M. Rush,et al.  Optimal Beam Search for Machine Translation , 2013, EMNLP.

[39]  D. Jeffery Higginbotham,et al.  Evaluation of keystroke savings across five assistive communication technologies , 1992 .

[40]  Shumin Zhai,et al.  A Cost-Benefit Study of Text Entry Suggestion Interaction , 2016, CHI.

[41]  Shumin Zhai,et al.  Performance and User Experience of Touchscreen and Gesture Keyboards in a Lab Setting and in the Wild , 2015, CHI.