User interaction with word prediction: the eects of prediction quality (preprint)

Word prediction systems can reduce the number of keystrokes required to form a message in a letter-based AAC system. It has been questioned, however, whether such savings translate into an enhanced communication rate due to the additional overhead (e.g., shifting of focus and repeated scanning of a prediction list) required in using such a system. Our hypothesis is that word prediction has high potential for enhancing AAC communication rate, but the amount is dependent in a complex way on the accuracy of the predictions. Due to significant user interface variations in AAC systems and the potential bias of prior word prediction experience on existing devices, this hypothesis is difficult to verify. We present a study of two different word prediction methods compared against letter-by-letter entry at simulated AAC communication rates. We find that word prediction systems can in fact speed communication rate (an advanced system gave a 58.6% improvement), and that a more accurate word prediction system can raise communication rate higher than is explained by the additional accuracy of the system alone due to better utilization (93.6% utilization for advanced vs. 78.2% for basic).

[1]  Simon P. Levine,et al.  Modeling of user performance with computer access and augmentative communication systems for handicapped people , 1990 .

[2]  Graeme Hirst,et al.  Testing the Efficacy of Part-of-Speech Information in Word Completion , 2003 .

[3]  D. Beukelman,et al.  Augmentative & Alternative Communication: Supporting Children & Adults With Complex Communication Needs , 2006 .

[4]  Kathleen F. McCoy,et al.  Word prediction and communication rate in AAC , 2008 .

[5]  H H Koester,et al.  Learning and performance of able-bodied individuals using scanning systems with and without word prediction. , 1994, Assistive technology : the official journal of RESNA.

[6]  Sheri Hunnicutt,et al.  Improving word prediction using Markov models and Heuristic methods , 2001 .

[7]  Dale Schuurmans,et al.  Semantic n-gram language modeling with the latent maximum entropy principle , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[8]  Jianhua Li,et al.  Semantic knowledge in word completion , 2005, Assets '05.

[9]  Hinrich Schütze,et al.  Book Reviews: Foundations of Statistical Natural Language Processing , 1999, CL.

[10]  Kathleen F. McCoy,et al.  Topic modeling in fringe word prediction for AAC , 2006, IUI '06.

[11]  Robert L. Mercer,et al.  Class-Based n-gram Models of Natural Language , 1992, CL.

[12]  H. H. Koester,et al.  A model of performance cost versus benefit for augmentative communication systems , 1993, Proceedings of the 15th Annual International Conference of the IEEE Engineering in Medicine and Biology Societ.

[13]  Horabail S. Venkatagiri.  Efficiency of lexical prediction as a communication acceleration technique , 1993 .

[14]  Alan F. Newell,et al.  The rôle of natural language processing in alternative and augmentative communication , 1998, Natural Language Engineering.

[15]  Marco Baroni,et al.  THE LANGUAGE COMPONENT OF THE FASTY TEXT PREDICTION SYSTEM , 2005, Appl. Artif. Intell..

[16]  H. S. Venkatagiri,et al.  Effect of window size on rate of communication in a lexical prediction AAC system , 1994 .

[17]  Georg Edelmayer,et al.  First User Test Results with the Predictive Typing System FASTY , 2004, ICCHP.

[18]  Alan F. Newell,et al.  Effect of the “PAL” word prediction system on the quality and quantity of text generation , 1992 .

[19]  James H. Martin,et al.  Speech and language processing: an introduction to natural language processing, computational linguistics, and speech recognition, 2nd Edition , 2000, Prentice Hall series in artificial intelligence.

[20]  Marco Baroni,et al.  Exploiting Long Distance Collocational Relations in Predictive Typing , 2003 .

[21]  Slava M. Katz,et al.  Estimation of probabilities from sparse data for the language model component of a speech recognizer , 1987, IEEE Trans. Acoust. Speech Signal Process..

[22]  Nestor Garay-Vitoria,et al.  A Comparison of Prediction Techniques to Enhance the Communication Rate , 2004, User Interfaces for All.

[23]  Simon P. Levine,et al.  Modeling the speed of text entry with a word prediction interface , 1994 .

[24]  Denis Anson,et al.  The Effects of Word Completion and Word Prediction on Typing Rates Using On-Screen Keyboards , 2006, Assistive technology : the official journal of RESNA.

[25]  Henry Lieberman,et al.  A commonsense approach to predictive text entry , 2004, CHI EA '04.

[26]  Simon P. Levine,et al.  Effect of a word prediction feature on user performance , 1996 .

[27]  Nestor Garay-Vitoria,et al.  Text prediction systems: a survey , 2006, Universal Access in the Information Society.

[28]  Tonio Wandmacher,et al.  SIBYLLE: a system for alternative communication adapting to the context and its user , 2007, Assets '07.

[29]  Brad A. Myers,et al.  From letters to words: efficient stroke-based word completion for trackball text entry , 2006, Assets '06.

[30]  Simon P. Levine,et al.  Model simulations of user performance with word prediction , 1998 .

[31]  Kathleen F. McCoy,et al.  The Effects of Word Prediction on Communication Rate for AAC , 2007, HLT-NAACL.

[32]  Tonio Wandmacher,et al.  Methods to Integrate a Language Model with Semantic Information for a Word Prediction Component , 2007, EMNLP.

[33]  Simon P. Levine,et al.  Keystroke-Level Models for User Performance with Word Prediction , 1997 .

[34]  Tonio Wandmacher,et al.  Training Language Models without Appropriate Language Resources: Experiments with an AAC System for Disabled People , 2006, LREC.

[35]  Gregory W. Lesher.  DOMAIN-SPECIFIC WORD PREDICTION FOR AUGMENTATIVE COMMUNICATION , 2001 .

[36]  Donald Hindle,et al.  Deterministic Parsing of Syntactic Non-fluencies , 1983, ACL.

[37]  Jun Gong.  Semantic & syntactic context-aware text entry methods , 2007, Assets '07.

[38]  Kathleen F. McCoy,et al.  Generating text from compressed input: an intelligent interface for people with severe motor impairments , 1992, CACM.

[39]  Kevin Knight,et al.  The Practical Value of N-Grams Is in Generation , 1998, INLG.