I 
Word Prediction for 
Nestor Garay-Vitoria 
Informatika Fakultatea 
Euskal Herriko Unibertsitatea 
649 Postakutxa 
E-20080 Donostia (Basque Country) 
E-mail: nestor@si, ehu. es 
Phone: +34 43 21 80 00 
Fax: +34 43 21 93 06 
Inflected Languages. Application to Basque 
Language 
Julio G. Abascal 
Informatika Fakultatea 
Euskal Herriko Unibertsitatea 
649 Postakutxa 
E-20080 Donostia (Basque Country) 
E-mail: julio©si, ehu. es 
Phone: +34 43 21 80 00 
Fax: +34 43 21 93 06 
Abstract 
Several word prediction methods to help 
the communication of people with disabil- 
ities can be found in the recent litera- 
ture. Most Of them have been developed for 
English or other non-inflected languages. 
While most of these methods can be modi- 
fied to be used in other languages with sim- 
ilar structures, they may not be directly 
adapted to inflected languages. In this pa- 
per some word prediction techniques are re- 
viewed and the difficulties to apply them to 
inflected languages are studied. Possibili- 
ties for word prediction methods that cope 
with the enormous number of different in- 
flexions of each word are proposed, using 
Basque as the target language. Finally, 
conclusions about word prediction for in- 
flected languages are extracted from the ex- 
perience with the Basque language. 
1 Introduction 
So far, word prediction methods have been devel- 
oped in order to increase message composition rate 
for people with severe motor and speech disabilities. 
These methods try to guess what is going to be the 
current or even the next word the user is trying to 
type. Their results are normally measured in terms 
of keystroke savings (Ks) 1 . 
To our knowledge, the design of word predic- 
tion methods is mainly focused on non-inflected lan- 
guages, like English. Words in these types of lan- 
guages have a small amount of variation, like the 
ones due to number (singular or plural) for instance, 
1 
Nchar - Nkeystrok Ks -- 
Nchar 
house/houses, spy~spies. Some other languages ad- 
mit differences in gender, for example in French: 
voisin/voisine. When the number of different forms 
of a word is small, it is possible to include all of them 
in the dictionary used in word prediction. Nonethe- 
less, inflected languages can have a huge number of 
affixes that determine the syntactic function of each 
word and therefore it is not possible to include ev- 
ery variation of a word in the dictionary. So, other 
methods have to be tried for languages that use ex- 
tensively prefixes, infixes or suffixes. 
As a starting point, let us show what the declen- 
sion of a word in Basque may be, by means of an ex- 
ample. The declension of the dictionary entry mendi 
(which means "mountain") can be seen in Table 1. 
This table is valid only for words referring to ob- 
jects, but there are different tables for declensions of 
words referring to living beings. Whether the last 
letter of the lemma is a vowel or a consonant, dif- 
ferent tables of declensions are also used. As shown, 
there are sixty-two possible word-forms for a single 
dictionary entry. In addition, most of the cases ad- 
mit the recursive concatenation of suffixes. So, the 
number of possible cases grows. It has been esti- 
mated that nouns may mathematically have even 
458,683 inflected forms in Basque language, taking 
into account two levels .of recursion, (Agirre et al., 
1992). 
There also are other suffixes which are not shown 
in Table 1, as those applied to a verb for subordinate 
sentences. 
Even if prefixes and infixes are possible the Basque 
language is declensed mainly by suffixes. There are 
some prefixes that can be used in some specific cases 
(for example, a prefix for verbs may indicate the 
absolutive case in the sentence), but in general their 
frequency of apparition is not very relevant. The 
same thing happens with the infixes: there are few 
of them in Basque and their frequency is not very 
relevant. Their prediction makes sense mainly if the 
29 
Absolutive 
Ergative 
Dative 
Possessive 
Genitive 
Comitative 
Benefactive 
Motivative 
Instrumental 
Inessive 
Indefinite 
mendi 
mendik 
mendiri 
mendiren 
mendirekin 
mendirentzat 
mendirengatik 
mendiz 
menditan 
Definite 
Singular 
mendia 
mendiak 
mendiari 
mendiaren 
mendiarekin 
mendiarentzat 
mendia(ren)gatik 
mendiaz 
mendian 
Definite 
Far Plural 
mendiak 
mendiek 
mendiei 
mendien 
mendiekin 
mendientzat 
mendiengatik 
mendiez 
mendietan 
Definite 
Near Plural 
mendiok 
mendiok 
mendioi 
mendion 
mendiokin 
mendiontzat 
mendiongatik 
mendioz 
mendiotan 
Ablative 
Allative 
Allative 
of Destination 
menditatik 
menditara 
menditaraino 
menditik 
mendira 
mendiraino 
mendietatik/rik 
mendietara 
mendietaraino 
mendiotatik/rik 
mendiotara 
mendiotaraino 
Allative menditarantz mendirantz mendietarantz mendiotarantz 
of Direction 
menditako mendiko mendietako mendiotako Local 
Genitive 
Destinative menditarako mendirako 
Partitive mendirik 
mendietarako mendiotarako 
Prolative menditzat 
Table 1: Declension of mendi (Kintana et al., 1988). 
word is an auxiliary or a declined verb. For the 
rest of the cases, it seems better to treat the affix in 
combination with the lemma as a new lemma, if this 
combination is usual. Doing this, the complexity of 
operations decrease because there is only the need 
to treat lemmas and suffixes. 
Thus, in this paper, the problem of suffixes will 
mainly be mentioned, because our target language 
is the Basque language. 
2 Word Prediction Methods for 
Non-Inflected Languages 
In this section some of the methods that have been 
used in word prediction for non-inflected languages 
are summarised. This small review will serve as a 
basis for coming sections in order to identify the key 
aspects that are involved in prediction. These meth- 
ods are going to be presented by increasing complex- 
ity, from the simplest to the most complex. 
2.1 Probabilistic Methods 
2.1.1 Word Prediction Using Frequencies 
The simplest word prediction method is to built 
a dictionary containing words and their relative fre- 
quencies of apparition. When the user starts typing 
a string of characters a the predictor offers the n 
most frequent words beginning by this string in the 
same way they are stored in the system. Then, the 
user can choose the word in the list he or she wanted 
to enter or continue typing if it is not in the list. 
There are several studies about word frequencies in 
different languages, for instance (Beukelman et al., 
1984) gives information about the frequency of word 
occurrence in English used by some disabled people. 
If the dictionary does not contain inflected words 
(that is, if there are just the lemmas), it may need 
some correction by the user (or by the system) in 
order to adjust its concordance with other related 
words. For instance, it may need to adjust the gen- 
der: "C'est une voiture fantastique" or the number 
"A lot of cars". The dictionary uses to be an alpha- 
betically ordered list of words and their frequencies, 
but other possible dictionary structures can be found 
in (Swiflin et al., 1987a). This prediction system can 
be adapted to the user by updating the frequency 
of the word in the dictionary each time this word 
is used. Words seldom employed can be replaced 
by others which are not in the dictionary. The in- 
clusion of new words is not difficult because all the 
information that is required is their frequency. Fur- 
ther information about this type of prediction can be 
seen in (Colby et al., 1982), (Garay et al., 1994a), 
(Heckathorne et al., 1983), (Hunnicutt, 1987), (Swif- 
30 
I 
fin et al., 1987a), (Venkatagiri, 1993). 
To enhance the results of this method, an indica- 
tion about the "recency" of use of each word may be 
added. In this way, the prediction system is able to 
offer the most recently used Words among the most 
probable ones beginning by a. Each entry in the 
dictionary is composed by a word, its frequency and 
its recency of use. Adaptation of the dictionary to 
the user's vocabulary is possible by updating the fre- 
quency and recency of the each word used. (Swiffin 
et al., 1987a) observes that this method produces 
small better savings in the number of keystrokes 
needed than in the previous approach, but more in- 
formation must be stored in the dictionary and the 
complexity is also increased. 
2.1.2 Word Prediction Using Probability 
Tables 
Another possibility is to use the relative probabil- 
ity of appearance of a word depending on the pre- 
vious one. To implement this system a two-entries 
table is needed to store the conditional probability 
of apparition of each word Wj after each Wi. If the 
dictionary contains N words the dimension of the 
table will be of N*N. That is, it will have N 2 en- 
tries, but most of the values in the table will be zero 
or close to zero. 
In some cases it could be possible for the system 
to give proposals before entering the beginning of a 
word. The recency of use may also be included in 
this approach. This method is hardly adaptable to 
include the user preferred words because the dimen- 
sions of the table cannot be changed. This difficulty 
leads to the design of modified versions, like the one 
that uses only the most probable pair of words, re- 
ported as in (Hunnicutt, 1987). 
2.2 Syntactic Word Prediction 
2.2.1 Syntactic Word Prediction Using 
Probability Tables 
This approach takes into account the syntactic in- 
formation inherent to the languages. To this end two 
l~inds of statistical data are used: the frequency of 
apparition of each word and the conditioned proba- 
bility of each syntactic category to follow every other 
syntactic category. In this way, the set of words that 
are candidates to be proposed by the predictor is re- 
stricted to the ones that match the most probable 
syntactic role in the current position of the sentence, 
thus increasing the hint rates. This syntactic table is 
smaller than the one used in the previous approach, 
and the proportion of probabilities which are close 
to zero is also smaller. Each entry in the dictionary 
will associate a word with its syntactic category, and 
its frequency of apparition. Words can be sorted by 
syntactic categories to facilitate the selection pro- 
cess. When a word is syntactically ambiguous, that 
is, when more than one category is possible for a 
given word, one entry for each possible category may 
be created. The table of conditional probabilities of 
syntactic categories has a fixed size and it is built 
before the use of the predictor. Adaptation to the 
user's lexicon is possible because there is no need 
to increase the size of the table. New words are in- 
cluded in the dictionary with a provisional syntactic 
category deducted from its use. Later on, the system 
may require some help from the user to verify if the 
categorisation was correct. It could be also possible 
to add some morphological information in the dictio- 
nary to propose the words with the most appropri- 
ate morphological characteristics (gender, number). 
This could increase the hint rate of the predictor. 
Some systems that use this approach are described 
in (Garay et al., 1994a) and (Swiffin et al., 1987b). 
2.2.2 Syntactic Word Prediction by Using 
Grammars 
In these approaches, the current sentence is being 
parsed using a grammar to get the most probable 
categories. Parsing methods for word prediction can 
be either "top-down" (Van Dyke, 1991) or "bottom- 
up" (Garay et al., 1994b), (Garay et al., 1997). So, 
there is a need to define the syntactic rules (typ- 
ically LEFT <- \[RIGHT\]+, usually being LEFT 
and RIGHT some syntactic categories defined in the 
system) that are used in a language. Within a rule, 
it could be possible to define concordance amongst 
the components of the right part (either in gender 
and/or in number). Then, the proposals may be of- 
fered with the most appropriate morphological char- 
acteristics. It is necessary to leave open to the user 
the possibility of changing the word's ending. For ex- 
ample, if there is a mismatch in the rule used by the 
system, it may be necessary to modify the end of an 
accepted proposal. The dictionary is similar to the 
one used in the previous approach with the addition 
of morphological information to allow concordance. 
The complexity of this system is also larger because 
in this case, all the words of the sentence that ap- 
pear before the current word are taken into account, 
while in the previous approaches only one previous 
word was used. The adaptation of the system for 
the new words is made increasing the word frequen- 
cies and the weights of the rules. The inclusion of 
new words is similar to the one in the previous ap- 
proach. The use of grammars for word prediction is 
also shown in (Hunnicutt, 1989), (Le P~v~dic et al., 
1996), (Morris et al., 1991) and (Wood et al., 1993). 
31 
2.3 Semantic Word Prediction 
These methods are not very used, because their re- 
sults are similar to those of the syntactic approaches, 
but the increase in complex!ty is great. Maybe the 
simplest method that can be used is the semantic 
word prediction by using parsing methods. In this 
approach each word has some associated semantic 
categories, while in the previous one categories were 
purely syntactic. The rest of the features (the proce- 
dure, complexity, structure of the dictionary, adapt- 
ability...) are similar to the previous one. Never- 
theless, the problem of giving semantic categories 
to the words is very complex and it results difficult 
to be programmed. Some authors propose semantic 
categorisation made "by hand" (Hunnicutt, 1989). 
There may be other methods to treat the seman- 
tic information, but their complexity is going to be 
very great for a real-time system as the word pre- 
dictors are intended to be, even the time require- 
ments (maybe a few seconds between two consecu- 
tive keystrokes of an impaired person) are not very 
strong for the computational capacities of today's 
• equipment. 
3 Application of Mentioned Word 
Prediction Methods to Inflected 
Languages 
In this section the use of previously reviewed word 
prediction methods for non-inflected languages is 
studied and their suitability for inflected languages 
is discussed. So, the key question is: Are the word 
prediction methods that we have previously shown 
useful for inflected languages? 
As we mentioned in the introduction, in non- 
inflected languages it is feasible to include in the 
dictionary all the forms derived from each lemma, 
taking into account that the number of variations is 
quite small. For instance, in English friends is the 
only variation (without creating composed words) 
of friend, and the verbs have a few variations too. 
• In Spanish, the word amigo (with the same mean- 
ing than friend) may vary in gender and number, 
giving the words: amiga, amigos and amigas. But 
the variations that the word adiskide (same mean- 
ing as friend or amigo) may have in Basque make 
it impossible to store them in the dictionary of the 
system. This is one of the changes to be taken into 
account for the design of a predictor for this type 
of languages. In inflected languages, the complexity 
in making the changes is very high, because of the 
number of possibilities. One possibility is to group 
the suffixes depending on their syntactic function to 
make it possible to have an easy automatisation. In 
addition, we shouldn't forget that suffixes may be 
recursively concatenated. 
In the previously presented prediction methods, 
the ones using probabilistic information mainly work 
with the words as isolated entities. That is, they 
work seeing each word in the dictionary as a whole to 
be guessed, without taking into account the morpho- 
syntactical information inherent to the languages. 
So, a word that is not at the lexicon can not be 
guessed. The impossibility to store all the combina- 
tions of a word, make these methods not very suit- 
able for inflected languages 2. 
Therefore, it would be very interesting to treat the 
entire sentence. Then, the first syntactic approach 
is not very useful, because it only takes into account 
the previous word. And the second one is very hard 
to implement, because of the number of variations a 
word may have. Maybe a great number of rules have 
to be defined to cope with all the variations, but in 
this way the probabilities to guess the rule which is 
being used are very small, because of their variety. 
The same thing happens with the semantic ap- 
proach, which has, as it has been said before, the 
same procedural characteristics as the second syn- 
tactic one. 
So, the complexity needed to create a correct 
word, including all the suffixes it needs, in inflected 
languages may make it necessary to search for other 
prediction methods, apart from all that were shown 
in the previous section. 
2To know what the suitability for the next shown ap- 
proaches can he, let us show a special case for Basque: 
verbs, mainly auxiliary verbs. They depend not only 
on the subject (which normally appears as absolutive or 
ergative cases) but also on the direct complement (if the 
sentence is transitive this complement has the absolutive 
case while the subject has the ergative case) and on the 
indirect complement (the dative case). For instance, the 
auxiliary dizut is related to the subject of the first per- 
son singular, the object complement in the singular and 
the indirect complement of the second person singular. 
But if the subject is in the third person plural, the indi- 
rect complement in the first person plural and the direct 
complement is in the plural, the needed auxiliary has to 
be dizkigute. Both cases are in the present of the indica- 
tive. If the tense of the verb changes, the verb itself also 
changes (for example, the past of the indicative of dizut 
is nizun and the past of dizkigute is zizkiguten). There 
also are some cases in which the verb depends on the 
gender of the absolutive, ergative or dative cases. 
32 
! 
4 Word Prediction in an Inflected 
Language. Application to Basque 
Language 
4.1 First Approach to Solving the 
Prediction Problem in an Inflected 
Language 
As we have seen in the previous section, it is very 
difficult to predict complete words in inflected lan- 
guages because of the variations a word may have. 
As there is a huge variety of inflected languages, let 
us concentrate on the particular characteristics of 
the Basque language, customising to this case the 
operational way. 
For this first approach, due to the above men- 
tioned primacy of suffies (over other affixes) in the 
Basque language, and to simplify the problem, pre- 
diction in Basque is divided in two parts: prediction 
of lemmas and prediction of suffixes. Thus, two dic- 
tionaries (one for lemmas and other for suffixes) are 
used. The first one includes the lemmas of the lan- 
guage alphabetically ordered with their frequencies 
and some morphologic information in order to know 
which possible declensions are possible for a word. 
The second one includes suffixes and their frequen- 
cies ordered by frequencies. 
To start the prediction, the system tries to an- 
ticipate the lemma of the next word. Most of the 
methods seen in previous sections can be used for 
this purpose. When the lemma is accepted (or typed 
entirely if the predictor fails), the system offers the 
suffixes that are correct for this lemma ordered by 
frequencies. As the acceptable suffixes for a noun 
can be about 62 (as we have seen in the Table 1) 
only the most probable n suffixes are offered 3. As 
can be seen, the operational way is very similar to 
word prediction using tables of probabilities, but 
there is some added complexity because the system 
(and also the user) has to distinguish between lem- 
mas and suffixes. In addition, more than one table 
of probabilities may be necessary to properly make 
predictions. Apart from the increase of the com- 
plexity, a decrease of the keystroke savings may be 
expected, because of the need of accepting at least 
two proposals for completing a word (while at least 
only one proposal is required with predictors for non- 
inflected languages). 
Even if some promising results have been ob- 
tained, there are still some problems to solve in this 
approach. 
• First of all, due to the possibility of recursively 
composed suffixes (concatenating the existing 
aWith n depending on the interaction method 
ones) the system has to again propose a list of 
suffixes until the user explicitly marks the end 
of the current word (maybe inserting a space 
character). 
• The recursive behaviour is one of the reasons 
to create more than one table of probabilities 
which stores the probability of apparition of a 
suffi immediately after the previous one. 
• The system may be adapted to the user updat- 
ing the frequencies in the lexicons and the prob- 
abilities of the tables. To include a new lemma 
in the dictionary, it is necessary to obtain its 
morphological characteristics. 
• Finally, due to the special characteristics of the 
verbs (that include any kind of affixes in concor- 
dance with other words in the entire sentence) 
their prediction requires a special treatment. 
Therefore, it seems interesting to do a syntax ap- 
proach for these types of languages, because other- 
wise, the problems of this approach are very dimcult 
to solve. 
4.2 Second Approach to Solving the 
Prediction Problem in an Inflected 
Language 
This approach will try to alleviate the above men- 
tioned problems. The lemmas and the suffies are 
still treated separately, but syntactical information 
is included in the system. This can be done by 
adding syntactic information to the entries of the 
dictionary of lemmas, and some weighted grammat- 
ical rules on the system. The main idea is to parse 
the sentence while it is being composed and to pro- 
pose the most appropriate lemmas and suffixes. In 
principle, the parsing allows storing/extracting the 
information that has influenced in forming the verb. 
There exist systems that verify the morphologic and 
syntactical correctness of a Basque sentence, but the 
complexity of the Basque verb avoids its anticipa- 
tion. To face this problem, the most frequent verb 
forms are included in the dictionary, and a morpho- 
logical generator permits their modification or the 
addition of suffixes when it is necessary. 
As there are no probability tables, there is no 
problem related to their extension. The adaptation 
of the system is made by updating the frequencies 
of the lemmas and suffies and the weights of the 
defined rules. The inclusion of a new lemma in the 
lexicon might cause some lack of syntactic informa- 
tion. To solve this problem, there are some possi- 
bilities. First, the predictor tries to guess the cat- 
egory, depending on the most highly weighted rule 
33 
at that point of the sentence. Second, the predictor 
asks the user directly about the information. The 
first approach can produce false assumptions, while 
the second one slows the message composition rate 
and demands a great knowledge of the syntax by 
the user. There is another possibility: the predictor 
marks the lemma and the user is asked to complete 
the needed information after ending the session. 
Finally, recursion may be included into the de- 
fined rules. Most of the grammars may have an im- 
plicit recursion which may be shown by rules. For 
instance, let us consider these rules: 
NP <- Noun PP 
PP <- Prep NP, 
where NP means Noun Phrase, PP, Prepositional 
Phrase, Noun is a noun and Prep, a preposition. As 
can be seen, these rules can be expanded to: 
NP <- Noun Prep NP. 
So, the NP is on the left and on the right of the 
same rule, and a recursion happens. This recursion 
may be used as a way to indicate the recursion of 
the concatenation of the suffixes, because they can 
express the syntactic role of a word in a sentence, as 
it was noted in the introduction. 
The operational way and the order of complex- 
ity are similar to the word prediction using gram- 
mars. Nevertheless higher complexity may be ex- 
pected mainly due to the existence of lemmas and 
suffixes. So, poorer keystroke savings are expected. 
To enhance this approach, it seems interesting to 
try to guess the entire word, that is, a lemma and its 
associated suffix. This system will be easier to use 
(there is no need to force users to know what the 
lemma and what the suffix of a word are) and may 
have better results, measured in terms of keystroke 
savings. 
4.3 Third Approach to Solving the 
Prediction Problem in an Inflected 
Language 
Taking into account the previous experience, a third 
approach could be tried. Built as a combination of 
the previous ones, the main idea is to guess the entire 
current word. It treats the beginning of the sentence 
like the first approach, using statistical information. 
While advancing in the composition of the sentence, 
the system parses it and uses this information to of- 
fer the most probable words, including both lemma 
and suffix, like the second approach does. The first 
word of the sentence is treated using the first ap- 
proach seen. But to minimise the problems related 
to that approach, tee rest of the sentence is treated 
using the second approach. 
In this way, only three tables would be needed: 
one with the probabilities of the syntactic categories 
of the lemmas to appear at the starting of a sen- 
tence, another with the probabilities of the basic 
suffixes to appear after those words and the third 
with the probabilities of the basic suffixes to appear 
after another basic suffix (and to make possible the 
recursion). All of these tables would have fixed sizes 
even when new lemmas are added to the system. 
The adaptation of the system would be made up- 
dating the first table and, while the suffixes would 
be added to the word, the other two tables would be 
also updated. With relation to the new lemmas that 
do not have the information completed, they might 
update, or not, the first of the tables if a entry for 
the unknown cases is included; otherwise they would 
remain unchanged. Finally, the problem of verb for- 
mation in Basque is not solved and the most frequent 
verb forms are included in the dictionary in the same 
way as in the second approach. 
5 Conclusions 
Our experience with the Basque language in word 
prediction applied to Alternative and Augmentative 
Communication for people with disabilities, shows 
that prediction methods successful for non-inflected 
languages are hardly applicable to inflected ones. 
The high number of in flexions for each word makes 
their inclusion in the lexicon impossible. Different 
approaches have been studied to overcome this prob- 
lem. To be able to predict whole words it is neces- 
sary to determine the syntactic role of the next word 
in the sentence. That can be done by means of a syn- 
tactic analysis "on the fly". Nevertheless the results 
of the evaluation of these methods with the Basque 
language are not as good as the ones obtained with 
non-inflected languages. 
6 Acknowledgements 
The authors would like to acknowledge the work of 
the rest of the members of the Laboratory of Human- 
Computer Interaction of the Computer Science Fac- 
ulty of the University of the Basque Country. They 
also would like to acknowledge the aid given by Jose 
Marl Arriola, Kepa Sarasola and Ruben Urizar, who 
work in the IXA Taldea of the Computer Science 
Faculty above mentioned. 

References 
Eneko Agirre, Ifiaki Alegria, Xabier Arregi, Xa- 
bier Artola, Arantza Diaz de Ilarraza, Montse 
Maritxalar, Kepa Sarasola, and Miriam Urkia. 
1992. XUXEN: A Spelling Checker/Corrector for 
Basque Based on Two-Level Morphology. In Pro- 
ceedinys of the 3rd Conference on Applied Nat- 
ural Language Processing, pages 119-125, Trento, 
Italy, March-April. Association for Computational 
Linguistics. 
David R. Beukelman, Kathryn M. Yorkston, Miguel 
Poblete and Carlos Naranjo. 1984. Frequency 
of Word Occurrence in Communication Samples 
Produced by Adult Communication Aid. Journal 
of Speech and hearing Disorders, Vol. 49: 360-367. 
Kenneth M. Colby, Daniel U. Christinaz, Roger C. 
Parkison, Mark Tiedemann. 1982. Predicting 
Word-Expressions to Increase Output Rates of 
Speech Prosthesis Used in Communication Disor- 
ders. In IEEEProc. ICASSP-8~, pages 751-754. 
Nestor Garay and Julio Gonz~lez Abascal. 1994. 
Using Statistical and Syntactic Information in 
Word Prediction for Input Speed Enhancement. 
In Information Systems Design and Hyperme- 
dia edited by Claude Chrisment, pages 223-230. 
C~padu~s-l~ditions. Toulouse (France). 
Nestor Garay-Vitoria and Julio Gonz~ilez Abascal. 
1994. Application of Artificial Intelligence Meth- 
ods in a Word-Prediction Aid. In Computers for 
Handicapped Persons edited by W. L. Zagler, G. 
Busby and R.R. Wagner, pages 363-370. Springer- 
Verlag. Berlin. 
Nestor Garay-Vitoria and Julio Gonz~ilez Abascal. 
1997. Intelligent Word-Prediction to Enhance 
Text Input Rate. In Proceedings of the Intelli- 
gent User Interfaces 97 International Conference, 
pages 241-244,: Orlando (Florida), January. ACM 
Press, New York. 
Craig W. Heckathorne, Dudley S. Childress. 1983. 
Applying Anticipatory Text Selection in a Writ- 
ing Aid for People with Severe Motor Impairment. 
IEEE MICRO, June, pages 17-23. 
Sheri Hunnicutt. 1987. Input and Output Alterna- 
tives in Word Prediction. STL/QPSR 2-3/1987, 
pages 17-29. Dept. of Speech Comm. and Music 
Acoustics. Royal Institute of Tech. Stockholm. 
Sheri Hunnicutt. 1989. Using Syntactic and Se- 
mantic Information in a Word Prediction Aid. In 
Proceedings of the European Conference on Speech 
Communication and Technology "Eurospeech 89", 
Vol. 1: 191-193. Paris, September. 
Xabier Kintana, Jon Aurre, Ricardo Badiola, Sabin 
Egileor, Juan Luis Goikoetxea, Juanjo Zearreta, 
Koldo Zuazo, Ekiten taldeko irakasleak. 1988. 
Hiztegia Bi Mila: Euskara-Espainiera/Espainiera- 
Euskara. ELKAR, Donostia, Basque Country. 
Brigitte Le P~v~dic and Denis Maurel. 1996. 
Un Logiciel d'Aide a la Communication pour 
des Personnes Handicape~s. In Proceedings of 
the NLP+IA '96 (TAL÷AI '96), pages 211-216. 
Moncton, N. B., Canada. 
Corinne Morris, Alan Newell, Lynda Booth and 
John Arnott. 1991. SYNTAX PAL-A System to 
Improve the Syntax of Those with Language Dys- 
function. In Proceedings of RESNA 14th Annual 
Conference, pages 105-106, Kansas City. 
Andrew L. Swiffin, John L. Arnott, J. Adrian Pick- 
ering and Alan F. Newell. 1987. Adaptive and 
Predictive Techniques in a Communication Pros- 
thesis. Augmentative and Alternative Communi- 
cation, pages 181-191. 
Andrew L. Swiffin, John L. Arnott and Alan F. 
Newell. 1987. The Use of Syntax in a Predic- 
tive Communication Aid for the Physically Hand- 
icapped. In Proceedings of RESNA lOth Annual 
Conference, pages 124-126, San Jose, Ca. 
Julie A. Van Dyke. 1991. Word Prediction for Dis- 
abled Users: Applying Natural Language Process- 
ing to Enhance Communication. Thesis submit- 
ted for the degree of Honors Bachelor of Arts in 
Cognitive Studies. University of Delaware, June. 
Horaball S. Venkatagiri. 1993. Efficiency of Lexi- 
cal Prediction as a Communication Acceleration 
Technique. AAC, Vol. 9: 161-167. 
Matthew E. J. Wood, Eric Lewis. 1993. Gram- 
matical Recognition in Computer Aided Conver- 
sation. In Proceedings of the European Speech 
Communication Association (ESCA) Workshop 
on Speech and Language Technology for Disabled 
People, pages 115-118. Stockholm, Sweden, May- 
June. 
