Ling, C. X. #1994#. Learning the past tense of English verbs: The symbolic pattern associator vs. connectionist models. Journal of Arti#cial Intelligence Research 1, 209#229. Ling, C. X. #1995#. Introducing new predicates to model scienti#c revolution. International Studies in the Philosophy of Science.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....Table 2: Results of different sizes of partial 10 models on BNC test data. The single mistake was on the pair (forbid,forbad) The model gave the regular past. 5. Experiment 1: English past tense For this experiment I used the standard English Past Tense data set , available as an appendix to (Ling, 1994), and also used by Mooney and Califf (1995) and various other researchers, which consists of 1389 pairs of words in the UNIBET phonetic alphabet. For test data I selected some unseen verbs from the British National Corpus(Aston Burnard, 1998) I chose all verbs that occurred ten times with the ....

....However the role of blocking and the choice between the regular and irregular forms depends crucially on the relative probabilities of the two transition sequences. 9.2 Comparison with Symbolic learning models Various symbolic learning algorithms have been presented for the English past tense. Ling (1994) proposes a symbolic pattern associator that he claims performs clearly better than the connectionist models. Unfortunately it suffers from many of the same problems. Mooney and Califf (1995) presents an inductive logic programming approach (Muggelton, 1999; Muggleton Bain, 1999) that performs ....

Ling, C. X. (1994). Learning the past tense of english verbs: The symbolic pattern associator vs. connectionist models. Journal of Artifical Intelligence Research, 1, 209--229.

....by the examples sing sang, ring rang, and spring sprang. Decision tree algorithms were applied to this task and found to signi cantly outperform previous neural network models in terms of producing correct past tense forms for independent test words (Ling Marinov, 1993; Ling, 1994). In this study, verbs were restricted to 15 phonemes encoded using the UNIBET ASCII standard, and 15 separate trees were induced, one for producing each of the output phoneme positions using all 15 of the input phonemes as features. Below is the encoding for the mapping act acted, where ....

Ling, C. X. (1994). Learning the past tense of English verbs: The symbolic pattern associator vs. connectionist models. Journal of Articial Intelligence Research, 1, 209-229.

....Learning from positive data The second problem is of a different nature. When learning from only positive data, predictive accuracy will be maximised by choosing the most general consistent hypothesis since this will always agree with new data. However, in applications such as grammar learning [25, 50], only positive data are available, though the grammar which produces all strings is not an acceptable hypothesis. Let us then suppose a modification to the U learning setting given in Appendix B. The teacher still draws instances randomly from distribution G but only gives them to the learner if ....

C.X. Ling. Learning the past tense of english verbs: the symbolic pattern associators vs. connectionist models. Journal of Artificial Intelligence Research, 1:209--229, 1994.

....and kaz man.tex; 15 06 2000; 17:55; p. 3 4 Bain, 1999) When trying to predict the past tense form of an English verb from its present, the current ILP approaches outperforms both the best application of connectionist approach (Rumelhart and McClelland, 1986) and propositional decision trees (Ling, 1994) known so far. The theories learnt come in two flavours, as first order decision trees (Mooney and Califf, 1995; Manandhar et al. 1998) or pure first order logic clauses (Muggleton and Bain, 1999) Of these two approaches, the former is a product of eager learning, whereas the latter is a halfway ....

Ling, C. X.: 1994, `Learning the Past Tense of English Verbs: The Symbolic Pattern Associatior vs. Connectionist Models'. Journal of Artificial Intelligence Research 1, 209--229.

....Figure 7: Form of examples and hypotheses for past tense domain 4.4 Morphology Mooney and Califf [18] have applied ILP to learning the past tense of English verbs. Learning of English past tense has become a benchmark problem in the computational modelling of human language acquisition [29, 14]. In [18] it was shown that a particular ILP system, FOIDL, could learn this transformation more effectively than previous neural network and decision tree methods. FOIDL s first order default rule style representation was demonstrated by the authors as producing a predictive accuracy advantage in ....

C.X. Ling. Learning the past tense of english verbs: the symbolic pattern associators vs. connectionist models. Journal of Artificial Intelligence Research, 1:209--229, 1994.

....Given a target and a context, determine whether and which boundary is associated with this target. Examples include syllabification, morphological analysis, 2 This restriction can be circumvented by having multiple classifiers predict a different part of the output pattern, see Ling ([1994]) for this approach in learning decision trees. syntactic analysis (in combination with tagging) etc. An approach often necessary to arrive at the context information needed is windowing approach (as in Sejnowski and Rosenberg [1986] for text to speech) in which an imaginary window is moved ....

....were replicated for English, French and Dutch, using the same lazy learning algorithm, which shows its reusability. 5. 3 Word Stress Assignment Another task we applied the lazy learning algorithm to, was stress assignment in Dutch monomorphematic, polysyllabic words (Daelemans et al. 1993] [1994]) A word was coded by assigning one feature to each part of the syllable structure of the last three syllables (if present) of the word (see the description of the diminutive formation task described earlier) There were three categories: final stress, penultimate stress, and antepenultimate ....

[Article contains additional citation context not shown here]

Ling, C.: Learning the past tense of English verbs: The symbolic Pattern Associator vs. Connectionist Models. Journal of Artificial Intelligence Research 1 (1994) 209--229.

....whether and which boundary is associated with this target. Examples include syllabification, morphological analysis, syntactic analysis (in combination with tagging) etc. 2 This restriction can be circumvented by having multiple classifiers predict a different part of the output pattern, see [18] for this approach in learning decision trees. An approach often necessary to arrive at the context information needed is the windowing approach (as in [22] for text to speech) in which an imaginary window is moved one item at a time over an input string where one item in the window (usually ....

Ling, C.: Learning the past tense of English verbs: The symbolic Pattern Associator vs. Connectionist Models. Journal of Artificial Intelligence Research 1, (1994) 209--229.

....from significantly fewer examples than all previous methods. 1 Introduction The problem of learning the past tense of English verbs has been widely studied as an interesting subproblem in language acquisition. Previous research has applied both connectionist and symbolic method to this problem [22, 12, 9]; however, previous efforts used specially designed feature based encodings that impose a fixed limit on the length of words and fail to capture the generativity and position independence of the underlying transformation. We believed that representing the problem as constructing a logic program ....

....few negatives. The papers referenced above provide details and information on additional features. 2. 2 Learning the Past Tense of English Verbs The problem of learning the English past tense has been attempted by both connectionist systems [22, 12] and systems based on decision tree induction [11, 9]. The task to be learned in these experiments is: given a phonetic encoding of the base form of an English verb, generate the phonetic encoding of the past tense form of that verb. The task can also be done using the alphabetic forms forms of the verbs, and we use that form of the task for the ....

[Article contains additional citation context not shown here]

C. X. Ling. Learning the past tense of English verbs: The symbolic pattern associator vs. connectionist models. Journal of Artificial Intelligence Research, 1:209--229, 1994.

....improvement in prediction accuracy. 6. 4 Past tense of English verbs Learning how to transform a verb from present to past tense was first studied in the connectionist community, but relational representations of the task have also been explored [11, 17, 19] These experiments imitate Ling s [10] in using ten training and test sets of verbs taken from a corpus of 1500 phonetic verbs. This formulation used two target relations, delete(A; B) and add(A; C) which jointly state that the past tense of verb A is found by removing the string B from its end and adding the string C. A single ....

Ling, C.X.: Learning the past tense of English verbs: the symbolic pattern associator versus connectionist models. Journal of Artificial Intelligence Research 1 (1994) 209-229

....by a failure we observed when applying existing ILP methods to a particular problem, that of learning the past tense of English verbs. This problem has been studied fairly extensively using both connectionist and symbolic methods (Rumelhart McClelland, 1986; MacWhinney Leinbach, 1991; Ling, 1994); however, previous efforts used specially designed feature based encodings that impose a fixed limit on the length of words and fail to capture the position independence of the underlying transformation. We believed that representing the problem as constructing a logic program for the predicate ....

.... Bever, 1988) MacWhinney and Leinbach (1991) attempted to address some of these criticisms by using a standard multi layer backpropagation learning algorithm and a simpler UNIBET encoding of phonemes (in which each of 36 phonemes is encoded as a single ASCII character) Ling and Marinov (1993) and Ling (1994) criticize all of the current connectionist models of past tense acquisition for heavily engineered representations and poor experimental methodology. They present more systematic results on a system called SPA (Symbolic Pattern Associator) which uses a slightly modified version of C4.5 (Quinlan, ....

[Article contains additional citation context not shown here]

Ling, C. X. (1994). Learning the past tense of English verbs: The symbolic pattern associator vs. connectionist models. Journal of Artificial Intelligence Research, 1, 209--229.

....data sets. Note that hypothesis 1 is not the negation of hypothesis 2. If both are rejected then it means simply that AP is better for some domains but not others. 4.2 Materials The following data sets were used for testing the experimental hypotheses. English past tense. This is described in [15, 6, 7]) The available example set E past has size 1390. KRK illegality. This was originally described in [9] The total instance space size is 8 6 = 262144. For both domains the form of examples are shown in Figure 3 and the form of hypothesised clauses in Figure 4. Note that in the KRK illegality ....

C.X. Ling. Learning the past tense of english verbs: the symbolic pattern associators vs. connectionist models. Journal of Articial Intelligence Research, 1:209-229, 1994.

....verb in phonetic notation from present to past tense, has more of a real world flavor in that any totally correct definition would be extremely complex. A considerable literature has built up around this task, starting in the connectionist community, moving to symbolic learning through the work of Ling (1994), then to relational learning (Quinlan, 1994; Mooney and Califf, 1995) Quinlan (1994) proposes representing this task as a relation past(A,B,C) interpreted as the past tense of verb A is formed by stripping off the ending B and then adding string C. The single background relation split(A,B,C) ....

....as a relation past(A,B,C) interpreted as the past tense of verb A is formed by stripping off the ending B and then adding string C. The single background relation split(A,B,C) shows all ways in which word A can be split into two non empty substrings B and C. Following the experiment reported in (Ling, 1994), a corpus of 1391 verbs is used to generate ten randomly selected learning tasks, each containing 500 verbs from which a definition is learned and 500 different verbs used to test the definition. A Prolog interpreter is used to evaluate the definitions learned by foil, each unseen word w being ....

Ling, C. X. (1994). Learning the past tense of english verbs: the symbolic pattern associator versus connectionist models. Journal of Artificial Intelligence Research, 1, 209--229.

.... the success and cognitive plausibility of different approaches (Rumelhart and McClelland (1986) MacWhinney and Leinbach (1991) arguing for connectionist models, Pinker and Prince (1988) Lachter and Bever (1988) Marcus et al. 1992) arguing against connectionist models, Ling and Marinov (1993) Ling (1994) using ID3 C4.5 decision trees, and Mooney and Califf (1995, 1996) using inductive logic programming decision lists, among others) However except for a couple of forays into German this literature has been exclusively concerned with the learning of the English past tense. This has not ....

Ling, C. X. 1994. Learning the past tense of english verbs: the symbolic pattern associator vs.

....whether the word being shown in the phonology units referred to a [ Human] concept or not. If it did and the referent was [ Male] then the first four units were set to an active 14 A similar representational problem was solved in a similar fashion by both MacWhinney Leinbach (1991) and Ling (1994) in their models of the English past tense formation, despite the fact that the former used a connectionist network and the latter a rule driven symbolic learning algorithm. Modeling Latin to French Gender Reanalysis 25 value (1) and the final four were set to 0. With a [ Female] referent these ....

Ling, C.X. 1994. Learning the Past Tense of English Verbs: The Symbolic Pattern Associator vs.

No context found.

Ling, C. X. #1994#. Learning the past tense of English verbs: The symbolic pattern associator vs. connectionist models. Journal of Arti#cial Intelligence Research 1, 209#229. Ling, C. X. #1995#. Introducing new predicates to model scienti#c revolution. International Studies in the Philosophy of Science.

No context found.

Ling, C.X., Learning the past tense of English verbs: the symbolic pattern associator versus connectionist models, Journal of Artificial Intelligence Research 1 (1994) 209-229.

No context found.

C. X. Ling (1994). Learning the past tense of English verbs: the symbolic pattern associator versus connectionist models. Journal of Artificial Intelligence Research, 1, 209-229.

No context found.

C.X. Ling. Learning the past tense of english verbs: The symbolic pattern associator vs. connectionist models. Journal of Artificial Intelligence Research, 1:209--229, 1994.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC