Evolutionary induction of stochastic context free grammars

This paper describes an evolutionary approach to the problem of inferring stochastic context-free grammars from finite language samples. The approach employs a distributed, steady-state genetic algorithm, with a fitness function incorporating a prior over the space of possible grammars. Our choice of prior is designed to bias learning towards structurally simpler grammars. Solutions to the inference problem are evolved by optimizing the parameters of a covering grammar for a given language sample. Full details are given of our genetic algorithm (GA) and of our fitness function for grammars. We present the results of a number of experiments in learning grammars for a range of formal languages. Finally we compare the grammars induced using the GA-based approach with those found using the inside-outside algorithm. We find that our approach learns grammars that are both compact and fit the corpus data well.     

Learning deterministic context free grammars: The Omphalos competition

This paper describes the winning entry to the Omphalos context free grammar learning competition. We describe a context-free grammatical inference algorithm operating on positive data only, which integrates an information theoretic constituent likelihood measure together with more traditional heuristics based on substitutability and frequency. The competition is discussed from the perspective of a competitor. We discuss a class of deterministic grammars, the Non-terminally Separated (NTS) grammars, that have a property relied on by our algorithm, and consider the possibilities of extending the algorithm to larger classes of languages. Editor: Georgios Paliouras and Yasubumi Sakakibara     

A survey of normal form covers for context free grammars

Summary An overview is given of cover results for normal forms of context-free grammars. The emphasis in this paper is on the possibility of constructing ɛ-free grammars, non-left-recursive grammars and grammars in Greibach normal form. Among others it is proved that any ɛ-free context-free grammar can be right covered with a context-free grammar in Greibach normal form. All the cover results concerning the ɛ-free grammars, the non-left-recursive grammars and the grammars in Greibach normal form are listed, with respect to several types of covers, in a cover-table.     

Incremental learning of context free grammars based on bottom-up parsing and search

This paper describes approaches for machine learning of context free grammars (CFGs) from positive and negative sample strings, which are implemented in Synapse system. The grammatical inference consists of a rule generation by “inductive CYK algorithm,” mechanisms for incremental learning, and search. Inductive CYK algorithm generates minimum production rules required for parsing positive samples, when the bottom-up parsing by CYK algorithm does not succeed. The incremental learning is used not only for synthesizing grammars by giving the system positive strings in the order of their length but also for learning grammars from other similar grammars. Synapse can synthesize fundamental ambiguous and unambiguous CFGs including nontrivial grammars such as the set of strings not of the form ww with w∈{a,b}⁺.     

Left context precedence grammars

Summary Left context precendence grammars which are defined in this paper are a proper subclass of the precedence grammars and contain properly the simple precedence grammars. Left context precedence grammars need not be uniquely invertible. The parsing algorithm developed for left context precedence languages works with linear time and has the viable prefix property which is a stronger property than the correct prefix property.     

One-sided random context grammars

The notion of a one-sided random context grammar is defined as a context-free-based regulated grammar, in which a set of permitting symbols and a set of forbidding symbols are attached to every rule, and its set of rules is divided into the set of left random context rules and the set of right random context rules. A left random context rule can rewrite a nonterminal if each of its permitting symbols occurs to the left of the rewritten symbol in the current sentential form while each of its forbidding symbols does not occur there. A right random context rule is applied analogically except that the symbols are examined to the right of the rewritten symbol. The paper demonstrates that without erasing rules, one-sided random context grammars characterize the family of context-sensitive languages, and with erasing rules, these grammars characterize the family of recursively enumerable languages. In fact, these characterization results hold even if the set of left random context rules coincides with the set of right random context rules. Several special cases of these grammars are considered, and their generative power is established. In its conclusion, some important open problems are suggested to study in the future.     

Expressing unrestricted grammars by extended definite clause grammars1

A definition of extended definite clause grammars and their relationship to unrestricted grammars are presented. A method for translating extended definite clause grammars describing unrestricted grammars into executable prolog programs is given. Three different parsing techniques are presented, and for each a complete presentation of how to incorporate unrestricted grammars in the actual formalism is done. Extended definite clause grammar is a powerful formalism usable for specifying grammars in natural language processing systems.     

Syntactic complexity of scattered context grammars

The syntactic complexity of scattered context grammars with respect to the number of nonterminals is investigated. First, the family of the recursively enumerable languages is characterized by some basic operations, such as quotient and coding, over the languages generated by propagating scattered context grammars with four nonterminals. Then, a new method of achieving the characterization of the family of recursively enumerable languages by scattered context grammars is given; in fact, this family is characterized by scattered context grammars with only five nonterminals and a single erasing production. Finally, it is demonstrated that the number of nonterminals can be decreased by one in the present characterizations if scattered context grammars start their derivations from a word rather than a single symbol.     

Nonterminal complexity of one-sided random context grammars

In the present paper, we study the nonterminal complexity of one-sided random context grammars. More specifically, we prove that every recursively enumerable language can be generated by a one-sided random context grammar with no more than ten nonterminals. An analogical result holds for thirteen nonterminals in terms of these grammars with the set of left random context rules coinciding with the set of right random context rules. Furthermore, we introduce the notion of a right random context nonterminal, defined as a nonterminal that appears on the left-hand side of a right random context rule. We demonstrate how to convert any one-sided random context grammar G to an equivalent one-sided random context grammar H with two right random context nonterminals. An analogical conversion is given in terms of (1) propagating one-sided random context grammars and (2) left random context nonterminals. In the conclusion, two open problems are stated.     

Model-based test generation using extended symbolic grammars

A novel, model-based test case generation approach for validating reactive systems, especially those supporting richly structured data inputs and/or interactions, is presented. Given an executable system model and an extended symbolic grammar specifying plausible system inputs, the approach performs a model-based simulation to (i) ensure the consistency of the model with respect to the specified inputs, and (ii) generate corresponding test cases for validating the system. The model-based simulation produces a state transition diagram (STD) automatically justifying the model runtime behaviors within the test case coverage. The STD can further be transformed to produce an evolved symbolic grammar, which can then be used to incrementally generate a refined set of test cases. As a case study, we present a live sequence chart (LSC) model-based test generator, named LCT in short, for LSC simulation and consistency testing. The evolved symbolic grammar produced by the simulator can either be used to generate practical test cases for software testing, or be further refined by applying our model-based test generation approach again with additional test coverage criteria. We further show that LSCs can also be used to specify and test certain temporal system properties during the model simulation. Their satisfaction, reflected in the STD, can either be served as a directive for selective test generation, or a basis for further temporal property model checking.     

Parsing extended LR(k) grammars

Summary An extended LR(k) (ELR(k)) grammar is a context free grammar in which the right sides of the productions are regular expressions and which can be parsed from left to right with k symbol look-ahead. We present a practical algorithm for producing small fast parsers directly from certain ELR(k) grammars, and an algorithm for converting the remaining ELR(k) grammars into a form that can be processed by the first algorithm. This method, when combined with previously developed methods for improving the efficiency of LR(k) parsers, usually produces parsers that are significantly smaller and faster than those produced by previous LR(k) and ELR(k) algorithms.     

On the descriptional complexity of scattered context grammars

This paper proves that every recursively enumerable language is generated by a scattered context grammar with no more than four nonterminals and three non-context-free productions. In its conclusion, it gives an overview of the results and open problems concerning scattered context grammars and languages.     

An LR parsing technique for extended context-free grammars

Extended context-free grammars allow regular expressions to appear in productions right hand sides, and are a clear and natural way to describe the syntax of programming languages.In this paper an LR parsing technique for extended context-free grammars is presented, which is based on an underlying transformation of the grammar into an equivalent context-free one.The technique is suitable for inclusion in one-pass compilers: the implementation requires little extensions to the algorithms working for normal LR grammars. Besides describing the parsing method, the paper shows also the algorithms for deriving the parsing tables; tables optimization is also discussed. Finally, this technique is compared with other proposals appeared in the literature.     

A note on a proposed LALR parser for extended context-free grammars

The lemma in “An improved LALR(k) parser generation for regular right part grammars” [Inform. Process. Lett. 47 (1993) 123-129] to test the applicability of the method is shown to be false by means of a counter-example grammar.