The problem of inducing, learning or inferring grammars has been studied for decades, but only in recent years has grammatical inference emerged as an independent field with connections to many scientific disciplines, including bio-informatics, computational linguistics and pattern recognition. This book meets the need for a comprehensive and unified summary of the basic techniques and results, suitable for researchers working in these various areas. In Part I, the objects of use for grammatical inference are studied in detail: strings and their topology, automata and grammars, whether probabilistic or not. Part II carefully explores the main questions in the field: What does learning mean? How can we associate complexity theory with learning? In Part III the author describes a number of techniques and algorithms that allow us to learn from text, from an informant, or through interaction with the environment. These concern automata, grammars, rewriting systems, pattern languages or transducers.

Grammatical Inference: Learning Automata and Grammars

A bibliographical study of grammatical inference

The natural language learning problem has attracted the attention of researchers for several decades Computational and formal models of language acquisition have provided some preliminary yet promising insights of how children learn the lan guage of their community Further these formal models also provide an operational framework for the numerous practical applications of language learning We will survey some of the key results in formal language learning In particular we will discuss the prominent computational approaches for learning di erent classes of for mal languages and discuss how these t in the broad context of natural language learning

Grammar Inference Automata Induction and Language Acquisition

Whereas there is a number of methods and algorithms to learn regular languages, moving up the Chomsky hierarchy is proving to be a challenging task. Indeed, several theoretical barriers make the class of context-free languages hard to learn. To tackle these barriers, we choose to change the way we represent these languages. Among the formalisms that allow the definition of classes of languages, the one of string-rewriting systems (SRS) has outstanding properties. We introduce a new type of SRS's, called Delimited SRS (DSRS), that are expressive enough to define, in a uniform way, a noteworthy and non trivial class of languages that contains all the regular languages, $$\{a^{n}b^{n}: n \geq 0 \}$$ , $$\{w\in \{a,b\}^{*}:|w|_{a}=|w|_{b}\}$$ , the parenthesis languages of Dyck, the language of Lukasiewicz, and many others. Moreover, DSRS's constitute an efficient (often linear) parsing device for strings, and are thus promising candidates in forthcoming applications of grammatical inference. In this paper, we pioneer the problem of their learnability. We propose a novel and sound algorithm (called LARS) which identifies a large subclass of them in polynomial time (but not data). We illustrate the execution of our algorithm through several examples, discuss the position of the class in the Chomsky hierarchy and finally raise some open questions and research directions.

/pdf/lars-a-learning-algorithm-for-rewriting-systems-fr6u6p5x67.pdf

LARS: A learning algorithm for rewriting systems

: Round trip engineering of software from source code and reverse engineering of software from binary files have both been extensively studied and the state-of-practice have documented tools and techniques. Forward engineering of protocols has also been extensively studied and there are firmly established techniques for generating correct protocols. While observation of protocol behavior for performance testing has been studied and techniques established, reverse engineering of protocol control flow from observations of protocol behavior has not received the same level of attention. State-of-practice in reverse engineering the control flow of computer network protocols is comprised of mostly ad hoc approaches. We examine state-of-practice tools and techniques used in three open source projects: Pidgin, Samba, and rdesktop. We examine techniques proposed by computational learning researchers for grammatical inference. We propose to extend the state-of-art by inferring protocol control flow using grammatical inference inspired techniques to reverse engineer automata representations from captured data flows. We present evidence that grammatical inference is applicable to the problem domain under consideration.

/pdf/dynamic-protocol-reverse-engineering-a-grammatical-inference-3lgusdfpe8.pdf

Dynamic Protocol Reverse Engineering: A Grammatical Inference Approach

The problem of inferring grammars from examples and counter-examples has been mostly studied for regular languages. Here we apply a classical approach, said by enumeration, to context-free grammars. Structural containment is used as an ordering relation, and associated operators are defined. Special attention is paid to computation time, hence our system cannot always reach all the solutions for a given problem. Though, our system is able to generalize efficiently, and the solutions found constitute a good description of the sample.

Inference of Context-Free Grammars by Enumeration: Structural Containment as an Ordering Bias

Grammatical inference is the problem of learning a language from examples and counter-examples of words in this language. The common formulations of this problem make the fundamental hypothesis of the existence of a formal language underlying the data, to be discovered. Here we follow another approach, based on the following remarks. First of all, the initial hypothesis of a formal language to be identified does not hold for real data. Secondly, the algorithmic complexity of language identification is huge, even in the case of the simplest class of languages. Our approach aims at removing these two limitations. It allows the grammars produced to discriminate the sample words imperfectly, while it introduces the use of classic optimization techniques. Here we apply Tabu Search to the inference of regular grammars.

Jean-Yves Giordano

Papers

Inference of Context-Free Grammars by Enumeration: Structural Containment as an Ordering Bias

Grammatical inference using tabu search