Incremental Evolution of Fuzzy Grammar Fragments to Enhance Instance Matching and Text Mining

Abstract

In many applications, it is useful to extract structured data from sections of unstructured text. A common approach is to use pattern matching (e.g., regular expressions) or more general grammar-based techniques. In cases where exact templates or grammar fragments are not known, it is possible to use machine learning approaches, based on words or n-grams, to identify the structured data. This is generally a two-stage (train/use) process that cannot easily cope with incremental extensions of the training set. In this paper, we combine a fuzzy grammar-based approach with incremental learning. This enables a set of grammar fragments to evolve incrementally, each time a new example is given, while guaranteeing that it can parse previously seen examples. We propose a novel measure of overlap between fuzzy grammar fragments that can also be used to determine the degree to which a string is parsed by a grammar fragment. This measure of overlap allows us to compare the range of two fuzzy grammar fragments (i.e., to estimate and compare the sets of strings that fuzzily conform to each grammar) without explicitly parsing any strings. A simple application shows the method's validity.

Bisociative Discovery in Business Process Models

“…It is difficult to anticipate all possible forms (including abbreviations) in a regular expression. Full details of the fuzzy grammar approach are given in [7,8] …”

“…Because there was so much flexibility in the question/answer patterns, we preprocessed the text to extract key features, using fuzzy grammar tagging [7] to add attributes. This went beyond a simple keyword-recognition approach (which was found to be inadequate) and was able to cope with the non-standard language employed in the questions.…”

Prediction of the MSCI EURO index based on fuzzy grammar fragments extracted from European Central Bank statements

“…Central to our work is the approach described in [1], where the evolution of fuzzy grammar fragments is studied for matching strings originating in free text. Also central to our approach are the methods described in [2], [3] for learning and extracting such fuzzy grammar fragments from text.…”

“…The focus of these statements is mainly on 1 Available at: http://www.ecb.int/press/html/index.en.html announcing the levels of the key interest rates set by the ECB, and providing the arguments that support this decision.…”

“…While our previous work [4] has focussed on a selected number of General Inquirer (GI) categories [5] and the frequency of words across these categories, the current work focuses on a more expressive approach. Building upon recent work relating to fuzzy grammars [1], [3], we focus on extracting such grammars from the text of the ECB statements, grouping these grammars based on the GI content category they represent, and identifying the frequency with which these fuzzy grammar categories are encountered in the texts of the individual ECB statements.…”