Evolutionary Learning of Syntax Patterns for Genic Interaction Extraction

Abstract

There is an increasing interest in the development of techniques for automatic relation extraction from unstructured text. The biomedical domain, in particular, is a sector that may greatly benefit from those techniques due to the huge and ever increasing amount of scientific publications describing observed phenomena of potential clinical interest.\ud In this paper, we consider the problem of automatically identifying sentences that contain interactions between genes and proteins, based solely on a dictionary of genes and proteins and a small set of sample sentences in natural language. We propose an evolutionary technique for learning a classifier that is capable of detecting the desired sentences within scientific publications with high accuracy. The key feature of our proposal, that is internally based on Genetic Programming, is the construction of a model of the relevant syntax patterns in terms of standard part-of-speech annotations. The model consists of a set of regular expressions that are learned automatically despite the large alphabet size involved.\ud We assess our approach on two realistic datasets and obtain 77% accuracy, a value sufficiently high to be of practical interest and that is in line with significant baseline methods

ARTE: Automated Generation of Realistic Test Inputs for Web APIs

“…Finally, ARTE can optionally generate regular expressions for each target parameter based on the input values obtained in the previous phases, by integrating the tool RegexGener-ator++ [44], [45], [46] (details in Section 4). This allows to refine the search for syntactically valid inputs, improving the effectiveness of the approach.…”

“…We implemented ARTE in Java, leveraging existing libraries 532 for specific tasks, namely: (1) Jena [47], for the creation 533 of SPARQL queries; (2) Stanford CoreNLP [48], for NLP 534 related tasks; and (3) RegexGenerator++ [44], [45], [46], for 535 the generation of regular expressions. RegexGenerator++ 536 uses search-based techniques for automatically generating 537 context-aware regular expressions based on strings tagged 538 as valid or invalid (i.e., matching or not matching the regular 539 expression, respectively) within a corpus, namely, a text 540 that provides some context.…”

Automated generation of realistic test inputs for web APIs

“…The list of input values may contain entries in different formats (e.g., websites starting with "https://" and "www"), with only one of them being accepted by the API. ARTE can optionally learn from previous API responses by automatically generating a regular expression that matches only the valid values (e.g., those URLs starting with "https://"), this regular expression (generated by leveraging a modified version of RegexGenerator++ [8][9][10]) is used to filter the list of input values. Optionally, the regular expression can be used to conduct a refined search for input values.…”

“…A prototype of ARTE has been implemented in Java leveraging the libraries Jena [5], for the creation of SPARQL queries, Stanford CoreNLP [6], for NLP related tasks, and RegexGenerator++ [8][9][10], for the generation of regular expressions.…”

Towards an evolutionary-based approach for natural language processing

“…Another task where GP was employed is the search and extraction of semantic relationships in texts, which is especially relevant in the medical domain. In this domain, GP was used to identify sentences that contain descriptions of interactions between genes and proteins [9]. More in detail, GP was employed to obtain a model of syntax patterns composed of part-of-speech tags.…”