Automatic extraction of concepts from texts and applications

The extraction of relevant terms from texts is an extensively researched task in Text- Mining. Relevant terms have been applied in areas such as Information Retrieval or document clustering and classification. However, relevance has a rather fuzzy nature since the classification of some terms as relevant or not relevant is not consensual. For instance, while words such as "president" and "republic" are generally considered relevant by human evaluators, and words like "the" and "or" are not, terms such as "read" and "finish" gather no consensus about their semantic and informativeness. Concepts, on the other hand, have a less fuzzy nature. Therefore, instead of deciding on the relevance of a term during the extraction phase, as most extractors do, I propose to first extract, from texts, what I have called generic concepts (all concepts) and postpone the decision about relevance for downstream applications, accordingly to their needs. For instance, a keyword extractor may assume that the most relevant keywords are the most frequent concepts on the documents. Moreover, most statistical extractors are incapable of extracting single-word and multi-word expressions using the same methodology. These factors led to the development of the ConceptExtractor, a statistical and language-independent methodology which is explained in Part I of this thesis. In Part II, I will show that the automatic extraction of concepts has great applicability. For instance, for the extraction of keywords from documents, using the Tf-Idf metric only on concepts yields better results than using Tf-Idf without concepts, specially for multi-words. In addition, since concepts can be semantically related to other concepts, this allows us to build implicit document descriptors. These applications led to published work. Finally, I will present some work that, although not published yet, is briefly discussed in this document.

Using qualia information to identify lexical semantic classes in an unsupervised clustering task

Acquiring lexical information is a complex problem, typically approached by relying on a number of contexts to contribute information for classification. One of the first issues to address in this domain is the determination of such contexts. The work presented here proposes the use of automatically obtained FORMAL role descriptors as features used to draw nouns from the same lexical semantic class together in an unsupervised clustering task. We have dealt with three lexical semantic classes (HUMAN, LOCATION and EVENT) in English. The results obtained show that it is possible to discriminate between elements from different lexical semantic classes using only FORMAL role information, hence validating our initial hypothesis. Also, iterating our method accurately accounts for fine-grained distinctions within lexical classes, namely distinctions involving ambiguous expressions. Moreover, a filtering and bootstrapping strategy employed in extracting FORMAL role descriptors proved to minimize effects of sparse data and noise in our task.

Automatic induction of lexical features

This thesis concerns artificially intelligent natural language processing systems that are capable of learning the properties of lexical items (properties like verbal valency or inflectional class membership) autonomously while they are fulfilling their tasks for which they have been deployed in the first place. Many of these tasks require a deep analysis of language input, which can be characterized as a mapping of utterances in a given input C to a set S of linguistically motivated structures with the help of linguistic information encoded in a grammar G and a lexicon L: G + L + C → S (1) The idea that underlies intelligent lexical acquisition systems is to modify this schematic formula in such a way that the system is able to exploit the information encoded in S to create a new, improved version of the lexicon: G + L + S → L' (2) Moreover, the thesis claims that a system can only be considered intelligent if it does not just make maximum usage of the learning opportunities in C, but if it is also able to revise falsely acquired lexical knowledge. So, one of the central elements in this work is the formulation of a couple of criteria for intelligent lexical acquisition systems subsumed under one paradigm: the Learn-Alpha design rule. The thesis describes the design and quality of a prototype for such a system, whose acquisition components have been developed from scratch and built on top of one of the state-of-the-art Head-driven Phrase Structure Grammar (HPSG) processing systems. The quality of this prototype is investigated in a series of experiments, in which the system is fed with extracts of a large English corpus. While the idea of using machine-readable language input to automatically acquire lexical knowledge is not new, we are not aware of a system that fulfills Learn-Alpha and is able to deal with large corpora. To instance four major challenges of constructing such a system, it should be mentioned that a) the high number of possible structural descriptions caused by highly underspeci ed lexical entries demands for a parser with a very effective ambiguity management system, b) the automatic construction of concise lexical entries out of a bulk of observed lexical facts requires a special technique of data alignment, c) the reliability of these entries depends on the system's decision on whether it has seen 'enough' input and d) general properties of language might render some lexical features indeterminable if the system tries to acquire them with a too high precision. The cornerstone of this dissertation is the motivation and development of a general theory of automatic lexical acquisition that is applicable to every language and independent of any particular theory of grammar or lexicon. This work is divided into five chapters. The introductory chapter first contrasts three different and mutually incompatible approaches to (artificial) lexical acquisition: cue-based queries, head-lexicalized probabilistic context free grammars and learning by unification. Then the postulation of the Learn-Alpha design rule is presented. The second chapter outlines the theory that underlies Learn-Alpha and exposes all the related notions and concepts required for a proper understanding of artificial lexical acquisition. Chapter 3 develops the prototyped acquisition method, called ANALYZE-LEARN-REDUCE, a framework which implements Learn-Alpha. The fourth chapter presents the design and results of a bootstrapping experiment conducted on this prototype: lexeme detection, learning of verbal valency, categorization into nominal count/mass classes, selection of prepositions and sentential complements, among others. The thesis concludes with a review of the conclusions and motivation for further improvements as well as proposals for future research on the automatic induction of lexical features.

Automatic Creation of Lexical Resources for an Interlingua-based System

The Universal Networking Language (UNL) is an interlingua designed to be the base of several natural language processing systems aiming to support multilinguality in internet. One of the main components of the language is the dictionary of Universal Words (UWs), which links the vocabularies of the different languages involved in the project. As any NLP system, coverage and accuracy in its lexical resources are crucial for the development of the system. In this paper, the authors describes how a large coverage UWs dictionary was automatically created, based on an existent and well known resource like the English WordNet. Other aspects like implementation details and the evaluation of the final UW set are also depicted.

Towards the Fully Automatic Merging of Lexical Resources: A Step Forward

This article reports on the results of the research done towards the fully automatically merging of lexical resources. Our main goal is to show the generality of the proposed approach, which have been previously applied to merge Spanish Subcategorization Frames lexica. In this work we extend and apply the same technique to perform the merging of morphosyntactic lexica encoded in LMF. The experiments showed that the technique is general enough to obtain good results in these two different tasks which is an important step towards performing the merging of lexical resources fully automatically.

Automatic Extraction of Polar Adjectives for the Creation of Polarity Lexicons

Automatic creation of polarity lexicons is a crucial issue to be solved in order to reduce time andefforts in the first steps of Sentiment Analysis. In this paper we present a methodology based onlinguistic cues that allows us to automatically discover, extract and label subjective adjectivesthat should be collected in a domain-based polarity lexicon. For this purpose, we designed abootstrapping algorithm that, from a small set of seed polar adjectives, is capable to iterativelyidentify, extract and annotate positive and negative adjectives. Additionally, the methodautomatically creates lists of highly subjective elements that change their prior polarity evenwithin the same domain. The algorithm proposed reached a precision of 97.5% for positiveadjectives and 71.4% for negative ones in the semantic orientation identification task.

Towards the Automatic Merging of Lexical Resources: Automatic Mapping

Lexical Resources are a critical component for Natural Language Processing applications. However, the high cost of comparing and merging different resources has been a bottleneck to have richer resources with a broad range of potential uses for a significant number of languages.With the objective of reducing cost byeliminating human intervention, we present a new method for automating the merging of resources,with special emphasis in what we call the mapping step. This mapping step, which converts the resources into a common format that allows latter the merging, is usually performed with huge manual effort and thus makes the whole process very costly. Thus, we propose a method to perform this mapping fully automatically. To test our method, we have addressed the merging of two verb subcategorization frame lexica for Spanish, The resultsachieved, that almost replicate human work, demonstrate the feasibility of the approach.

A Method Towards the Fully Automatic Merging of Lexical Resources

Lexical Resources are a critical component for Natural Language Processing applications. However, the high cost of comparing and merging different resources has been a bottleneck to obtain richer resources and a broader range of potential uses for a significant number of languages. With the objective of reducing cost by eliminating human intervention, we present a new method towards the automatic merging of resources. This method includes both, the automatic mapping of resources involved to a common format and merging them, once in this format. This paper presents how we have addressed the merging of two verb subcategorization frame lexica for Spanish, but our method will be extended to cover other types of Lexical Resources. The achieved results, that almost replicate human work, demonstrate the feasibility of the approach.

Automatic extraction of road seeds from high-resolution aerial images

This article presents an automatic methodology for extraction of road seeds from high-resolution aerial images. The method is based on a set of four road objects and another set of connection rules among road objects. Each road object is a local representation of an approximately straight road fragment and its construction is based on a combination of polygons describing all relevant image edges, according to some rules embodying road knowledge. Each one of the road seeds is composed by a sequence of connected road objects, in which each sequence of this type can be geometrically structured as a chain of contiguous quadrilaterals. Experiments carried out with high-resolution aerial images showed that the proposed methodology is very promising in extracting road seeds. This article presents the fundamentals of the method and the experimental results, as well.

Automatic extraction of the mid-facial plane for cranio-maxillofacial surgery planning

Recently developed computer applications provide tools for planning cranio-maxillofacial interventions based on 3-dimensional (3D) virtual models of the patient's skull obtained from computed-tomography (CT) scans. Precise knowledge of the location of the mid-facial plane is important for the assessment of deformities and for planning reconstructive procedures. In this work, a new method is presented to automatically compute the mid-facial plane on the basis of a surface model of the facial skeleton obtained from CT. The method matches homologous surface areas selected by the user on the left and right facial side using an iterative closest point optimization. The symmetry plane which best approximates this matching transformation is then computed. This new automatic method was evaluated in an experimental study. The study included experienced and inexperienced clinicians defining the symmetry plane by a selection of landmarks. This manual definition was systematically compared with the definition resulting from the new automatic method: Quality of the symmetry planes was evaluated by their ability to match homologous areas of the face. Results show that the new automatic method is reliable and leads to significantly higher accuracy than the manual method when performed by inexperienced clinicians. In addition, the method performs equally well in difficult trauma situations, where key landmarks are unreliable or absent.

Automatic extraction of femur contours from calibrated fluoroscopic images

Automatic identification and extraction of bone contours from X-ray images is an essential first step task for further medical image analysis. In this paper we propose a 3D statistical model based framework for the proximal femur contour extraction from calibrated X-ray images. The automatic initialization is solved by an estimation of Bayesian network algorithm to fit a multiple component geometrical model to the X-ray data. The contour extraction is accomplished by a non-rigid 2D/3D registration between a 3D statistical model and the X-ray images, in which bone contours are extracted by a graphical model based Bayesian inference. Preliminary experiments on clinical data sets verified its validity