Capturing Emergent Semantics from Social Annotation Systems

The ongoing growth of the World Wide Web, catalyzed by the increasing possibility of ubiquitous access via a variety of devices, continues to strengthen its role as our prevalent information and commmunication medium. However, although tools like search engines facilitate retrieval, the task of finally making sense of Web content is still often left to human interpretation. The vision of supporting both humans and machines in such knowledge-based activities led to the development of different systems which allow to structure Web resources by metadata annotations. Interestingly, two major approaches which gained a considerable amount of attention are addressing the problem from nearly opposite directions: On the one hand, the idea of the Semantic Web suggests to formalize the knowledge within a particular domain by means of the "top-down" approach of defining ontologies. On the other hand, Social Annotation Systems as part of the so-called Web 2.0 movement implement a "bottom-up" style of categorization using arbitrary keywords. Experience as well as research in the characteristics of both systems has shown that their strengths and weaknesses seem to be inverse: While Social Annotation suffers from problems like, e. g., ambiguity or lack or precision, ontologies were especially designed to eliminate those. On the contrary, the latter suffer from a knowledge acquisition bottleneck, which is successfully overcome by the large user populations of Social Annotation Systems. Instead of being regarded as competing paradigms, the obvious potential synergies from a combination of both motivated approaches to "bridge the gap" between them. These were fostered by the evidence of emergent semantics, i. e., the self-organized evolution of implicit conceptual structures, within Social Annotation data. While several techniques to exploit the emergent patterns were proposed, a systematic analysis - especially regarding paradigms from the field of ontology learning - is still largely missing. This also includes a deeper understanding of the circumstances which affect the evolution processes. This work aims to address this gap by providing an in-depth study of methods and influencing factors to capture emergent semantics from Social Annotation Systems. We focus hereby on the acquisition of lexical semantics from the underlying networks of keywords, users and resources. Structured along different ontology learning tasks, we use a methodology of semantic grounding to characterize and evaluate the semantic relations captured by different methods. In all cases, our studies are based on datasets from several Social Annotation Systems. Specifically, we first analyze semantic relatedness among keywords, and identify measures which detect different notions of relatedness. These constitute the input of concept learning algorithms, which focus then on the discovery of synonymous and ambiguous keywords. Hereby, we assess the usefulness of various clustering techniques. As a prerequisite to induce hierarchical relationships, our next step is to study measures which quantify the level of generality of a particular keyword. We find that comparatively simple measures can approximate the generality information encoded in reference taxonomies. These insights are used to inform the final task, namely the creation of concept hierarchies. For this purpose, generality-based algorithms exhibit advantages compared to clustering approaches. In order to complement the identification of suitable methods to capture semantic structures, we analyze as a next step several factors which influence their emergence. Empirical evidence is provided that the amount of available data plays a crucial role for determining keyword meanings. From a different perspective, we examine pragmatic aspects by considering different annotation patterns among users. Based on a broad distinction between "categorizers" and "describers", we find that the latter produce more accurate results. This suggests a causal link between pragmatic and semantic aspects of keyword annotation. As a special kind of usage pattern, we then have a look at system abuse and spam. While observing a mixed picture, we suggest that an individual decision should be taken instead of disregarding spammers as a matter of principle. Finally, we discuss a set of applications which operationalize the results of our studies for enhancing both Social Annotation and semantic systems. These comprise on the one hand tools which foster the emergence of semantics, and on the one hand applications which exploit the socially induced relations to improve, e. g., searching, browsing, or user profiling facilities. In summary, the contributions of this work highlight viable methods and crucial aspects for designing enhanced knowledge-based services of a Social Semantic Web.

Estruturação do conhecimento e relações semânticas: uma ontologia para o domínio da naonociência e nanotecnologia

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Communities and emerging semantics in semantic link network:discovery and learning

The World Wide Web provides plentiful contents for Web-based learning, but its hyperlink-based architecture connects Web resources for browsing freely rather than for effective learning. To support effective learning, an e-learning system should be able to discover and make use of the semantic communities and the emerging semantic relations in a dynamic complex network of learning resources. Previous graph-based community discovery approaches are limited in ability to discover semantic communities. This paper first suggests the Semantic Link Network (SLN), a loosely coupled semantic data model that can semantically link resources and derive out implicit semantic links according to a set of relational reasoning rules. By studying the intrinsic relationship between semantic communities and the semantic space of SLN, approaches to discovering reasoning-constraint, rule-constraint, and classification-constraint semantic communities are proposed. Further, the approaches, principles, and strategies for discovering emerging semantics in dynamic SLNs are studied. The basic laws of the semantic link network motion are revealed for the first time. An e-learning environment incorporating the proposed approaches, principles, and strategies to support effective discovery and learning is suggested.

Semantic patterns for sentiment analysis of Twitter

Most existing approaches to Twitter sentiment analysis assume that sentiment is explicitly expressed through affective words. Nevertheless, sentiment is often implicitly expressed via latent semantic relations, patterns and dependencies among words in tweets. In this paper, we propose a novel approach that automatically captures patterns of words of similar contextual semantics and sentiment in tweets. Unlike previous work on sentiment pattern extraction, our proposed approach does not rely on external and fixed sets of syntactical templates/patterns, nor requires deep analyses of the syntactic structure of sentences in tweets. We evaluate our approach with tweet- and entity-level sentiment analysis tasks by using the extracted semantic patterns as classification features in both tasks. We use 9 Twitter datasets in our evaluation and compare the performance of our patterns against 6 state-of-the-art baselines. Results show that our patterns consistently outperform all other baselines on all datasets by 2.19% at the tweet-level and 7.5% at the entity-level in average F-measure.

Automatic identification of whole-part relations in Portuguese

Dissertação de Mestrado, Ciências da Linguagem, Faculdade de Ciências Humanas e Sociais, Universidade do Algarve, 2014

Health-related quality of life in Australian children with asthma: lessons for the cross-cultural use of quality of life instruments

The development of Australian forms of the Childhood Asthma Questionnaires (CAQs) is reported. Focus group methods and psychometric analyses were used to establish the conceptual, semantic and technical equivalence of these forms with the UK versions. Both versions also provide for data collection from non-asthmatic youngsters. The internal consistency was found to be acceptable (Cronbach's alpha 0.52-0.90) and the health-related quality of life (HRQoL) scores were found to vary with asthma severity (p < 0.05). Comparison with the UK data revealed that the non-asthmatic scores were higher for Australian than British children (p < 0.001) but that the scores for children with asthma did not differ between the two countries. It was only In the Australian sample that the group with asthma reported impaired HRQoL when compared to their healthy peers. These findings were interpreted in the context of cultural expectations of life quality and conclusions are presented regarding the importance of the gap between experience and expectations. The difficulties raised by the developmental and cultural issues inherent in paediatric HRQoL research were discussed. Qual. Life Res. 7:409-419 (C) 1998 Kluwer Academic Publishers

Validation of psychosocial scales for physical activity in university students

OBJECTIVE Translate the Patient-centered Assessment and Counseling for Exercise questionnaire, adapt it cross-culturally and identify the psychometric properties of the psychosocial scales for physical activity in young university students.METHODS The Patient-centered Assessment and Counseling for Exercise questionnaire is made up of 39 items divided into constructs based on the social cognitive theory and the transtheoretical model. The analyzed constructs were, as follows: behavior change strategy (15 items), decision-making process (10), self-efficacy (6), support from family (4), and support from friends (4). The validation procedures were conceptual, semantic, operational, and functional equivalences, in addition to the equivalence of the items and of measurements. The conceptual, of items and semantic equivalences were performed by a specialized committee. During measurement equivalence, the instrument was applied to 717 university students. Exploratory factor analysis was used to verify the loading of each item, explained variance and internal consistency of the constructs. Reproducibility was measured by means of intraclass correlation coefficient.RESULTS The two translations were equivalent and back-translation was similar to the original version, with few adaptations. The layout, presentation order of the constructs and items from the original version were kept in the same form as the original instrument. The sample size was adequate and was evaluated by the Kaiser-Meyer-Olkin test, with values between 0.72 and 0.91. The correlation matrix of the items presented r < 0.8 (p < 0.05). The factor loadings of the items from all the constructs were satisfactory (> 0.40), varying between 0.43 and 0.80, which explained between 45.4% and 59.0% of the variance. Internal consistency was satisfactory (α ≥ 0.70), with support from friends being 0.70 and 0.92 for self-efficacy. Most items (74.3%) presented values above 0.70 for the reproducibility test.CONCLUSIONS The validation process steps were considered satisfactory and adequate for applying to the population.

Improving search engines with open Web-based SKOS vocabularies

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

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.

Anatomical correlates for category-specific naming of objects and actions: A brain stimulation mapping study.

The production of object and action words can be dissociated in aphasics, yet their anatomical correlates have been difficult to distinguish in functional imaging studies. To investigate the extent to which the cortical neural networks underlying object- and action-naming processing overlap, we performed electrostimulation mapping (ESM), which is a neurosurgical mapping technique routinely used to examine language function during brain-tumor resections. Forty-one right-handed patients who had surgery for a brain tumor were asked to perform overt naming of object and action pictures under stimulation. Overall, 73 out of the 633 stimulated cortical sites (11.5%) were associated with stimulation-induced language interferences. These interference sites were very much localized (<1 cm(2) ), and showed substantial variability across individuals in their exact localization. Stimulation interfered with both object and action naming over 44 sites, whereas it specifically interfered with object naming over 19 sites and with action naming over 10 sites. Specific object-naming sites were mainly identified in Broca's area (Brodmann area 44/45) and the temporal cortex, whereas action-naming specific sites were mainly identified in the posterior midfrontal gyrus (Brodmann area 6/9) and Broca's area (P = 0.003 by the Fisher's exact test). The anatomical loci we emphasized are in line with a cortical distinction between objects and actions based on conceptual/semantic features, so the prefrontal/premotor cortex would preferentially support sensorimotor contingencies associated with actions, whereas the temporal cortex would preferentially underpin (functional) properties of objects. Hum Brain Mapp 35:429-443, 2014. © 2012 Wiley Periodicals, Inc.

Social tagging within task management

Social tagging evolved in response to a need to tag heterogeneous objects, the automated tagging of which is usually not feasible by current technological means. Social tagging can be used for more flexible competence management within organizations. The profiles of employees can be built in the form of groups of tags, as employees tag each other, based on their familiarity of each other’s expertise. This can serve as a replacement for the more traditional competence management approaches, which usually become outdated due to social and organizational hurdles, and obsolete data. These limitations can be overcome by people tagging, as the information revealed by such tags is usually based on most recent employee interaction and knowledge. Task management as part of personal information management aims at the support of users’ individual task handling. This can include collaborating with other individuals, sharing one’s knowledge, both functional and process-related, and distributing documents and web resources. In this context, Task patterns can be used as templates that collect information and experience around tasks associated to it during run time, facilitating agility. The effective collaboration among contributors necessitates the means to find the appropriate individuals to work with on the task, and this can be made possible by using social tagging to describe individual competencies. The goal of this study is to support finding and tagging people within task management, through the effective exploitation of the work/task context. This involves the utilization of knowledge of the workers’ expertise, nature of the task/task pattern and information available from the documents and web resources attached to the task. Vice versa, task management provides an excellent environment for social tagging due to the task context that already provides suitable tags. The study also aims at assisting users of the task management solution with the collaborative construction of light-weight ontology by inferring semantic relations between tags. The thesis project aims at an implementation of people finding & tagging within the java application for task management that consumes web services, which provide the required ontology for the organization.

Biomedical Event Extraction with Machine Learning

Biomedical natural language processing (BioNLP) is a subfield of natural language processing, an area of computational linguistics concerned with developing programs that work with natural language: written texts and speech. Biomedical relation extraction concerns the detection of semantic relations such as protein-protein interactions (PPI) from scientific texts. The aim is to enhance information retrieval by detecting relations between concepts, not just individual concepts as with a keyword search. In recent years, events have been proposed as a more detailed alternative for simple pairwise PPI relations. Events provide a systematic, structural representation for annotating the content of natural language texts. Events are characterized by annotated trigger words, directed and typed arguments and the ability to nest other events. For example, the sentence “Protein A causes protein B to bind protein C” can be annotated with the nested event structure CAUSE(A, BIND(B, C)). Converted to such formal representations, the information of natural language texts can be used by computational applications. Biomedical event annotations were introduced by the BioInfer and GENIA corpora, and event extraction was popularized by the BioNLP'09 Shared Task on Event Extraction. In this thesis we present a method for automated event extraction, implemented as the Turku Event Extraction System (TEES). A unified graph format is defined for representing event annotations and the problem of extracting complex event structures is decomposed into a number of independent classification tasks. These classification tasks are solved using SVM and RLS classifiers, utilizing rich feature representations built from full dependency parsing. Building on earlier work on pairwise relation extraction and using a generalized graph representation, the resulting TEES system is capable of detecting binary relations as well as complex event structures. We show that this event extraction system has good performance, reaching the first place in the BioNLP'09 Shared Task on Event Extraction. Subsequently, TEES has achieved several first ranks in the BioNLP'11 and BioNLP'13 Shared Tasks, as well as shown competitive performance in the binary relation Drug-Drug Interaction Extraction 2011 and 2013 shared tasks. The Turku Event Extraction System is published as a freely available open-source project, documenting the research in detail as well as making the method available for practical applications. In particular, in this thesis we describe the application of the event extraction method to PubMed-scale text mining, showing how the developed approach not only shows good performance, but is generalizable and applicable to large-scale real-world text mining projects. Finally, we discuss related literature, summarize the contributions of the work and present some thoughts on future directions for biomedical event extraction. This thesis includes and builds on six original research publications. The first of these introduces the analysis of dependency parses that leads to development of TEES. The entries in the three BioNLP Shared Tasks, as well as in the DDIExtraction 2011 task are covered in four publications, and the sixth one demonstrates the application of the system to PubMed-scale text mining.

Extraction de motifs dans la rédaction collaborative sur les Wikis

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Diversified query expansion

La diversification des résultats de recherche (DRR) vise à sélectionner divers documents à partir des résultats de recherche afin de couvrir autant d’intentions que possible. Dans les approches existantes, on suppose que les résultats initiaux sont suffisamment diversifiés et couvrent bien les aspects de la requête. Or, on observe souvent que les résultats initiaux n’arrivent pas à couvrir certains aspects. Dans cette thèse, nous proposons une nouvelle approche de DRR qui consiste à diversifier l’expansion de requête (DER) afin d’avoir une meilleure couverture des aspects. Les termes d’expansion sont sélectionnés à partir d’une ou de plusieurs ressource(s) suivant le principe de pertinence marginale maximale. Dans notre première contribution, nous proposons une méthode pour DER au niveau des termes où la similarité entre les termes est mesurée superficiellement à l’aide des ressources. Quand plusieurs ressources sont utilisées pour DER, elles ont été uniformément combinées dans la littérature, ce qui permet d’ignorer la contribution individuelle de chaque ressource par rapport à la requête. Dans la seconde contribution de cette thèse, nous proposons une nouvelle méthode de pondération de ressources selon la requête. Notre méthode utilise un ensemble de caractéristiques qui sont intégrées à un modèle de régression linéaire, et génère à partir de chaque ressource un nombre de termes d’expansion proportionnellement au poids de cette ressource. Les méthodes proposées pour DER se concentrent sur l’élimination de la redondance entre les termes d’expansion sans se soucier si les termes sélectionnés couvrent effectivement les différents aspects de la requête. Pour pallier à cet inconvénient, nous introduisons dans la troisième contribution de cette thèse une nouvelle méthode pour DER au niveau des aspects. Notre méthode est entraînée de façon supervisée selon le principe que les termes reliés doivent correspondre au même aspect. Cette méthode permet de sélectionner des termes d’expansion à un niveau sémantique latent afin de couvrir autant que possible différents aspects de la requête. De plus, cette méthode autorise l’intégration de plusieurs ressources afin de suggérer des termes d’expansion, et supporte l’intégration de plusieurs contraintes telles que la contrainte de dispersion. Nous évaluons nos méthodes à l’aide des données de ClueWeb09B et de trois collections de requêtes de TRECWeb track et montrons l’utilité de nos approches par rapport aux méthodes existantes.

Selecting object pairs for action: is the active object always first?

Perception is linked to action via two routes: a direct route based on affordance information in the environment and an indirect route based on semantic knowledge about objects. The present study explored the factors modulating the recruitment of the two routes, in particular which factors affecting the selection of paired objects. In Experiment 1, we presented real objects among semantically related or unrelated distracters. Participants had to select two objects that can interact. The presence of distracters affected selection times, but not the semantic relations of the objects with the distracters. Furthermore, participants first selected the active object (e.g. teaspoon) with their right hand, followed by the passive object (e.g. mug), often with their left hand. In Experiment 2, we presented pictures of the same objects with no hand grip, congruent or incongruent hand grip. Participants had to decide whether the two objects can interact. Action decisions were faster when the presentation of the active object preceded the presentation of the passive object, and when the grip was congruent. Interestingly, participants were slower when the objects were semantically but not functionally related; this effect increased with congruently gripped objects. Our data showed that action decisions in the presence of strong affordance cues (real objects, pictures of congruently gripped objects) relied on sensory-motor representation, supporting the direct route from perception-to-action that bypasses semantic knowledge. However, in the case of weak affordance cues (pictures), semantic information interfered with action decisions, indicating that semantic knowledge impacts action decisions. The data support the dual-route account from perception-to-action.