Tratamiento inteligente de la información para ayuda a la toma de decisiones

Proyecto emergente centrado en el tratamiento inteligente de información procedente de diversas fuentes tales como micro-blogs, blogs, foros, portales especializados, etc. La finalidad es generar conocimiento a partir de la información semántica recuperada. Como resultado se podrán determinar las necesidades de los usuarios o mejorar la reputación de diferentes organizaciones. En este artículo se describen los problemas abordados, la hipótesis de trabajo, las tareas a realizar y los objetivos parciales alcanzados.

The role of context in producing item interactions and false memories

Cued recall with an extralist cue poses a challenge for contemporary memory theory in that there is a need to explain how episodic and semantic information are combined. A parallel activation and intersection approach proposes one such means by assuming that an experimental cue will elicit its preexisting semantic network and a context cue will elicit a list memory. These 2 sources of information are then combined by focusing on information that is common to the 2 sources. Two key predictions of that approach are examined: (a) Combining semantic and episodic information can lead to item interactions and false memories, and (b) these effects are limited to memory tasks that involve an episodic context cue. Five experiments demonstrate such item interactions and false memories in cued recall but not in free association. Links are drawn between the use of context in this setting and in other settings.

Intelligent encoding of concepts in web development retrieval

The main aim of the proposed approach presented in this paper is to improve Web information retrieval effectiveness by overcoming the problems associated with a typical keyword matching retrieval system, through the use of concepts and an intelligent fusion of confidence values. By exploiting the conceptual hierarchy of the WordNet (G. Miller, 1995) knowledge base, we show how to effectively encode the conceptual information in a document using the semantic information implied by the words that appear within it. Rather than treating a word as a string made up of a sequence of characters, we consider a word to represent a concept.

Lexical database enrichment through semi-automated morphological analysis

Derivational morphology proposes meaningful connections between words and is largely unrepresented in lexical databases. This thesis presents a project to enrich a lexical database with morphological links and to evaluate their contribution to disambiguation. A lexical database with sense distinctions was required. WordNet was chosen because of its free availability and widespread use. Its suitability was assessed through critical evaluation with respect to specifications and criticisms, using a transparent, extensible model. The identification of serious shortcomings suggested a portable enrichment methodology, applicable to alternative resources. Although 40% of the most frequent words are prepositions, they have been largely ignored by computational linguists, so addition of prepositions was also required. The preferred approach to morphological enrichment was to infer relations from phenomena discovered algorithmically. Both existing databases and existing algorithms can capture regular morphological relations, but cannot capture exceptions correctly; neither of them provide any semantic information. Some morphological analysis algorithms are subject to the fallacy that morphological analysis can be performed simply by segmentation. Morphological rules, grounded in observation and etymology, govern associations between and attachment of suffixes and contribute to defining the meaning of morphological relationships. Specifying character substitutions circumvents the segmentation fallacy. Morphological rules are prone to undergeneration, minimised through a variable lexical validity requirement, and overgeneration, minimised by rule reformulation and restricting monosyllabic output. Rules take into account the morphology of ancestor languages through co-occurrences of morphological patterns. Multiple rules applicable to an input suffix need their precedence established. The resistance of prefixations to segmentation has been addressed by identifying linking vowel exceptions and irregular prefixes. The automatic affix discovery algorithm applies heuristics to identify meaningful affixes and is combined with morphological rules into a hybrid model, fed only with empirical data, collected without supervision. Further algorithms apply the rules optimally to automatically pre-identified suffixes and break words into their component morphemes. To handle exceptions, stoplists were created in response to initial errors and fed back into the model through iterative development, leading to 100% precision, contestable only on lexicographic criteria. Stoplist length is minimised by special treatment of monosyllables and reformulation of rules. 96% of words and phrases are analysed. 218,802 directed derivational links have been encoded in the lexicon rather than the wordnet component of the model because the lexicon provides the optimal clustering of word senses. Both links and analyser are portable to an alternative lexicon. The evaluation uses the extended gloss overlaps disambiguation algorithm. The enriched model outperformed WordNet in terms of recall without loss of precision. Failure of all experiments to outperform disambiguation by frequency reflects on WordNet sense distinctions.

Aspects of the learner's dictionary with special reference to advanced Pakistani learners of English

The present work is an empirical investigation into the lq`reference skills' of Pakistani learners and their language needs on semantic, phonetic, lexical and pragmatic levels in the dictionary. The introductory chapter discusses the relatively problematic nature of lexis in comparison with the other aspects in EFL learning and spells out the aim of this study. Chapter two provides an analytical survey of the various types of research undertaken in different contexts of the dictionary and explains the eclectic approach adopted in the present work. Chapter three studies the `reference skills' of this category of learners in the background of highly sophisticated information structure of learners' dictionaries under evaluation and suggests some measures for improvement in this context. Chapter four considers various criteria, eg. pedagogic, linguistic and sociolinguistic for determining the macro-structure of learner's dictionary with a focus on specific Ll speakers. Chapter five is concerned with various aspects of the semantic information provided in the dictionaries matched against the needs of Pakistani learners with regard to both comprehension and production. The type, scale and presentation of grammatical information in the dictionary is analysed in chapter six with the object of discovering their role and utility for the learner. Chapter seven explores the rationale for providing phonological information, the extent to which this guidance is vital and the problems of phonetic symbols employed in the dictionaries. Chapter eight brings into perspective the historical background of English-Urdu bilingual lexicography and evalutes the currently popular bilingual dictionaries among the student community, with the aim of discovering the extent to which they have taken account of the modern tents of lexicography and investigating their validity as a useful reference tool in the learning of English language. The final chapter concludes the findings of individual aspects in a coherent fashion to assess the viability of the original hypothesis that learners' dictionaries if compiled with a specific set of users in mind would be more useful.

The effective use of implicit parallelism through the use of an object-oriented programming language

This thesis explores translating well-written sequential programs in a subset of the Eiffel programming language - without syntactic or semantic extensions - into parallelised programs for execution on a distributed architecture. The main focus is on constructing two object-oriented models: a theoretical self-contained model of concurrency which enables a simplified second model for implementing the compiling process. There is a further presentation of principles that, if followed, maximise the potential levels of parallelism. Model of Concurrency. The concurrency model is designed to be a straightforward target for mapping sequential programs onto, thus making them parallel. It aids the compilation process by providing a high level of abstraction, including a useful model of parallel behaviour which enables easy incorporation of message interchange, locking, and synchronization of objects. Further, the model is sufficient such that a compiler can and has been practically built. Model of Compilation. The compilation-model's structure is based upon an object-oriented view of grammar descriptions and capitalises on both a recursive-descent style of processing and abstract syntax trees to perform the parsing. A composite-object view with an attribute grammar style of processing is used to extract sufficient semantic information for the parallelisation (i.e. code-generation) phase. Programming Principles. The set of principles presented are based upon information hiding, sharing and containment of objects and the dividing up of methods on the basis of a command/query division. When followed, the level of potential parallelism within the presented concurrency model is maximised. Further, these principles naturally arise from good programming practice. Summary. In summary this thesis shows that it is possible to compile well-written programs, written in a subset of Eiffel, into parallel programs without any syntactic additions or semantic alterations to Eiffel: i.e. no parallel primitives are added, and the parallel program is modelled to execute with equivalent semantics to the sequential version. If the programming principles are followed, a parallelised program achieves the maximum level of potential parallelisation within the concurrency model.

A deficit in encoding the order of visual sequences in developmental dyslexia: a non -phonological deficit

The present thesis tested the hypothesis of Stanovich, Siegel, & Gottardo (1997) that surface dyslexia is the result of a milder phonological deficit than that seen in phonological dyslexia coupled with reduced reading experience. We found that a group of adults with surface dyslexia showed a phonological deficit that was commensurate with that shown by a group of adults with phonological dyslexia (matched for chronological age and verbal and non-verbal IQ) and normal reading experience. We also showed that surface dyslexia cannot be accounted for by a semantic impairment or a deficit in the verbal learning and recall of lexical-semantic information (such as meaningful words), as both dyslexic subgroups performed the same. This study has replicated the results of our published study that surface dyslexia is not the consequence of a mild retardation or reduced learning opportunities but a separate impairment linked to a deficit in written lexical learning, an ability needed to create novel lexical representations from a series of unrelated visual units, which is independent from the phonological deficit (Romani, Di Betta, Tsouknida & Olson, 2008). This thesis also provided evidence that a selective nonword reading deficit in developmental dyslexia persists beyond poor phonology. This was shown by finding a nonword reading deficit even in the presence of normal regularity effects in the dyslexics (when compared to both reading and spelling-age matched controls). A nonword reading deficit was also found in the surface dyslexics. Crucially, this deficit was as strong as in the phonological dyslexics despite better functioning of the sublexical route for the former. These results suggest that a nonword reading deficit cannot be solely explained by a phonological impairment. We, thus, suggested that nonword reading should also involve another ability relating to the processing of novel visual orthographic strings, which we called 'orthographic coding'. We then investigated the ability to process series of independent units within multi-element visual arrays and its relationship with reading and spelling problems. We identified a deficit in encoding the order of visual sequences (involving both linguistic and nonlinguistic information) which was significantly associated with word and nonword processing. More importantly, we revealed significant contributions to orthographic skills in both dyslexic and control individuals, even after age, performance IQ and phonological skills were controlled. These results suggest that spelling and reading do not only tap phonological skills but also order encoding skills.

Progressive structural analysis for dynamic recognition of on-line handwritten mathematical expressions

Structural analysis in handwritten mathematical expressions focuses on interpreting the recognized symbols using geometrical information such as relative sizes and positions of the symbols. Most existing approaches rely on hand-crafted grammar rules to identify semantic relationships among the recognized mathematical symbols. They could easily fail when writing errors occurred. Moreover, they assume the availability of the whole mathematical expression before being able to analyze the semantic information of the expression. To tackle these problems, we propose a progressive structural analysis (PSA) approach for dynamic recognition of handwritten mathematical expressions. The proposed PSA approach is able to provide analysis result immediately after each written input symbol. This has an advantage that users are able to detect any recognition errors immediately and correct only the mis-recognized symbols rather than the whole expression. Experiments conducted on 57 most commonly used mathematical expressions have shown that the PSA approach is able to achieve very good performance results.

A generalized multidimensional index structure for multimedia data to support content-based similarity searches in a collaborative search environment

Since multimedia data, such as images and videos, are way more expressive and informative than ordinary text-based data, people find it more attractive to communicate and express with them. Additionally, with the rising popularity of social networking tools such as Facebook and Twitter, multimedia information retrieval can no longer be considered a solitary task. Rather, people constantly collaborate with one another while searching and retrieving information. But the very cause of the popularity of multimedia data, the huge and different types of information a single data object can carry, makes their management a challenging task. Multimedia data is commonly represented as multidimensional feature vectors and carry high-level semantic information. These two characteristics make them very different from traditional alpha-numeric data. Thus, to try to manage them with frameworks and rationales designed for primitive alpha-numeric data, will be inefficient. An index structure is the backbone of any database management system. It has been seen that index structures present in existing relational database management frameworks cannot handle multimedia data effectively. Thus, in this dissertation, a generalized multidimensional index structure is proposed which accommodates the atypical multidimensional representation and the semantic information carried by different multimedia data seamlessly from within one single framework. Additionally, the dissertation investigates the evolving relationships among multimedia data in a collaborative environment and how such information can help to customize the design of the proposed index structure, when it is used to manage multimedia data in a shared environment. Extensive experiments were conducted to present the usability and better performance of the proposed framework over current state-of-art approaches.

Select, Link and Rank: Diversified Query Expansion and Entity Ranking using Wikipedia

A search query, being a very concise grounding of user intent, could potentially have many possible interpretations. Search engines hedge their bets by diversifying top results to cover multiple such possibilities so that the user is likely to be satisfied, whatever be her intended interpretation. Diversified Query Expansion is the problem of diversifying query expansion suggestions, so that the user can specialize the query to better suit her intent, even before perusing search results. We propose a method, Select-Link-Rank, that exploits semantic information from Wikipedia to generate diversified query expansions. SLR does collective processing of terms and Wikipedia entities in an integrated framework, simultaneously diversifying query expansions and entity recommendations. SLR starts with selecting informative terms from search results of the initial query, links them to Wikipedia entities, performs a diversity-conscious entity scoring and transfers such scoring to the term space to arrive at query expansion suggestions. Through an extensive empirical analysis and user study, we show that our method outperforms the state-of-the-art diversified query expansion and diversified entity recommendation techniques.

Towards structured, block-based PDF

The Portable Document Format (PDF), defined by Adobe Systems Inc. as the basis of its Acrobat product range, is discussed in some detail. Particular emphasis is given to its flexible object-oriented structure, which has yet to be fully exploited. It is currently used to represent not logical structure but simply a series of pages and associated resources. A definition of an Encapsulated PDF (EPDF) is presented, in which EPDF blocks carry with them their own resource requirements, together with geometrical and logical information. A block formatter called Juggler is described which can lay out EPDF blocks from various sources onto new pages. Future revisions of PDF supporting uniquely-named EPDF blocks tagged with semantic information would assist in composite-pagemakeup and could even lead to fully revisable PDF.

RICH AND EFFICIENT VISUAL DATA REPRESENTATION

Increasing the size of training data in many computer vision tasks has shown to be very effective. Using large scale image datasets (e.g. ImageNet) with simple learning techniques (e.g. linear classifiers) one can achieve state-of-the-art performance in object recognition compared to sophisticated learning techniques on smaller image sets. Semantic search on visual data has become very popular. There are billions of images on the internet and the number is increasing every day. Dealing with large scale image sets is intense per se. They take a significant amount of memory that makes it impossible to process the images with complex algorithms on single CPU machines. Finding an efficient image representation can be a key to attack this problem. A representation being efficient is not enough for image understanding. It should be comprehensive and rich in carrying semantic information. In this proposal we develop an approach to computing binary codes that provide a rich and efficient image representation. We demonstrate several tasks in which binary features can be very effective. We show how binary features can speed up large scale image classification. We present learning techniques to learn the binary features from supervised image set (With different types of semantic supervision; class labels, textual descriptions). We propose several problems that are very important in finding and using efficient image representation.

Effects of efficient fronto-temporal circuitry on lexical ambiguity resolution: converging evidence from cross-age comparisons in eye-tracking and ERP data

Eye-tracking was used to examine how younger and older adults use syntactic and semantic information to disambiguate noun/verb (NV) homographs (e.g., park). We find that young adults exhibit inflated first fixations to NV-homographs when only syntactic cues are available for disambiguation (i.e., in syntactic prose). This effect is eliminated with the addition of disambiguating semantic information. Older adults (60+) as a group fail to show the first fixation effect in syntactic prose; they instead reread NV homographs longer. This pattern mirrors that in prior event-related potential work (Lee & Federmeier, 2009, 2011), which reported a sustained frontal negativity to NV-homographs in syntactic prose for young adults, which was eliminated by semantic constraints. The frontal negativity was not observed in older adults as a group, although older adults with high verbal fluency showed the young-like pattern. Analyses of individual differences in eye-tracking patterns revealed a similar effect of verbal fluency in both young and older adults: high verbal fluency groups of both ages show larger first fixation effects, while low verbal fluency groups show larger downstream costs (rereading and/or refixating NV homographs). Jointly, the eye-tracking and ERP data suggest that effortful meaning selection recruits frontal brain areas important for suppressing contextually inappropriate meanings, which also slows eye movements. Efficacy of fronto-temporal circuitry, as captured by verbal fluency, predicts the success of engaging these mechanisms in both young and older adults. Failure to recruit these processes requires compensatory rereading or leads to comprehension failures (Lee & Federmeier, in press).

Exploring Hidden Coherent Feature Groups and Temporal Semantics for Multimedia Big Data Analysis

Thanks to the advanced technologies and social networks that allow the data to be widely shared among the Internet, there is an explosion of pervasive multimedia data, generating high demands of multimedia services and applications in various areas for people to easily access and manage multimedia data. Towards such demands, multimedia big data analysis has become an emerging hot topic in both industry and academia, which ranges from basic infrastructure, management, search, and mining to security, privacy, and applications. Within the scope of this dissertation, a multimedia big data analysis framework is proposed for semantic information management and retrieval with a focus on rare event detection in videos. The proposed framework is able to explore hidden semantic feature groups in multimedia data and incorporate temporal semantics, especially for video event detection. First, a hierarchical semantic data representation is presented to alleviate the semantic gap issue, and the Hidden Coherent Feature Group (HCFG) analysis method is proposed to capture the correlation between features and separate the original feature set into semantic groups, seamlessly integrating multimedia data in multiple modalities. Next, an Importance Factor based Temporal Multiple Correspondence Analysis (i.e., IF-TMCA) approach is presented for effective event detection. Specifically, the HCFG algorithm is integrated with the Hierarchical Information Gain Analysis (HIGA) method to generate the Importance Factor (IF) for producing the initial detection results. Then, the TMCA algorithm is proposed to efficiently incorporate temporal semantics for re-ranking and improving the final performance. At last, a sampling-based ensemble learning mechanism is applied to further accommodate the imbalanced datasets. In addition to the multimedia semantic representation and class imbalance problems, lack of organization is another critical issue for multimedia big data analysis. In this framework, an affinity propagation-based summarization method is also proposed to transform the unorganized data into a better structure with clean and well-organized information. The whole framework has been thoroughly evaluated across multiple domains, such as soccer goal event detection and disaster information management.

Semantic modeling method for configurable enterprise information systems

Business practices vary from one company to another and business practices often need to be changed due to changes of business environments. To satisfy different business practices, enterprise systems need to be customized. To keep up with ongoing business practice changes, enterprise systems need to be adapted. Because of rigidity and complexity, the customization and adaption of enterprise systems often takes excessive time with potential failures and budget shortfall. Moreover, enterprise systems often drag business behind because they cannot be rapidly adapted to support business practice changes. Extensive literature has addressed this issue by identifying success or failure factors, implementation approaches, and project management strategies. Those efforts were aimed at learning lessons from post implementation experiences to help future projects. This research looks into this issue from a different angle. It attempts to address this issue by delivering a systematic method for developing flexible enterprise systems which can be easily tailored for different business practices or rapidly adapted when business practices change. First, this research examines the role of system models in the context of enterprise system development; and the relationship of system models with software programs in the contexts of computer aided software engineering (CASE), model driven architecture (MDA) and workflow management system (WfMS). Then, by applying the analogical reasoning method, this research initiates a concept of model driven enterprise systems. The novelty of model driven enterprise systems is that it extracts system models from software programs and makes system models able to stay independent of software programs. In the paradigm of model driven enterprise systems, system models act as instructors to guide and control the behavior of software programs. Software programs function by interpreting instructions in system models. This mechanism exposes the opportunity to tailor such a system by changing system models. To make this true, system models should be represented in a language which can be easily understood by human beings and can also be effectively interpreted by computers. In this research, various semantic representations are investigated to support model driven enterprise systems. The significance of this research is 1) the transplantation of the successful structure for flexibility in modern machines and WfMS to enterprise systems; and 2) the advancement of MDA by extending the role of system models from guiding system development to controlling system behaviors. This research contributes to the area relevant to enterprise systems from three perspectives: 1) a new paradigm of enterprise systems, in which enterprise systems consist of two essential elements: system models and software programs. These two elements are loosely coupled and can exist independently; 2) semantic representations, which can effectively represent business entities, entity relationships, business logic and information processing logic in a semantic manner. Semantic representations are the key enabling techniques of model driven enterprise systems; and 3) a brand new role of system models; traditionally the role of system models is to guide developers to write system source code. This research promotes the role of system models to control the behaviors of enterprise.