Scene Classification Using a Hybrid Generative/Discriminative Approach

We investigate whether dimensionality reduction using a latent generative model is beneficial for the task of weakly supervised scene classification. In detail, we are given a set of labeled images of scenes (for example, coast, forest, city, river, etc.), and our objective is to classify a new image into one of these categories. Our approach consists of first discovering latent ";topics"; using probabilistic Latent Semantic Analysis (pLSA), a generative model from the statistical text literature here applied to a bag of visual words representation for each image, and subsequently, training a multiway classifier on the topic distribution vector for each image. We compare this approach to that of representing each image by a bag of visual words vector directly and training a multiway classifier on these vectors. To this end, we introduce a novel vocabulary using dense color SIFT descriptors and then investigate the classification performance under changes in the size of the visual vocabulary, the number of latent topics learned, and the type of discriminative classifier used (k-nearest neighbor or SVM). We achieve superior classification performance to recent publications that have used a bag of visual word representation, in all cases, using the authors' own data sets and testing protocols. We also investigate the gain in adding spatial information. We show applications to image retrieval with relevance feedback and to scene classification in videos

Mesoclitic Romance futures and conditionals vs participle preposing

(INFINITIVE + CLITIC + AUX) is an evidential configuration in Old Spanish and Old Catalan, whereas (PARTICIPLE + CLITIC + AUX) is an instance of weak or unmarked focus fronting. The evidentiality of mesoclitic structures can be put forward on the bases of three main arguments: a) mesoclisis is not compulsory (i.e., whenever you have a clitic, you can either have mesoclisis or proclisis/enclisis); b) mesoclitic futures and conditionals areattested in interrogative sentences (with wh- elements); and c) they are not found in derived adverbial clauses (which is what you expect if they have an evidential value, since they bring about intervention effects corresponding to the derivational account of conditional and temporal sentences, for example - see Haegeman 2007 and ff.), and are related to high modal expressions (thus interfering with MoodPIrrealis)

Estudio contrastivo de transposición semántica entre el lenguaje metafórico en lengua inglesa y española

El projecte de tesi doctoral que estem portant a terme requereix una mitjana de quatre anys de durada. A hores d‟ara, nou mesos després de la concessió de l‟ajut, estem treballant per a la defensa del projecte, que farem en el primer trimestre de l‟any 2012. El nostre treball pertany al camp de la lingüística i està directament relacionat amb la lexicografia i la semàntica. L‟objectiu final és el disseny de materials per a la traducció i l‟ensenyament de llengües. L‟ús de lèxic és el tema escollit. D‟una banda, volem esbrinar com autors consagrats de literatura utilitzen el llenguatge. Per això, mirem de diferenciar el que es consideran usos “normals” del llenguatge del que serien “explotacions”, especialmente aquelles relacionades amb l‟extensió de significat i, més específicament, amb la metáfora. D‟altra banda, perseguim fer una proposta de material destinat tant a l‟ensenyament de llengües estrangeres com a la traducció (els idiomas amb els quals treballem són anglès i espanyol). Les teories que utilitzem en la nostra recerca són la teoria del lèxic generatiu (de J. Pustejovsky) i la teoria de normes i explotacions (de P. Hanks). Encara no hem decidit quins autors estudiarem, tot i estar consultant obres d‟escriptors com V. S. Naipaul i J. Marías. El treball de recerca desenvolupat fins avui ha consistit en la redacció de la memòria final de màster (Màster Oficial en Formació de Professors d‟Espanyol com a Llengua Estrangera) en què s‟estudiava el tractament de la metáfora en diferents diccionaris d‟ELE i d‟EFL i en la cerca, lectura i selección de documents que podran ajudar-nos en la nostra empresa.

Emphatic Polarity from Latin to Romance

This paper offers an analysis of some emphatic polarity constructions in Spanish and Catalan from a diachronic and comparative perspective. We focus on the syntactic processes involved in the verbalization of this polarity, in both its positive and negative expression. Our main proposal is that the markers used in Spanish and Catalan (as well as other Romance languages) to reinforce polarity have acquired their status as the consequence of a focalization process. It will be argued that these elements have undergone leftward movement from their base-generated hierarchically low position to a prominent position in the left periphery of the sentence (see Rizzi 1997) and subsequently experienced a progressive bleaching of their original value (which is related to a verbal denotation) to end up being reanalyzed / grammaticalized as polarity markers

Individual differences in control of language interference in late bilinguals are mainly related to general executive abilities

Background: Recent research based on comparisons between bilinguals and monolinguals postulates that bilingualism enhances cognitive control functions, because the parallel activation of languages necessitates control of interference. In a novel approach we investigated two groups of bilinguals, distinguished by their susceptibility to cross-language interference, asking whether bilinguals with strong language control abilities ('non-switchers") have an advantage in executive functions (inhibition of irrelevant information, problem solving, planning efficiency, generative fluency and self-monitoring) compared to those bilinguals showing weaker language control abilities ('switchers"). Methods: 29 late bilinguals (21 women) were evaluated using various cognitive control neuropsychological tests [e.g., Tower of Hanoi, Ruff Figural Fluency Task, Divided Attention, Go/noGo] tapping executive functions as well as four subtests of the Wechsler Adult Intelligence Scale. The analysis involved t-tests (two independent samples). Non-switchers (n = 16) were distinguished from switchers (n = 13) by their performance observed in a bilingual picture-naming task. Results: The non-switcher group demonstrated a better performance on the Tower of Hanoi and Ruff Figural Fluency task, faster reaction time in a Go/noGo and Divided Attention task, and produced significantly fewer errors in the Tower of Hanoi, Go/noGo, and Divided Attention tasks when compared to the switchers. Non-switchers performed significantly better on two verbal subtests of the Wechsler Adult Intelligence Scale (Information and Similarity), but not on the Performance subtests (Picture Completion, Block Design). Conclusions: The present results suggest that bilinguals with stronger language control have indeed a cognitive advantage in the administered tests involving executive functions, in particular inhibition, self-monitoring, problem solving, and generative fluency, and in two of the intelligence tests. What remains unclear is the direction of the relationship between executive functions and language control abilities.

Online brands: Branding, possible worlds, and interactive grammars

This paper proposes to reflect from a semiotic perspective on the transformation that brands have undergone since the rise of the Internet. After a brief theoretical introduction to digital communication and the semiotics of brands, the case of the Google brand is analyzed by applying concepts of generative and interpretive semiotics. The paper holds that the iconic and linguistic enunciations are secondary with respect to interaction. In digital media interaction — the interactive experience that the Internet user lives — is a fundamental component of the hypermedia cocktail and occupies a central position in the brand building process. The article concludes with some of the questions and special characteristics raised by so-called eBranding.

Sobre l'aplicació de les transformacions en una gramàtica generativa i transformacional

Nota on s'intenta mostrar, amb pretensions essencialment pedagògiques, alguns fets lingüístics que permeten determinar l'ordre en què s'apliquen les transformacions en una gramàtica generativa i transformacional

La noció d''Illa transparent': defensa de les regles transderivacionals

El present treball pretén contribuir, des d'una òptica relativa a la llengua catalana, a la revolució lingüística de dues maneres distintes: primer de tot, contrastant algunes hipòtesis assumides pel corrent del transformacionalisme dins del qual operaré -la semàntica generativa- per a mostrar-ne les febleses; segonament, proposant una modificació teòrica consistent en la introducció en la gramòtica de la noció d'illa transparent, la definició de la qual requereix regles transderivacionals, definides com 'unes regles que operen, no en derivacions individuals, sinó en un conjunt de derivacions', és a dir, que relacionen, per a definir condicions de bona formació, estructures pertanyents a derivacions distintes

Semàntica i sintaxi (aspectes de la complementació en català)

Aspectes de la complementació en català

Modeling and Classifying Breast Tissue Density in Mammograms

We present a new approach to model and classify breast parenchymal tissue. Given a mammogram, first, we will discover the distribution of the different tissue densities in an unsupervised manner, and second, we will use this tissue distribution to perform the classification. We achieve this using a classifier based on local descriptors and probabilistic Latent Semantic Analysis (pLSA), a generative model from the statistical text literature. We studied the influence of different descriptors like texture and SIFT features at the classification stage showing that textons outperform SIFT in all cases. Moreover we demonstrate that pLSA automatically extracts meaningful latent aspects generating a compact tissue representation based on their densities, useful for discriminating on mammogram classification. We show the results of tissue classification over the MIAS and DDSM datasets. We compare our method with approaches that classified these same datasets showing a better performance of our proposal

Individual differences in control of language interference in late bilinguals are mainly related to general executive abilities

Background: Recent research based on comparisons between bilinguals and monolinguals postulates that bilingualism enhances cognitive control functions, because the parallel activation of languages necessitates control of interference. In a novel approach we investigated two groups of bilinguals, distinguished by their susceptibility to cross-language interference, asking whether bilinguals with strong language control abilities ('non-switchers") have an advantage in executive functions (inhibition of irrelevant information, problem solving, planning efficiency, generative fluency and self-monitoring) compared to those bilinguals showing weaker language control abilities ('switchers"). Methods: 29 late bilinguals (21 women) were evaluated using various cognitive control neuropsychological tests [e.g., Tower of Hanoi, Ruff Figural Fluency Task, Divided Attention, Go/noGo] tapping executive functions as well as four subtests of the Wechsler Adult Intelligence Scale. The analysis involved t-tests (two independent samples). Non-switchers (n = 16) were distinguished from switchers (n = 13) by their performance observed in a bilingual picture-naming task. Results: The non-switcher group demonstrated a better performance on the Tower of Hanoi and Ruff Figural Fluency task, faster reaction time in a Go/noGo and Divided Attention task, and produced significantly fewer errors in the Tower of Hanoi, Go/noGo, and Divided Attention tasks when compared to the switchers. Non-switchers performed significantly better on two verbal subtests of the Wechsler Adult Intelligence Scale (Information and Similarity), but not on the Performance subtests (Picture Completion, Block Design). Conclusions: The present results suggest that bilinguals with stronger language control have indeed a cognitive advantage in the administered tests involving executive functions, in particular inhibition, self-monitoring, problem solving, and generative fluency, and in two of the intelligence tests. What remains unclear is the direction of the relationship between executive functions and language control abilities.