A Classification of Sandhi Rules for Suffix Separation in Malayalam

Suffix separation plays a vital role in improving the quality of training in the Statistical Machine Translation from English into Malayalam. The morphological richness and the agglutinative nature of Malayalam make it necessary to retrieve the root word from its inflected form in the training process. The suffix separation process accomplishes this task by scrutinizing the Malayalam words and by applying sandhi rules. In this paper, various handcrafted rules designed for the suffix separation process in the English Malayalam SMT are presented. A classification of these rules is done based on the Malayalam syllable preceding the suffix in the inflected form of the word (check_letter). The suffixes beginning with the vowel sounds like ആല, ഉെെ, ഇല etc are mainly considered in this process. By examining the check_letter in a word, the suffix separation rules can be directly applied to extract the root words. The quick look up table provided in this paper can be used as a guideline in implementing suffix separation in Malayalam language

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

Effect of phonetic complexity on word reading and repetition in deep dyslexia

This investigation moves beyond the traditional studies of word reading to identify how the production complexity of words affects reading accuracy in an individual with deep dyslexia (JO). We examined JO’s ability to read words aloud while manipulating both the production complexity of the words and the semantic context. The classification of words as either phonetically simple or complex was based on the Index of Phonetic Complexity. The semantic context was varied using a semantic blocking paradigm (i.e., semantically blocked and unblocked conditions). In the semantically blocked condition words were grouped by semantic categories (e.g., table, sit, seat, couch,), whereas in the unblocked condition the same words were presented in a random order. JO’s performance on reading aloud was also compared to her performance on a repetition task using the same items. Results revealed a strong interaction between word complexity and semantic blocking for reading aloud but not for repetition. JO produced the greatest number of errors for phonetically complex words in semantically blocked condition. This interaction suggests that semantic processes are constrained by output production processes which are exaggerated when derived from visual rather than auditory targets. This complex relationship between orthographic, semantic, and phonetic processes highlights the need for word recognition models to explicitly account for production processes.

Word sense disambiguation via high order of learning in complex networks

Complex networks have been employed to model many real systems and as a modeling tool in a myriad of applications. In this paper, we use the framework of complex networks to the problem of supervised classification in the word disambiguation task, which consists in deriving a function from the supervised (or labeled) training data of ambiguous words. Traditional supervised data classification takes into account only topological or physical features of the input data. On the other hand, the human (animal) brain performs both low- and high-level orders of learning and it has facility to identify patterns according to the semantic meaning of the input data. In this paper, we apply a hybrid technique which encompasses both types of learning in the field of word sense disambiguation and show that the high-level order of learning can really improve the accuracy rate of the model. This evidence serves to demonstrate that the internal structures formed by the words do present patterns that, generally, cannot be correctly unveiled by only traditional techniques. Finally, we exhibit the behavior of the model for different weights of the low- and high-level classifiers by plotting decision boundaries. This study helps one to better understand the effectiveness of the model. Copyright (C) EPLA, 2012

Content Classification of Multimedia Documents using Partitions of Low-Level Features

Audio-visual documents obtained from German TV news are classified according to the IPTC topic categorization scheme. To this end usual text classification techniques are adapted to speech, video, and non-speech audio. For each of the three modalities word analogues are generated: sequences of syllables for speech, “video words” based on low level color features (color moments, color correlogram and color wavelet), and “audio words” based on low-level spectral features (spectral envelope and spectral flatness) for non-speech audio. Such audio and video words provide a means to represent the different modalities in a uniform way. The frequencies of the word analogues represent audio-visual documents: the standard bag-of-words approach. Support vector machines are used for supervised classification in a 1 vs. n setting. Classification based on speech outperforms all other single modalities. Combining speech with non-speech audio improves classification. Classification is further improved by supplementing speech and non-speech audio with video words. Optimal F-scores range between 62% and 94% corresponding to 50% - 84% above chance. The optimal combination of modalities depends on the category to be recognized. The construction of audio and video words from low-level features provide a good basis for the integration of speech, non-speech audio and video.

Anatomy of word and sentence meaning

Reading and listening involve complex psychological processes that recruit many brain areas. The anatomy of processing English words has been studied by a variety of imaging methods. Although there is widespread agreement on the general anatomical areas involved in comprehending words, there are still disputes about the computations that go on in these areas. Examination of the time relations (circuitry) among these anatomical areas can aid in understanding their computations. In this paper, we concentrate on tasks that involve obtaining the meaning of a word in isolation or in relation to a sentence. Our current data support a finding in the literature that frontal semantic areas are active well before posterior areas. We use the subject’s attention to amplify relevant brain areas involved either in semantic classification or in judging the relation of the word to a sentence to test the hypothesis that frontal areas are concerned with lexical semantics and posterior areas are more involved in comprehension of propositions that involve several words.

The acute effects of mild traumatic brain injury on finger tapping with and without word repetition

This study aimed to investigate the acute effects of mild Traumatic Brain Injury (mTBI) on the performance of a finger tapping and word repetition dual task in order to determine working memory impairment in mTBI Sixty-four (50 male, 14 female) right-handed cases of mTBI and 26 (18 male and 8 female) right-handed cases of orthopaedic injuries were tested within 24 hours of injury. Patients with mTBI completed fewer correct taps in 10 seconds than patients with orthopaedic injuries, and female mTBI cases repeated fewer words. The size of the dual task decrement did not vary between groups. When added to a test battery including the Rapid Screen of Concussion (RSC; Comerford, Geffen, May, Medland T Geffen, 2002) and the Digit Symbol Substitution Test,finger tapping speed accounted for 1% of between groups variance and did not improve classification rates of male participants. While the addition of tapping rate did not improve the sensitivity and specificity of the RSC and DSST to mTBI in males, univariate analysis of motor performance in females indicated. that dual task performance might be diagnostic. An increase in female sample Size is warranted. These results confirm the view that there is a generalized slowing of processing ability following mTBI.

Learning sentiment classification model from labeled features

We propose a novel framework where an initial classifier is learned by incorporating prior information extracted from an existing sentiment lexicon. Preferences on expectations of sentiment labels of those lexicon words are expressed using generalized expectation criteria. Documents classified with high confidence are then used as pseudo-labeled examples for automatical domain-specific feature acquisition. The word-class distributions of such self-learned features are estimated from the pseudo-labeled examples and are used to train another classifier by constraining the model's predictions on unlabeled instances. Experiments on both the movie review data and the multi-domain sentiment dataset show that our approach attains comparable or better performance than exiting weakly-supervised sentiment classification methods despite using no labeled documents.

Automatically extracting polarity-bearing topics for cross-domain sentiment classification

Joint sentiment-topic (JST) model was previously proposed to detect sentiment and topic simultaneously from text. The only supervision required by JST model learning is domain-independent polarity word priors. In this paper, we modify the JST model by incorporating word polarity priors through modifying the topic-word Dirichlet priors. We study the polarity-bearing topics extracted by JST and show that by augmenting the original feature space with polarity-bearing topics, the in-domain supervised classifiers learned from augmented feature representation achieve the state-of-the-art performance of 95% on the movie review data and an average of 90% on the multi-domain sentiment dataset. Furthermore, using feature augmentation and selection according to the information gain criteria for cross-domain sentiment classification, our proposed approach performs either better or comparably compared to previous approaches. Nevertheless, our approach is much simpler and does not require difficult parameter tuning.

Automatic Classification of Bulgarian Adjectives on Emotional Semantic Axes

Report published in the Proceedings of the National Conference on "Education and Research in the Information Society", Plovdiv, May, 2014

A gloss composition and context clustering based distributed word sense representation model

In recent years, there has been an increasing interest in learning a distributed representation of word sense. Traditional context clustering based models usually require careful tuning of model parameters, and typically perform worse on infrequent word senses. This paper presents a novel approach which addresses these limitations by first initializing the word sense embeddings through learning sentence-level embeddings from WordNet glosses using a convolutional neural networks. The initialized word sense embeddings are used by a context clustering based model to generate the distributed representations of word senses. Our learned representations outperform the publicly available embeddings on half of the metrics in the word similarity task, 6 out of 13 sub tasks in the analogical reasoning task, and gives the best overall accuracy in the word sense effect classification task, which shows the effectiveness of our proposed distributed distribution learning model.

Detection and Classification of Whale Acoustic Signals

This dissertation focuses on two vital challenges in relation to whale acoustic signals: detection and classification.

In detection, we evaluated the influence of the uncertain ocean environment on the spectrogram-based detector, and derived the likelihood ratio of the proposed Short Time Fourier Transform detector. Experimental results showed that the proposed detector outperforms detectors based on the spectrogram. The proposed detector is more sensitive to environmental changes because it includes phase information.

In classification, our focus is on finding a robust and sparse representation of whale vocalizations. Because whale vocalizations can be modeled as polynomial phase signals, we can represent the whale calls by their polynomial phase coefficients. In this dissertation, we used the Weyl transform to capture chirp rate information, and used a two dimensional feature set to represent whale vocalizations globally. Experimental results showed that our Weyl feature set outperforms chirplet coefficients and MFCC (Mel Frequency Cepstral Coefficients) when applied to our collected data.

Since whale vocalizations can be represented by polynomial phase coefficients, it is plausible that the signals lie on a manifold parameterized by these coefficients. We also studied the intrinsic structure of high dimensional whale data by exploiting its geometry. Experimental results showed that nonlinear mappings such as Laplacian Eigenmap and ISOMAP outperform linear mappings such as PCA and MDS, suggesting that the whale acoustic data is nonlinear.

We also explored deep learning algorithms on whale acoustic data. We built each layer as convolutions with either a PCA filter bank (PCANet) or a DCT filter bank (DCTNet). With the DCT filter bank, each layer has different a time-frequency scale representation, and from this, one can extract different physical information. Experimental results showed that our PCANet and DCTNet achieve high classification rate on the whale vocalization data set. The word error rate of the DCTNet feature is similar to the MFSC in speech recognition tasks, suggesting that the convolutional network is able to reveal acoustic content of speech signals.

"Lancea", Lusitanian Word, and the "Lancienses" Ethnogenesis

Analysis of the word lancea, of Hispanic origin after Varro, and of place names, people´s names and personal names derived from it. It confirms that the spear was the most important weapon in the Bronze Age, belonging to the iuventus and used as heroic and divine symbol. This analysis confirms also the personality of the Lusitanians, a people related to the Celts but with more archaic archaeological, linguistic and cultural characteristics originated in the tradition of the Atlantic Bronze in the II millennium BC. It is also relevant to better know the organisation of Broze and Iron Age societies and the origin of Indo-Europeans peoples in Western Europe and of pre-Roman peoples of Iberia.

Machine learning classification of surgical pathology reports and chunk recognition for information extraction noise reduction

Background and aims: Machine learning techniques for the text mining of cancer-related clinical documents have not been sufficiently explored. Here some techniques are presented for the pre-processing of free-text breast cancer pathology reports, with the aim of facilitating the extraction of information relevant to cancer staging.

Materials and methods: The first technique was implemented using the freely available software RapidMiner to classify the reports according to their general layout: ‘semi-structured’ and ‘unstructured’. The second technique was developed using the open source language engineering framework GATE and aimed at the prediction of chunks of the report text containing information pertaining to the cancer morphology, the tumour size, its hormone receptor status and the number of positive nodes. The classifiers were trained and tested respectively on sets of 635 and 163 manually classified or annotated reports, from the Northern Ireland Cancer Registry.

Results: The best result of 99.4% accuracy – which included only one semi-structured report predicted as unstructured – was produced by the layout classifier with the k nearest algorithm, using the binary term occurrence word vector type with stopword filter and pruning. For chunk recognition, the best results were found using the PAUM algorithm with the same parameters for all cases, except for the prediction of chunks containing cancer morphology. For semi-structured reports the performance ranged from 0.97 to 0.94 and from 0.92 to 0.83 in precision and recall, while for unstructured reports performance ranged from 0.91 to 0.64 and from 0.68 to 0.41 in precision and recall. Poor results were found when the classifier was trained on semi-structured reports but tested on unstructured.

Conclusions: These results show that it is possible and beneficial to predict the layout of reports and that the accuracy of prediction of which segments of a report may contain certain information is sensitive to the report layout and the type of information sought.

On the effect of word frequency on distributional similarity

The dependency of word similarity in vector space models on the frequency of words has been noted in a few studies, but has received very little attention. We study the influence of word frequency in a set of 10 000 randomly selected word pairs for a number of different combinations of feature weighting schemes and similarity measures. We find that the similarity of word pairs for all methods, except for the one using singular value decomposition to reduce the dimensionality of the feature space, is determined to a large extent by the frequency of the words. In a binary classification task of pairs of synonyms and unrelated words we find that for all similarity measures the results can be improved when we correct for the frequency bias.