Abstract

This paper describes the BUT ‘Jilebi’ team’s speech recognition systems created for the 2018 low resource speech recognition challenge for Indian languages. We investigate modifications of multilingual time-delay neural network (TDNN) architectures with transfer learning and compare them to bi-directional residual memory networks (BRMN) and bi-directional LSTM. Our best submission based on system combination achieved word error rates of 13.92% (Tamil), 14.71% (Telugu) and 14.06% (Gujarati). We present the details of submitted systems and also the post-evaluation analysis done for lexicon discovery using unsupervised word segmentation.

Abstract

Fricatives are produced by creating a turbulence in the air– flow by passing it through a stricture in the vocal tract cavity. Fricatives are characterized by their noise–like behavior, which makes it difficult to analyze. Difference in the place of articulation leads to different classes of fricatives. Identification of fricative segment boundaries in speech helps in improving the performance of several applications. The present study attempts towards the identification and classification of fricative segments in continuous speech, based on the statistical behavior of instantaneous spectral characteristics. The proposed method uses parameters such as the dominant resonance frequencies, the center of gravity along with the statistical moments of the spectrum obtained using the zero time windowing (ZTW) method. The ZTW spectra exhibits a high temporal resolution and therefore gives accurate segment boundaries in speech. The proposed algorithm is tested on the TIMIT dataset for English language. A high identification rate of 97.5% is achieved for segment boundaries of the sibilant fricative class. Voiced nonsibilants show a lower identification rate than their voiceless counterparts due to their vowel–like spectral characteristics. A high classification rate of 93.2% is achieved be

Abstract

The task of fact-checking has been formalised as the assessment of the truthfulness of a claim. Be it a political proclamation or a technological development, verification of a new tidbit of information before its propagation to the general public is of utmost importance. Failing to do so leads to the spread of misinformation, which is a devious tool. Fact-checking is commonly performed by journalists, manually looking up information pertaining to the statement in question. This is a drawn out and tedious process with a chance of the concerned person not covering the domain exhaustively. Some of this effort is reduced by the use of knowledge bases created over a period of time. In this work under Task 2 (Factuality) of the CLEF 2018 CheckThat! Lab, we detail a neural network based methodology which models the textual data of a claim based on various representations of its words and characters. An affixed attention mechanism allows us to encapsulate linguistic features common in false claims. We achieve an accuracy of 39.57% on the task dataset.

Abstract

The domain of cultural mapping deals with capturing and representing knowledge regarding diverse tangible and intangible aspects of cultural heritage, which involve hybrid data sets from incompatible sources with different terminologies and representational schemas. The complex schema integration problem in this domain can be addressed efficiently by use of formal ontologies which provide universal schema to represent data from heterogeneous sources. The paper describes an abstraction over the CIDOC CRM (ISO 21127:2014) ontology for the purpose of the cultural mapping of villages of India and discuss with case studies how cultural information in context of Indian villages can be captured, preserved and new relationships between cultural entities visualized using this ontology.

Abstract

Event detection is a key aspect of story development which is composed of multiple narrative layers. Most of the narratives are template-based and follow a narration theory. In this paper, we demonstrate a narrative from events detected in the international relations domain along with classification of events using our novel mediation ontology. We also introduce a novel method of classifying events through the mediation ontology. Our methodology involves action classification based on the verb categorization of Beth Levin, its arguments determined by universal dependencies and word2vec. The selected feature space is a result of mapping language entities to ontological entities where we obtain substantially good results. The narration also presents interactions of international actors over various topics as well as other visualizations.

Abstract

Prediction of trust relations between users of social networks is critical for finding credible information. Inferring trust is challenging, since userspecified trust relations are highly sparse and power-law distributed. In this paper, we explore utilizing community memberships for trust prediction in a principled manner. We also propose a novel method to model homophily that complements existing work. To the best of our knowledge, this is the first work that mathematically formulates an insight based on community memberships for unsupervised trust prediction. We propose and model the hypothesis that a user is more likely to develop a trust relation within the user’s community than outside it. Unlike existing work, our approach for encoding homophily directly links user-user similarities with the pair-wise trust model. We derive mathematical factors that model our hypothesis relating community memberships to trust relations and the homophily effect. Along with low-rank matrix factorization, they are combined into chTrust, the proposed multi-faceted optimization framework. Our experiments on the standard Ciao and Epinions datasets show that the proposed framework outperforms multiple unsupervised trust prediction baselines for all test user pairs as well as the low-degree segment, across evaluation setting

Abstract

Following approaches for understanding lexical meaning de- veloped by Y¯aska, Patanjali and Bhartrihari from Indian linguistic tra- ditions and extending approaches developed by Leibniz and Brentano in the modern times, a framework of formal ontology of language was devel- oped. This framework proposes that meaning of words are in-formed by intrinsic and extrinsic ontological structures. The paper aims to capture such intrinsic and extrinsic meanings of words for two major Indian lan- guages, namely, Hindi and Telugu. Parts-of-speech have been rendered into sense-types and sense-classes. Using them we have developed a gold- standard annotated lexical resource to support semantic understanding of a language. The resource has collection of Hindi and Telugu lexicons, which has been manually annotated by native speakers of the languages following our annotation guidelines. Further, the resource was utilised to derive adverbial sense-class distribution of verbs and k¯araka-verb sense- type distribution. Different corpora (news, novels) were compared using verb sense-types distribution. Word Embedding was used as an aid for the enrichment of the resource. This is a work in progress that aims at lexical coverage of language extensively

Abstract

CESNA model consists of twin Bi-directional Long Short-Term Memory Recurrent Neural Networks (Bi-LSTM RNN) with shared parameters joined by a contrastive loss function based on a similarity metric. The model learns the representations of resource-poor and resource-rich language in a common emoji space by using a similarity metric based on the emojis present in sentences from both languages. The model, hence, projects sentences with similar emojis closer to each other and the sentences with different emojis farther from one another. Experiments on large-scale Twitter datasets of resource-rich languages - English and Spanish and resource-poor languages - Hindi and Telugu reveal that CESNA outperforms the state-of-the-art emoji prediction approaches based on distributional semantics,

Abstract

The paper describes the enrichment of OntoSenseNet- a verb- centric lexical resource for Indian Languages. This resource contains a newly developed Telugu-Telugu dictionary. It is important because na- tive speakers can better annotate the senses when both the word and its meaning are in Telugu. Hence efforts are made to develop a soft copy of Telugu dictionary. Our resource also has manually annotated gold standard corpus consisting 8483 verbs, 253 adverbs and 1673 adjectives. Annotations are done by native speakers according to defined annotation guidelines. In this paper, we provide an overview of the annotation proce- dure and present the validation of our resource through inter-annotator agreement. Concepts of sense-class and sense-type are discussed. Addi- tionally, we discuss the potential of lexical sense-annotated corpora in improving word sense disambiguation (WSD) tasks. Telugu WordNet is crowd-sourced for annotation of individual words in synsets and is compared with the developed sense-annotated lexicon (OntoSenseNet) to examine the improvement. Also, we present a special categorization (spatio-temporal classification) of adjectives.

Abstract

With the increasing usage of internet, more and more data is being digitized including parliamentary debates but they are in an unstructured format. There is a need to convert them into structured format for linguistic analysis. Much work has been done on parliamentary data such as Hansard, American congressional floor-debate data on various aspects but less on pragmatics. In this paper, we provide a dataset for synopsis of Indian parliamentary debates and perform stance classification of speeches i.e identifying if the speaker is supporting the bill/issue or is against it. We also analyze the intention of the speeches beyond mere sentences i.e pragmatics in the parliament. Based on thorough manual analysis of the debates, we developed an annotation scheme of 4 mutually exclusive categories to analyze the purpose of