MOTIVATION: This article describes NEREL-BIO-an annotation scheme and corpus of PubMed abstracts in Russian and smaller number of abstracts in English. NEREL-BIO extends the general domain dataset NEREL by introducing domain-specific entity types. NEREL-BIO annotation scheme covers both general and biomedical domains making it suitable for domain transfer experiments. NEREL-BIO provides annotation for nested named entities as an extension of the scheme employed for NEREL. Nested named entities may cross entity boundaries to connect to shorter entities nested within longer entities, making them harder to detect. RESULTS: NEREL-BIO contains annotations for 700+ Russian and 100+ English abstracts. All English PubMed annotations have corresponding Russian counterparts. Thus, NEREL-BIO comprises the following specific features: annotation of nested named entities, it can be used as a benchmark for cross-domain (NEREL → NEREL-BIO) and cross-language (English → Russian) transfer. We experiment with both transformer-based sequence models and machine reading comprehension models and report their results. AVAILABILITY AND IMPLEMENTATION: The dataset and annotation guidelines are freely available at https://github.com/nerel-ds/NEREL-BIO.

In this paper, we describe entity linking annotation over nested named entities in the recently released Russian NEREL dataset for information extraction. The NEREL collection is currently the largest Russian dataset annotated with entities and relations. It includes 933 news texts with annotation of 29 entity types and 49 relation types. The paper describes the main design principles behind NEREL’s entity linking annotation, provides its statistics, and reports evaluation results for several entity linking baselines. To date, 38,152 entity mentions in 933 documents are linked to Wikidata. The NEREL dataset is publicly available.

Abstract We propose an integrated deep learning model for morphological segmentation, morpheme tagging, part-of-speech (POS) tagging, and syntactic parsing onto dependencies, using cross-level contextual information flow for every word, from segments to dependencies, with an attention mechanism at horizontal flow. Our model extends the work of Nguyen and Verspoor ((2018). Proceedings of the CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. The Association for Computational Linguistics, pp. 81–91.) on joint POS tagging and dependency parsing to also include morphological segmentation and morphological tagging. We report our results on several languages. Primary focus is agglutination in morphology, in particular Turkish morphology, for which we demonstrate improved performance compared to models trained for individual tasks. Being one of the earlier efforts in joint modeling of syntax and morphology along with dependencies, we discuss prospective guidelines for future comparison.

The application of deep learning to 3D point clouds is challenging due to its lack of order. Inspired by the point embeddings of PointNet and the edge embeddings of DGCNNs, we propose three improvements to the task of point cloud analysis. First, we introduce a novel feature-attentive neural network layer, a FAT layer, that combines both global point-based features and local edge-based features in order to generate better embeddings. Second, we find that applying the same attention mechanism across two different forms of feature map aggregation, max pooling and average pooling, gives better performance than either alone. Third, we observe that residual feature reuse in this setting propagates information more effectively between the layers, and makes the network easier to train. Our architecture achieves state-of-the-art results on the task of point cloud classification, as demonstrated on the ModelNet40 dataset, and an extremely competitive performance on the ShapeNet part segmentation challenge.

This article presents a probabilistic hierarchical clustering model for morphological segmentation. In contrast to existing approaches to morphology learning, our method allows learning hierarchical organization of word morphology as a collection of tree structured paradigms. The model is fully unsupervised and based on the hierarchical Dirichlet process. Tree hierarchies are learned along with the corresponding morphological paradigms simultaneously. Our model is evaluated on Morpho Challenge and shows competitive performance when compared to state-of-the-art unsupervised morphological segmentation systems. Although we apply this model for morphological segmentation, the model itself can also be used for hierarchical clustering of other types of data.

We propose a structured generative latent variable model that integrates information from multiple contextual representations for Word Sense Induction. Our approach jointly models global lexical, local lexical and dependency syntactic context. Each context type is associated with a latent variable and the three types of variables share a hierarchical structure. We use skip-gram based word and dependency context embeddings to construct all three types of representations, reducing the total number of parameters to be estimated and enabling better generalization. We describe an EM algorithm to efficiently estimate model parameters and use the Integrated Complete Likelihood criterion to automatically estimate the number of senses. Our model achieves state-of-the-art results on the SemEval-2010 and SemEval-2013 Word Sense Induction datasets.

We compare different word embeddings from a standard window based skipgram model, a skipgram model trained using dependency context features and a novel skipgram variant that utilizes additional information from dependency graphs.We explore the effectiveness of the different types of word embeddings for word similarity and sentence classification tasks.We consider three common sentence classification tasks: question type classification on the TREC dataset, binary sentiment classification on Stanford's Sentiment Treebank and semantic relation classification on the SemEval 2010 dataset.For each task we use three different classification methods: a Support Vector Machine, a Convolutional Neural Network and a Long Short Term Memory Network.Our experiments show that dependency based embeddings outperform standard window based embeddings in most of the settings, while using dependency context embeddings as additional features improves performance in all tasks regardless of the classification method.Our embeddings and code are available at https://www.

The research on community question answering (CQA) has been paid increasing attention in recent years. In CQA, to reduce the number of unanswered questions and the time for askers to wait, it is very necessary to identify relevant experts or best answers for these questions. Generally, the experts' answers are more likely to be the best answers. Existing studies considered that user expertise is reflected by the voting scores of both answers and questions. However, voting scores of questions are not really related to user expertise. In this paper, we proposed a new probabilistic model to depict users' expertise based on answers and their descriptive ability based on questions. To exploit social information in CQA, the link analysis is also considered. Extensive experiments on the large Stack Overflow dataset demonstrate that our methods can achieve comparable or even better performance than the state-of-the-art models.

Social networking services are attracting increasing interest in the domain of community discovery. In social networks, the interactions among users are very frequent by sending emails, posting tweets, and sharing comments online, etc. Such networks usually include rich sentiment information, which can provide us with useful resources for identifying communities with different sentiment-topic distributions. Most conventional community discovery methods only consider the social links among users, which ignore the valuable content information. Recent studies have focused on community detection by integrating both links and content. However, most of these methods are not available for identifying sentiment-topic based communities. In this paper, we propose two novel community discovery models by combining social links, author based topics and sentiment information to identify communities with different sentiment-topic distributions. We evaluate our models on two real-world datasets, and the experimental results demonstrate the effectiveness of our proposed models.

Community question answering (CQA) sites provide us online platforms to post questions or answers. Generally, there are a great number of questions waiting to be answered by expert users. However, most of answerers are ordinary with just basic background knowledge in certain areas. To help askers to get their preferable answers, a set of possible expert users should be recommended. There have been some studies on the expert recommendation in CQA, the latest work models the user expertise under topics, where each topic is learnt based on the content and tags of questions and answers. Practically, such topics are too general, whereas question tags can be more informative and valuable than the topic of each question. In this paper, we study the user expertise under tags. Experimental analysis on a large data set from Stack Overflow demonstrates that our method performs better than the up-to-date method.

Abstract Interactive question answering (QA), where a dialogue interface enables follow-up and clarification questions, is a recent although long-advocated field of research. We report on the design and implementation of YourQA, our open-domain, interactive QA system. YourQA relies on a Web search engine to obtain answers to both fact-based and complex questions, such as descriptions and definitions. We describe the dialogue moves and management model making YourQA interactive, and discuss the architecture, implementation and evaluation of its chat-based dialogue interface. Our Wizard-of-Oz study and final evaluation results show how the designed architecture can effectively achieve open-domain, interactive QA.

Word Sense Induction (WSI) is the task of identifying the different senses (uses) of a target word in a given text. Traditional graph-based approaches create and then cluster a graph, in which each vertex corresponds to a word that co-occurs with the target word, and edges between vertices are weighted based on the co-occurrence frequency of their associated words. In contrast, in our approach each vertex corresponds to a collocation that co-occurs with the target word, and edges between vertices are weighted based on the co-occurrence frequency of their associated collocations. A smoothing technique is applied to identify more edges between vertices and the resulting graph is then clustered. Our evaluation under the framework of SemEval-2007 WSI task shows the following: (a) our approach produces less sense-conflating clusters than those produced by traditional graph-based approaches, (b) our approach outperforms the existing state-of-the-art results.

Segmenting the speech signals on the basis of time-frequency analysis is the most natural approach. Boundaries are located in places where energy of some frequency subband rapidly changes. Speech segmentation method which bases on discrete wavelet transform, the resulting power spectrum and its derivatives is presented. This information allows to locate the boundaries of phonemes. A statistical classification method was used to check which features are useful. The efficiency of segmentation was verified on a male speaker taken from a corpus of Polish language.

We examine the implementation of clarification dialogues, a mechanism for ensuring that question answering systems take into account user goals by allowing them to ask series of related questions either by refining or expanding on previous questions with follow-up questions, in the context of open domain Question Answering systems. We develop an algorithm for clarification dialogue recognition through the analysis of collected data on clarification dialogues and examine the importance of clarification dialogue recognition for question answering. The algorithm is evaluated and shown to successfully recognize the start and continuation of clarification dialogues in 94% of cases. We then show the usefulness of the algorithm by demonstrating how the recognition of clarification dialogues can simplify the task of answer retrieval.

We propose a formal system for representing the available readings of sentences displaying quantifier scope ambiguity, in which partial scopes may be expressed.We show that using a theory of scope availability based upon the functionargument structure of a sentence allows a deterministic, polynomial time test for the availability of a reading, while solving the same problem within theories based on the well-formedness of sentences in the meaning language has been shown to be NP-hard.

We describe an implementation in Carpenter's typed feature formalism, ALE, of a discourse grammar of the kind proposed by Scha, Polanyi, et al.We examine their method for resolving

This paper provides a model theoretic semantics to feature terms augmented with set descriptions. We provide constraints to specify HPSG style set descriptions, fixed cardinality set descriptions, set-membership constraints, restricted universal role quantifications, set union, intersection, subset and disjointness. A sound, complete and terminating consistency checking procedure is provided to determine the consistency of any given term in the logic. It is shown that determining consistency of terms is a NP-complete problem.