SUMMARY: Named entity recognition (NER) is an important step in biomedical information extraction pipelines. Tools for NER should be easy to use, cover multiple entity types, be highly accurate and be robust toward variations in text genre and style. We present HunFlair, a NER tagger fulfilling these requirements. HunFlair is integrated into the widely used NLP framework Flair, recognizes five biomedical entity types, reaches or overcomes state-of-the-art performance on a wide set of evaluation corpora, and is trained in a cross-corpus setting to avoid corpus-specific bias. Technically, it uses a character-level language model pretrained on roughly 24 million biomedical abstracts and three million full texts. It outperforms other off-the-shelf biomedical NER tools with an average gain of 7.26 pp over the next best tool in a cross-corpus setting and achieves on-par results with state-of-the-art research prototypes in in-corpus experiments. HunFlair can be installed with a single command and is applied with only four lines of code. Furthermore, it is accompanied by harmonized versions of 23 biomedical NER corpora. AVAILABILITY AND IMPLEMENTATION: HunFlair ist freely available through the Flair NLP framework (https://github.com/flairNLP/flair) under an MIT license and is compatible with all major operating systems. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Tables are a popular and efficient means of presenting structured information. They are used extensively in various kinds of documents including web pages. Tables display information as a two-dimensional matrix, the semantics of which is conveyed by a mixture of structure (rows, columns), headers, caption, and content. Recent research has started to consider tables as first class objects, not just as an addendum to texts, yielding interesting results for problems like table matching, table completion, or value imputation. All of these problems inherently rely on an accurate measure for the semantic similarity of two tables. We present TabSim, a novel method to compute table similarity scores using deep neural networks. Conceptually, TabSim represents a table as a learned concatenation of embeddings of its caption, its content, and its structure. Given two tables in this representation, a Siamese neural network is trained to compute a score correlating with the tables' semantic similarity. To train and evaluate our method, we created a gold standard corpus consisting of 1500 table pairs extracted from biomedical articles and manually scored regarding their degree of similarity, and adopted two other corpora originally developed for a different yet similar task. Our evaluation shows that TabSim outperforms other table similarity measures on average by app. 7% pp F1-score in a binary similarity classification setting and by app. 1.5% pp in a ranking scenario. The codes and resources introduced here are available at https://github.com/Marhabibi/TabSim.

MOTIVATION: Several recent studies showed that the application of deep neural networks advanced the state-of-the-art in named entity recognition (NER), including biomedical NER. However, the impact on performance and the robustness of improvements crucially depends on the availability of sufficiently large training corpora, which is a problem in the biomedical domain with its often rather small gold standard corpora. RESULTS: We evaluate different methods for alleviating the data sparsity problem by pretraining a deep neural network (LSTM-CRF), followed by a rather short fine-tuning phase focusing on a particular corpus. Experiments were performed using 34 different corpora covering five different biomedical entity types, yielding an average increase in F1-score of ∼2 pp compared to learning without pretraining. We experimented both with supervised and semi-supervised pretraining, leading to interesting insights into the precision/recall trade-off. Based on our results, we created the stand-alone NER tool HUNER incorporating fully trained models for five entity types. On the independent CRAFT corpus, which was not used for creating HUNER, it outperforms the state-of-the-art tools GNormPlus and tmChem by 5-13 pp on the entity types chemicals, species and genes. AVAILABILITY AND IMPLEMENTATION: HUNER is freely available at https://hu-ner.github.io. HUNER comes in containers, making it easy to install and use, and it can be applied off-the-shelf to arbitrary texts. We also provide an integrated tool for obtaining and converting all 34 corpora used in our evaluation, including fixed training, development and test splits to enable fair comparisons in the future. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract Tables are a common way to present information in an intuitive and concise manner. They are used extensively in media such as scientific articles or web pages. Automatically analyzing the content of tables bears special challenges. One of the most basic tasks is determination of the orientation of a table: In column tables, columns represent one entity with the different attribute values present in the different rows; row tables are vice versa, and matrix tables give information on pairs of entities. In this paper, we address the problem of classifying a given table into one of the three layouts horizontal (for row tables), vertical (for column tables), and matrix. We describe DeepTable, a novel method based on deep neural networks designed for learning from sets. Contrary to previous state-of-the-art methods, this basis makes DeepTable invariant to the permutation of rows or columns, which is a highly desirable property as in most tables the order of rows and columns does not carry specific information. We evaluate our method using a silver standard corpus of 5500 tables extracted from biomedical articles where the layout was determined heuristically. DeepTable outperforms previous methods in both precision and recall on our corpus. In a second evaluation, we manually labeled a corpus of 300 tables and were able to confirm DeepTable to reach superior performance in the table layout classification task. The codes and resources introduced here are available at https://github.com/Marhabibi/DeepTable .

MOTIVATION: Text mining has become an important tool for biomedical research. The most fundamental text-mining task is the recognition of biomedical named entities (NER), such as genes, chemicals and diseases. Current NER methods rely on pre-defined features which try to capture the specific surface properties of entity types, properties of the typical local context, background knowledge, and linguistic information. State-of-the-art tools are entity-specific, as dictionaries and empirically optimal feature sets differ between entity types, which makes their development costly. Furthermore, features are often optimized for a specific gold standard corpus, which makes extrapolation of quality measures difficult. RESULTS: We show that a completely generic method based on deep learning and statistical word embeddings [called long short-term memory network-conditional random field (LSTM-CRF)] outperforms state-of-the-art entity-specific NER tools, and often by a large margin. To this end, we compared the performance of LSTM-CRF on 33 data sets covering five different entity classes with that of best-of-class NER tools and an entity-agnostic CRF implementation. On average, F1-score of LSTM-CRF is 5% above that of the baselines, mostly due to a sharp increase in recall. AVAILABILITY AND IMPLEMENTATION: The source code for LSTM-CRF is available at https://github.com/glample/tagger and the links to the corpora are available at https://corposaurus.github.io/corpora/ . CONTACT: [email protected].

This paper addresses the problem of keyword extraction from conversations, with the goal of using these keywords to retrieve, for each short conversation fragment, a small number of potentially relevant documents, which can be recommended to participants. However, even a short fragment contains a variety of words, which are potentially related to several topics; moreover, using an automatic speech recognition (ASR) system introduces errors among them. Therefore, it is difficult to infer precisely the information needs of the conversation participants. We first propose an algorithm to extract keywords from the output of an ASR system (or a manual transcript for testing), which makes use of topic modeling techniques and of a submodular reward function which favors diversity in the keyword set, to match the potential diversity of topics and reduce ASR noise. Then, we propose a method to derive multiple topically separated queries from this keyword set, in order to maximize the chances of making at least one relevant recommendation when using these queries to search over the English Wikipedia. The proposed methods are evaluated in terms of relevance with respect to conversation fragments from the Fisher, AMI, and ELEA conversational corpora, rated by several human judges. The scores show that our proposal improves over previous methods that consider only word frequency or topic similarity, and represents a promising solution for a document recommender system to be used in conversations.

In this paper, we present an approach for topic-level search and hyperlinking of video snippets, which relies on content-based recommendation and multimodal re-ranking techniques. We identify topic-level segments using transcripts or subtitles and enrich them with other metadata. Segments are indexed in a word vector space. Given a text query or an anchor, the most similar segments are retrieved using cosine similarity scores, which are then combined with visual similarity scores, computed as the distance from the anchor's visual concept vector. This approach has performed well on the MediaEval 2013 Search and Hyperlinking task, evaluated over 1260 hours of BBC TV broadcast, in terms of overall mean average precision. Experiments showed that topic-segments based on transcripts from automatic speech recognition level systems (ASR) led to better performance than the ones based on subtitles for both search and hyperlinking. Moreover, by analyzing the effect of multimodal re-ranking on hyperlinking performance, we emphasize the merits of rich visual information available in the anchors for the hyperlinking task, and the merits of ASR for large-scale search and hyperlinking.

This paper presents the MUST-VIS system for the MediaMixer/VideoLectures.NET Temporal Segmentation and Annotation Grand Challenge. The system allows users to visualize a lecture as a series of segments represented by keyword clouds, with relations to other similar lectures and segments. Segmentation is performed using a multi-factor algorithm which takes advantage of the audio (through automatic speech recognition and word-based segmentation) and video (through the detection of actions such as writing on the blackboard). The similarity across segments and lectures is computed using a content-based recommendation algorithm. Overall, the graph-based representation of segment similarity appears to be a promising and cost-effective approach to navigating lecture databases.

In this paper, a problem in spoken dialogue systems namely the menu problem, is introduced and solved by a POMDP model. To overcome the large size of the menu problem, a new method for achieving an optimal policy called divided POMDP method is introduced. Conditions for the problem to be solved by the proposed method are specified and the problem properties resulting in the given conditions are presented. The proposed method is evaluated using a typical menu problem with different menu sizes and it is shown that this method is superior to the conventional methods such as FRTDP for the problems it is capable to solve. Moreover, it converges faster in getting to an optimal policy.