Search

Cognitive modeling involves the creation of computer simulations that emulate the internal processes of the mind. This set of papers are the five best representatives of the papers presented at the 17th International Conference on Cognitive Modeling, ICCM 2019. While they represent a diversity of techniques and tasks, they all also share a striking similarity: They make strong statements about the importance of accounting for individual differences.

Predicting the impact of academic papers can help scholars quickly identify the high-quality papers in the field. How to develop efficient predictive model for evaluating potential papers has attracted increasing attention in academia. Many studies have shown that early citations contribute to improving the performance of predicting the long-term impact of a paper. Besides early citations, some bibliometric features and altmetric features have also been explored for predicting the impact of academic papers. Furthermore, paper metadata text such as title, abstract and keyword contains valuable information which has effect on its citation count. However, present studies ignore the semantic information contained in the metadata text. In this paper, we propose a novel citation prediction model based on paper metadata text to predict the long-term citation count, and the core of our model is to obtain the semantic information from the metadata text. We use deep learning techniques to encode the metadata text, and then further extract high-level semantic features for learning the citation prediction task. We also integrate early citations for improving the prediction performance of the model. We show that our proposed model outperforms the state-of-the-art models in predicting the long-term citation count of the papers, and metadata semantic features are effective for improving the accuracy of the citation prediction models.

The existing plethora of document classification techniques exploits different data sources either from the content or metadata of research articles. Various journal publishers like Springer, Elsevier, IEEE, etc., do not provide open access to the content of research articles, whereas metadata is freely available there. Metadata like title, keyword, and abstract can serve as a better alternative to the content in various scenarios. In the current literature, researchers have assessed the role of some of the metadata individually. We believe that the collective contribution of metadata parameters can play a significant role in classifying research papers. This paper presents a comprehensive evaluation of the role of metadata, individually as well as in combinations to achieve the objective of research paper classification. Moreover, we have classified the research articles into ACM hierarchy root categories (e.g. general literature, hardware, software, etc.). In this comprehensive evaluation, we have assessed all the possible combinations of metadata features against different classifiers such as Random Forest, K Nearest Neighbor, and Decision Tree. The results of this research reveal that the title & keywords combination outperforms other combinations with an F-measure score of 0.88.

From the past half of a century, identification of the relevant documents is deemed an active area of research due to the rapid increase of data on the web. The traditional models to retrieve relevant documents are based on bibliographic information such as Bibliographic coupling, Co-citations, and Direct citations. However, in the recent past, the scientific community has started to employ textual features to improve existing models’ accuracy. In our previous study, we found that analysis of citations at a deep level (i.e., content level) can play a paramount role in finding more relevant documents than surface level (i.e., just bibliography details). We found that cited and citing papers have a high degree of relevancy when in-text citations frequency of the cited paper is more than five times in the citing paper’s text. This paper is an extension of our previous study in terms of its evaluation of a comprehensive dataset. Moreover, the study results are also compared with other state-of-the-art approaches i.e., content, metadata, and bibliography. For evaluation, a user study is conducted on selected papers from 1,200 documents (comprise about 16,000 references) of an online journal, Journal of Computer Science (J.UCS). The evaluation results indicate that in-text citation frequency has attained higher precision in finding relevant papers than other state-of-the-art techniques such as content, bibliographic coupling, and metadata-based techniques. The use of in-text citation may help in enhancing the quality of existing information systems and digital libraries. Further, more sophisticated measure may be redefined be considering the use of in-text citations.

To support nanocrystal device development, we have been working on a computational framework to utilize information in research papers on nanocrystal devices. We developed an annotated corpus called “ NaDev” (Nanocrystal Device Development) for this purpose. We also proposed an automatic information extraction system called “NaDevEx” (Nanocrystal Device Automatic Information Extraction Framework). NaDevEx aims at extracting information from research papers on nanocrystal devices using the NaDev corpus and machine-learning techniques. However, the characteristics of NaDevEx were not examined in detail. In this paper, we conduct system evaluation experiments for NaDevEx using the NaDev corpus. We discuss three main issues: system performance, compared with human annotators; the effect of paper type (synthesis or characterization) on system performance; and the effects of domain knowledge features (e.g., a chemical named entity recognition system and list of names of physical quantities) on system performance. We found that overall system performance was 89% in precision and 69% in recall. If we consider identification of terms that intersect with correct terms for the same information category as the correct identification, i.e., loose agreement (in many cases, we can find that appropriate head nouns such as temperature or pressure loosely match between two terms), the overall performance is 95% in precision and 74% in recall. The system performance is almost comparable with results of human annotators for information categories with rich domain knowledge information (source material). However, for other information categories, given the relatively large number of terms that exist only in one paper, recall of individual information categories is not high (39–73%); however, precision is better (75–97%). The average performance for synthesis papers is better than that for characterization papers because of the lack of training examples for characterization papers. Based on these results, we discuss future research plans for improving the performance of the system.

Projective Mapping has recently attracted a lot of attention and the main sensory data acquisition software packages have developed interfaces to collect projective mapping data. However, the comparison between paper and computer based projective mapping has never been reported. The objectives of this research were to 1) compare the consensus maps and panelists' performances for paper and computer based projective mapping and 2) analyze the panelists' strategies while performing either tasks. In the first part of the study, 32 panelists were asked to perform both paper and computer based projective mapping on 8 beer samples. In a second part of the study, 10 panelists were asked to repeat the tasks whilst “thinking aloud” their strategy. There was no significant difference in panelists' performance as assessed by the People Performance Index (PPI) between the paper and computer tasks. The consensus maps obtained were similar with respect to sample groupings, RV coefficients and variation explained by the first 2 dimensions. Individual panelists adopted similar strategies on paper and computer but strategies differed greatly between panelists.

Literature reviews are essential for any researcher trying to keep up to date with the burgeoning software engineering literature. FAST$^2$ is a novel tool for reducing the effort required for conducting literature reviews by assisting the researchers to find the next promising paper to read (among a set of unread papers). This paper describes FAST$^2$ and tests it on four large software engineering literature reviews conducted by Wahono (2015), Hall (2012), Radjenovi\'c (2013) and Kitchenham (2017). We find that FAST$^2$ is a faster and robust tool to assist researcher finding relevant SE papers which can compensate for the errors made by humans during the review process. The effectiveness of FAST$^2$ can be attributed to three key innovations: (1) a novel way of applying external domain knowledge (a simple two or three keyword search) to guide the initial selection of papers---which helps to find relevant research papers faster with less variances; (2) an estimator of the number of remaining relevant papers yet to be found---which in practical settings can be used to decide if the reviewing process needs to be terminated; (3) a novel self-correcting classification algorithm---automatically corrects itself, in cases where the researcher wrongly classifies a paper., Comment: 20+3 pages, 6 figures, 5 tables, and 4 algorithms. Accepted by Journal of Expert Systems with Applications

From the first issue of 2015 to the third issue of 2018, the journal published 116 papers, all of which were indexed by the core collection of Web of Sciences with an index rate of 100%. Among thes...

Keeping up with the rapid growth of Deep Learning (DL) research is a daunting task. While existing scientific literature search systems provide text search capabilities and can identify similar papers, gaining an in-depth understanding of a new approach or an application is much more complicated. Many publications leverage multiple modalities to convey their findings and spread their ideas - they include pseudocode, tables, images and diagrams in addition to text, and often make publicly accessible their implementations. It is important to be able to represent and query them as well. We utilize RDF Knowledge graphs (KGs) to represent multimodal information and enable expressive querying over modalities. In our demo we present an approach for extracting KGs from different modalities, namely text, architecture images and source code. We show how graph queries can be used to get insights into different facets (modalities) of a paper, and its associated code implementation. Our innovation lies in the multimodal nature of the KG we create. While our work is of direct interest to DL researchers and practitioners, our approaches can also be leveraged in other scientific domains.

Which are the segmentation algorithms proposed during 2018-2019 in CVPR that have CNN architecture?’ Answering this question involves identifying and analyzing the deep learning architecture diagrams from several research papers. Retrieving such information poses significant challenge as most of the existing academic search engines are primarily based on only the text content. In this paper, we introduce Diag2Graph, an end-to-end framework for parsing deep learning diagram-figures, that enables powerful search and retrieval of architectural details in research papers. Our proposed approach automatically localizes figures from research papers, classifies them, and analyses the content of the diagram-figures. The key steps in analyzing the Figure content is the extraction of the different components data and finding their structural relation. Finally, the extracted components and their relations are represented in the form of a deep knowledge graph. A thorough evaluation on a real-word annotated dataset has been done to demonstrate the efficacy of our approach.

Citation recommendation can help researchers quickly find supplementary or alternative references in massive academic resources. Current research on citation recommendation mainly focuses on the citing papers, resulting in the enormous cited papers are ignored, including the relations among cited papers and their citation context cited in citing papers. Moreover, cited paper’s content is often denoted with its original title and abstract, which is hard to acquire and rarely considers different citation motivations. Furthermore, the most appropriate method for semantic representation of cited papers’ relations and content is uncertain. Therefore, this paper studies citation recommendation from the perspective of semantic representation of cited papers’ relations and content. Firstly, four forms of citation context are designed and extracted as cited papers’ content considering citation motivations, as well as co-citation relationships are extracted as cited papers’ relations. Secondly, 132 methods are designed for generating semantic vector of cited paper, including four network embedding methods, 16 methods by combining four text representation algorithms with four forms of citation content, and 112 fusion methods. Finally, similarity among cited papers is calculated for citation recommendation and a quantitative evaluation method based on link prediction is designed, to find the most appropriate form of citation content and the optimal method. The result shows that doc2vecC (Document to Vector through Corruption) with the form of CS&SS (Current Sentences and Surrounding Sentences) performs best, in which the AUC (Area Under Curve) and MAP (Macro Average Precision) reach 0.877 and 0.889 and have increased by 0.462 and 0.370 compared with the worst-performing method. This performance is slightly improved by parameters adjustment, and a case study is performed whose results have further proved the effectiveness of this method. In addition, among four forms of cited papers’ content, CS&SS performs best in almost all methods. Furthermore, the fusion methods not always perform better than the single methods, where doc2vecC (CS&SS) performs better than the best fusion method GCN (Graph Convolutional Network). These results not only prove the effectiveness of citation recommendation from the perspective of cited paper, but also provide helpful and useful suggestions for method selection and citation content selection. The data and conclusions can be extended to other text mining-related tasks. Simultaneously, it is a preliminary research which needs to be further studied in other domains using emerging semantic representation methods.

The rapid growth of scientific literature is presenting several challenges for the search and discovery of research artifacts. Datasets are the backbone of scientific experiments. It is crucial to locate the datasets used or generated by previous research as building suitable datasets is costly in terms of time, money, and human labor. Hence automated mechanisms to aid the search and discovery of datasets from scientific publications can aid reproducibility and reusability of these valuable scientific artifacts. Here in this work, utilizing the next sentence prediction capability of language models, we show that a BERT-based entity recognition model with POS aware embedding can be effectively used to address this problem. Our investigation shows that identifying sentences containing dataset mentions in the first place proves critical to the task. Our method outperforms earlier ones and achieves an F1 score of 56.2 in extracting dataset mentions from research papers on a popular corpus of social science publications. We make our codes available at https://github.com/sandeep82945/data_discovery.

International audience; The task of tagging papers with semantic metadata in order to analyze their relatedness represents a good foundation for a paper recommender system. The analysis from this paper extends from previous research in order to create a graph of papers from a specific domain with the purpose of determining each article's importance within the considered corpus of papers. Moreover, as non-latent representations are powerful when used in conjunction with latent ones, our system retrieves semantically close words, not present in the paper, in order to improve the retrieval of papers. Our previous analyses used the semantic representation of papers in different semantic models with the purpose of creating visual graphs based on the semantic relatedness links between the abstracts. The current analysis takes a step forward by proposing a model that can suggest which papers are of the highest relevance, share similar concepts, and are semantically related with the initial query. Our study is performed using paper abstracts in the field of information technology extracted from the Web of Science citation index. The research includes a use case and its corresponding results by using interactive and exploratory network graph representations.

In recent years, the volume of research papers has become enormous. Therefore, it is difficult for researchers to select their required papers. To lighten this problem, a method of describing abstracts called "Structured Abstract" is used in the fields of physiology and medicine. This paper proposes a method that extracts sentences matching with each heading of Structured Abstract. Specifically, SciBERT and BioBERT, which are specialized in the science and technology fields, are used in the method. The effectiveness of the proposed method is evaluated by comparing automatically generated abstracts and human-written abstracts.

As the spread of science and the number of researchers working in academic fields increase, there is also a considerable increase in the number of academic studies. Researchers always follow new works published for keeping their knowledge up to date. However, with many different sources and thousands of academic publications published every day, academics are not always able to find publications about their subjects. Today, almost all of online academic databases employ a recommendation module which only considers the studies similar to the paper that the user looked at. However, a recommendation system based on the information of a single article is often not enough. In this study, the proposed method recommends by considering user's publications, user’s co-authors and co-authors’ papers. Therefore, meta-data of the articles published by the researcher in the past are scored on a time-base basis with the method we propose. With the help of the sum of scores, there is a score of the user profile in the subject matter. It aims to find the closest studies to the profile of the user by searching with the method propsoed in the data pool which we created from the exact contents of hundreds of thousands of academic works. In the proposed method, TF-IDF is used from frequency-based similarity analysis methods. In the evaluation phase, the performance of the proposed method was examined. The success test of the method was measured by several different methods. These are to be evaluated by presenting them to real users and the other is to compare with existing data. The results are very promising and demonstrate that the method can produce accurate and quality results.

Researchers and scientists read articles to improve their studies. Researchers spend a lot of time trying to find the suitable article from among a large number of publications. The purpose of the article recommendation system is to shorten the working time and to present the research articles to the researchers. Classic article recommendation systems do not consider the user’s information. They show the same results in the same sort for each researcher.

E-commerce customers and general Web users should not believe that the products suggested by Recommender systems or results displayed by Web search engines are either complete or relevant to their search aspirations. The economic priority of Web related businesses requires a higher rank on Web snippets or product suggestions in order to receive additional customers; furthermore, Web search engines and recommender systems revenue is obtained from advertisements and pay-per-click. This survey paper presents a review of Web Search Engines, Ranking Algorithms, Citation Analysis and Recommender Systems. In addition, Neural networks and Deep Learning are also analyzed including their use in learning relevance and ranking. Finally, this survey paper also introduces the Random Neural Network with its practical applications.

Tremendous academic information causes serious information overload problems while supporting scientific research. Scientific paper and citation recommendation systems have been developed to relieve this problem and work as a filter to furnish only relevant papers to researchers. Although previous studies have made comparative progress, this problem is still challenging because current paper recommendation systems rely on heterogeneous and multi-sourced features, thereby requiring a unified learning representation to cover different types and modalities of information. Additionally, the implicit influence of scholars’ previous preferences of writing and citing on his/her new manuscript has not been well considered in the previous studies. Facing the issue from these two aspects, in this paper, a heterogeneous knowledge embedding-based attentive RNN model is proposed to recommend scientific paper citations. First, the preparation of features consists of two parts: (1) building a unified learning representation of structural entities and relations for recommending paper citations; and (2) defining and constructing a bibliographic network comprising five types of entities and five relations. The bibliographic network enables learning a unified representation so that all graphical entities and relations can be vectorized using TransD. To establish textual representations, the PV-DM model is utilized to generate numeric features for the title of each paper. Second, by combining structural and textual representations focusing on the “author-text query” scenario, an attentive bidirectional RNN is constructed to recommend paper and citation based on an user’s identity with a length-limited inquiry to capture the scholars’ previous writing and citing preferences, thereby reducing recommendation error. Through the DBLP dataset, our experiment results show the feasibility and effectiveness of our method, both in terms of the number as well as the quality of the first few recommended items. In specific, compared with existing models, our model has improved MRR and NDCG by approximately 4.8% and 2.4%, respectively.

Researchers face millions of research papers on various digital libraries. Therefore, finding relevant research work that meets the preferences of a researcher is a challenging task. Hence, different paper recommendation models have been proposed to address this issue. However, these models lack in exploiting prominent information factors, namely: papers’ citations proximity, authors’ information, papers’ topical relevance, venues’ information, researchers’ preference dynamics, and labels information to produce quality recommendations. Additionally, these models encounter problems such as cold start papers and data sparsity. To overcome these problems, this paper presents a weighted probabilistic paper recommendation model termed as PR-HNE, which jointly learns researchers’ and papers’ dynamics by encoding information from six information networks into a joint latent space. Specifically, it captures papers’ citation proximity, authors’ collaboration proximity, venues’ information, labeled information, and topical relevance to generate personalized paper recommendations. Compared to state-of-the-art models, the results generated by PR-HNE over publicly available datasets prove 4% and 6% improvement in Mean Average Precision (MAP) and Mean Reciprocal Rank (MRR) metrics, respectively. Further, in the cold-start papers problem, the proposed model produced 8% better recall score than its counterparts.

To compute the similarity of scientific papers, text-based similarity measures, link-based similarity measures, and hybrid methods can be applied. The text-based and link-based similarity measures take into account only a single aspect of scientific papers, content or citations, respectively. The hybrid methods consider both content and citations; however, they do not carefully consider the relation between the content of a pair of papers involved in a citation relationship. In this paper, we propose a novel method, SimCC (similarity based on content and citations), that considers both aspects, content and citations, to compute the similarity of scientific papers. Unlike previous methods, SimCC effectively reflects both content and authority of scientific papers simultaneously in similarity computation by applying a new RA (relevance and authority) weighting scheme. Also, we propose an RA+R weighting scheme to consider the recency of papers and an RA+E weighting scheme to take into account the author expertise of papers in similarity computation. The effectiveness of our proposed method is demonstrated by extensive experiments on a real-world dataset of scientific papers. The results show that our method achieves more than 100% improvement in accuracy in comparison with previous methods.

This paper proposes a system called CFP Manager specialized on IT field and designed to ease the process of searching conference suitable to one's need. At present, the handling of CFP faces two problems: for emails, the huge quantity of CFP received can be easily skimmed through. For websites, the reviewing of some of the main CFP aggregators available online points out the lack of usable criteria. This system proposes to answer to these problems via its architecture consisting of three components: firstly an Information Extraction module extracting relevant information (as date, location, etc...) from CFP using rule based text mining algorithm. The second component enriches the now extracted data with external one from ontology models. Finally the last one displays the said data and allows the end user to perform complex queries on the CFP dataset and thus allow him to only access to CFP suitable for him. In order to validate the authors' proposal, they eventually process the well-known precision / recall metric on our information extraction component with an average of 0.95 for precision and 0.91 for recall on three different 100 CFP dataset. This paper finally discusses the validity of our approach by confronting our system for different queries with two systems already available online (WikiCFP and IEEE Conference Search) and basic text searching approach standing for searching in an email box. On a 100 CFP dataset with the wide variety of usable data and the possibility to perform complex queries we surpass basic text searching method and WikiCFP by not returning the false positive usually returned by them and find a result close to the IEEE system.

International audience; The work presented in this paper deals with the use of ontologies to compare scientific texts. It particularly deals with scientific papers, specifically their abstracts, short texts that are relatively well structured and normally provide enough knowledge to allow a community of readers to assess the content of the associated scientific papers. The problem is, therefore, to determine how to assess the semantic proximity/similarity of two papers by examining their respective abstracts. Given that a domain ontology provides a useful way to represent knowledge relative to a given domain, this work considers ontologies relative to scientific domains. Our process begins by defining the relevant domain for an abstract through an automatic classification that makes it possible to associate this abstract to its relevant scientific domain, chosen from several candidate domains. The content of an abstract is represented in the form of a conceptual graph which is enriched to construct its semantic perimeter. As presented below, this notion of semantic perimeter usefully allows us to assess the similarity between the texts by matching their graphs. Detecting plagiarism is the main application field addressed in this paper, among the many possible application fields of our approach.

The paper addresses the task of information extraction from scientific literature with machine learning methods. In particular, the tasks of definition and result extraction from scientific publications in Russian are considered. We note that annotation of scientific texts for creation of training dataset is very labor insensitive and expensive process. To tackle this problem, we propose methods and tools based on active learning. We describe and evaluate a novel adaptive density-weighted sampling (ADWeS) meta-strategy for active learning. The experiments demonstrate that active learning can be a very efficient technique for scientific text mining, and the proposed meta-strategy can be beneficial for corpus annotation with strongly skewed class distribution. We also investigate informative task-independent features for information extraction from scientific texts and present an openly available tool for corpus annotation, which is equipped with ADWeS and compatible with well-known sampling strategies.

With the increasing number of scientific publications, research paper recommendation has become increasingly important for scientists. Most researchers rely on keyword-based search or following citations in other papers, in order to find relevant research articles. And usually they spend a lot of time without getting satisfactory results. This study aims to propose a personalized research paper recommendation system, that facilitate this task by recommending papers based on users' explicit and implicit feedback. The users will be allowed to explicitly specify the papers of interest. In addition, user activities (e.g., viewing abstracts or full-texts) will be analyzed in order to enhance users' profiles. Most of the current research paper recommendation and information retrieval systems use the classical bag-of-words methods, which don't consider the context of the words and the semantic similarity between the articles. This study will use Recurrent Neural Networks (RNNs) to discover continuous and latent semantic features of the papers, in order to improve the recommendation quality. The proposed approach utilizes PubMed so far, since it is frequently used by physicians and scientists, but it can easily incorporate other datasets in the future.

Researchers often summarize their work in the form of scientific posters. Posters provide a coherent and efficient way to convey core ideas expressed in scientific papers. Generating a good scientific poster, however, is a complex and time consuming cognitive task, since such posters need to be readable, informative, and visually aesthetic. In this paper, for the first time, we study the challenging problem of learning to generate posters from scientific papers. To this end, a data-driven framework, that utilizes graphical models, is proposed. Specifically, given content to display, the key elements of a good poster, including attributes of each panel and arrangements of graphical elements are learned and inferred from data. During the inference stage, an MAP inference framework is employed to incorporate some design principles. In order to bridge the gap between panel attributes and the composition within each panel, we also propose a recursive page splitting algorithm to generate the panel layout for a poster. To learn and validate our model, we collect and release a new benchmark dataset, called NJU-Fudan Paper-Poster dataset, which consists of scientific papers and corresponding posters with exhaustively labelled panels and attributes. Qualitative and quantitative results indicate the effectiveness of our approach., 10 pages, submission to IEEE TPAMI. arXiv admin note: text overlap with arXiv:1604.01219

In this paper we proposed a novel Deep Learning approach to realize a Word Embeddings (WEs) similarity based search tool, considering the medical literature as case study. Using the compositional properties of the WEs we defined a methodology to aggregate the information coming from each word to obtain a vector corresponding to the abstracts of each PubMed article. Through this paradigm it is possible to capture the semantic content of the papers and, consequently, to evaluate and rank the similarity among them. The preliminary results with the proposed approach are obtained analysing a subset of the whole the PubMed collection. The results correctness has been verified by human domain experts, showing that the methodology is promising.

Rhetorical sentence classification is an interesting approach for making extractive summaries but this technique still needs to be developed because the performance of automatic rhetorical sentence classification is still poor. Rhetorical sentences are sentences that contain rhetorical words or phrases. Rhetorical sentences not only appear in the contents of a paper but also in the title. In this study, features related to section class and title class that have been proposed in a previous research were further developed. Our method uses different techniques to reach automatic section class extraction for which we introduce new, format-based features. Furthermore, we propose automatic rhetoric phrase extraction from the title. The corpus we used was a collection of technical-experimental scientific papers. Our method uses the Support Vector Machine (SVM) algorithm and the Naive Bayesian algorithm for classification. The four categories used were: Problem, Method, Data, and Result . It was hypothesized that these features would be able to improve classification accuracy compared to previous methods. The F-measure for these categories reached up to 14%.

This research was supported by the NRF grant (NRF-2014R1A2A1A10054151) and the ITRC support program (IITP-2015-H8501-15-1013).

Extraction of structured information from biomedical literature is a complex and challenging problem due to the complexity of biomedical domain and lack of appropriate natural language processing (NLP) techniques. High quality domain ontologies model both data and metadata information at a fine level of granularity, which can be effectively used to accurately extract structured information from biomedical text. Extraction of provenance metadata, which describes the history or source of information, from published articles is an important task to support scientific reproducibility. Reproducibility of results reported by previous research studies is a foundational component of scientific advancement. This is highlighted by the recent initiative by the US National Institutes of Health called “Principles of Rigor and Reproducibility”. In this paper, we describe an effective approach to extract provenance metadata from published biomedical research literature using an ontology-enabled NLP platform as part of the Provenance for Clinical and Healthcare Research (ProvCaRe). The ProvCaRe-NLP tool extends the clinical Text Analysis and Knowledge Extraction System (cTAKES) platform using both provenance and biomedical domain ontologies. We demonstrate the effectiveness of ProvCaRe-NLP tool using a corpus of 20 peer-reviewed publications. The results of our evaluation demonstrate that the ProvCaRe-NLP tool has significantly higher recall in extracting provenance metadata as compared to existing NLP pipelines such as MetaMap.

This paper describes our TTI-COIN system that participated in SemEval-2017 Task 10. We investigated appropriate embeddings to adapt a neural end-to-end entity and relation extraction system LSTM-ER to this task. We participated in the full task setting of the entity segmentation, entity classification and relation classification (scenario 1) and the setting of relation classification only (scenario 3). The system was directly applied to the scenario 1 without modifying the codes thanks to its generality and flexibility. Our evaluation results show that the choice of appropriate pre-trained embeddings affected the performance significantly. With the best embeddings, our system was ranked third in the scenario 1 with the micro F1 score of 0.38. We also confirm that our system can produce the micro F1 score of 0.48 for the scenario 3 on the test data, and this score is close to the score of the 3rd ranked system in the task.

This paper attempts to use more than a million academic papers available in institutional repositories in Japan, and their possibilities as a language resource are verified. The verification is performed with (1) word length distribution of extracted noun words from collected academic papers in PDF and (2) word analogy tasks by using word2vec. The results suggest that words are separated at meaningless positions without pre-processing where full text are extracted from academic papers in PDF, and related words can be analogized by similarity calculation among words on a feature space. We conclude that academic papers in institutional repositories in Japan have possibilities as a language resource, although preprocessing for distributed files should be addressed.

This paper proposes a new method for query expansion based on bidirectional extraction of phrases as word n-grams from research paper titles. The proposed method aims to extract information relevant to users’ needs and interests and thus to provide a useful system for technical paper retrieval. The outcome of proposed method are the trigrams as phrases that can be used for query expansion. First, word trigrams are extracted from research paper titles. Second, a co-occurrence graph of the extracted trigrams is constructed. To construct the co-occurrence graph, the direction of edges is considered in two ways: forward and reverse. In the forward and reverse co-occurrence graphs, the trigrams point to other trigrams appearing after and before them in a paper title,respectively. Third, Jaccard similarity is computed between trigrams as the weight of the graph edge. Fourth, the weighted version of PageRank is applied. Consequently, the following two types of phrases can be obtained as the trigrams associated with the higher PageRank scores. The trigrams of the one type, which are obtained from the forward co-occurrence graph, can form a more specific query when users add a technical word or words before them. Those of the other type, obtained from the reverse co-occurrence graph, can form a more specific query when users add a technical word or words after them. The extraction of phrases is evaluated as additional features in the paper title classification task using SVM. The experimental results show that the classification accuracy is improved than the accuracy achieved when the standard TF-IDF text features are only used. Moreover, the trigrams extracted by the proposed method can be utilized to expand query words in research paper retrieval.

Sentence ranking is one of the most important research issues in text analysis. It can be used in text summarization and information retrieval. Graph-based methods are a common way of ranking and extracting sentences. In graph based methods, sentences are nodes of graph and edges are built based on the sentence similarities or on sentence co-occurrence relationship. PageRank style algorithms can be applied to get sentence ranks. In this paper, we focus on how to rank sentences in a single scientific paper. A scientific literature has more structural information than general texts and this structural information has not been fully explored yet in graph based ranking models. We investigated several different methods that used the is-part-of link on paragraph and section and similar link and co-occurrence link to construct a heterogeneous graph for ranking sentences. We conducted experiments on these methods to compare the results on sentence ranking. It shows that structural information can help identify more representative sentences.

International audience; Each domain, along with its knowledge base, changes over time and every timeframe is centered on specific topics that emerge from different ongoing research projects. As searching for relevant resources is a time-consuming process, the automatic extraction of the most important and relevant articles from a domain becomes essential in supporting researchers in their day-today activities. The proposed analysis extends other previous researches focused on extracting co-citations between the papers, with the purpose of comparing their overall importance within the domain from a semantic perspective. Our method focuses on the semantic analysis of paper abstracts by using Natural Language Processing (NLP) techniques such as Latent Semantic Analysis, Latent Dirichlet Allocation or specific ontology distances, i.e., WordNet. Moreover, the defined mechanisms are enforced on two different subdomains from the corpora generated around the keywords " e-learning " and " computer ". Graph visual representations are used to highlight the keywords of each subdomain, links among concepts and between articles, as well as specific document similarity views, or scores reflecting the keyword-abstract overlaps. In the end, conclusions and future improvements are presented, emphasizing nevertheless the key elements of our research support framework.

This paper proposes a method for exploring technical phrase frames by extracting word n-grams that match our information needs and interests from research paper titles. Technical phrase frames, the outcome of our method, are phrases with wildcards that may be substituted for any technical term. Our method, first of all, extracts word trigrams from research paper titles and constructs a co-occurrence graph of the trigrams. Even by simply applying Page Rank algorithm to the co-occurrence graph, we obtain the trigrams that can be regarded as technical key phrases at the higher ranks in terms of Page Rank score. In contrast, our method assigns weights to the edges of the co-occurrence graph based on Jaccard similarity between trigrams and then apply weighted Page Rank algorithm. Consequently, we obtain widely different but more interesting results. While the top-ranked trigrams obtained by unweighted Page Rank have just a self-contained meaning, those obtained by our method are technical phrase frames, i.e., A word sequence that forms a complete technical phrase only after putting a technical word (or words) before or/and after it. We claim that our method is a useful tool for discovering important phrase logical patterns, which can expand query keywords for improving information retrieval performance and can also work as candidate phrasings in technical writing to make our research papers attractive., 29th IEEE International Conference on Advanced Information Networking and Applications Workshops, WAINA 2015; Gwangju; South Korea; 25 March 2015 through 27 March 2015, 2015 IEEE 29th International Conference on Advanced Information Networking and Applications Workshops (WAINA), 7096236; 2015

We study the problem of extracting terms from research papers, which is an important step towards building knowledge graphs in research domain. Existing terminology extraction approaches are mostly domain dependent. They use domain specific linguistic rules, supervised machine learning techniques or a combination of the two to extract the terms. Using domain knowledge requires much human effort, e.g., manually composing a set of linguistic rules or labeling a large corpus, and hence limits the applicability of the existing approaches. To overcome this limitation, we propose a new terminology extraction approach that makes use of no knowledge from any specific domain. In particular, we use the title words and the keywords in research papers as the seeding terms and word2vec to identify similar terms from an open-domain corpus as the candidate terms, which are then filtered by checking their occurrence in the research papers. We repeat this process using the newly found terms until no new candidate term can be found. We conduct extensive experiments on the proposed approach. The results show that our approach can extract the terms effectively, while being domain independent.

An effective way to reveal the contents of research papers is assigning a group of terms against a controlled vocabulary. To the best of our knowledge, a variety of automatic indexing techniques have been studied to enhance the effectiveness and the efficiency. However, the current approaches depended on the content of a research paper, such as title, abstract, etc., which suffering from limitations on the automatic indexing performance. In this paper, we propose a new approach of automatic indexing for single research paper with its references based on Genetic Algorithm (GA). The extensive experiments on four subjects show the effectiveness of the proposed approach.

The problem to retrieve most relevant research papers for a given academic is studied. Existing solutions cannot adequately address the recommendation of new papers due to their lack of history information, the so-called cold start problem. Using the graphical model built from citation information between a new paper pi and published papers, toward this challenge, we propose a novel approach based on a probabilistic inference algorithm, the Belief Propagation (BP), to predict the likelihood of pi's relevance to a target academic. Compared to item-based collaborative filtering method using a DBLP data set, the empirical validation shows an improvement in accuracy up to 26% in F1 score.

A considerable amount of research is being conducted by many people (researchers, graduate students, professors etc) everyday. Finding information about a specific topic is one of the most time consuming activities of those people. People doing research have to search, read and analyze multiple research papers, e-books and other documents and then determine what they contain and discover knowledge from them. Many available resources are in the form of unstructured text format of long text pages which require long time to read and analyze. In this paper we propose a two-stage method for scientific paper analysis. The method uses information extraction to extract the main idea key sentences (mainly needed by the most readers) from the paper and the extracted paper’s information is then organized in a structured format and grouped in different clusters according to their topics using a multi-word based clustering method. The proposed method combines different features in paper’s topics extraction and uses multi-word matching feature in selection of initial centroids for clustering. The proposed method can help readers to access and analyze multiple research papers documents timely and efficiently. Conducted experiments show the effectiveness and usefulness of our proposed approach.

Understanding the structure of scientific discourse is of paramount importance for the development of appropriate Natural Language Processing tools able to extract and summarize information from research articles. In this paper we present an annotated corpus of scientific discourse in the domain of Computer Graphics. We describe the way we built our corpus by designing an annotation schema and relying on three annotators for manually classifying all sentences into the defined categories. Our corpus constitutes a semantically rich resource for scientific text mining. In this respect, we also present the results of our initial experiments of automatic classification of sentences into the 5 main categories in our corpus.

One of some ways to summarize scientific papers is by employing rhetorical structure of sentences. Determining rhetorical sentence itself passes through the process of text categorization. In order to get good performance, some works in text categorization have been done by employing semantic similarity words. Therefore, this paper aims to present the rhetorical sentence categorization from scientific paper by using selected features, added previous label, and Word2Vec to capture semantic similarity words. Then, this paper shows the result of employing resampling for balancing the existing instances per class and combining resampling and Word2Vec representation itself. Every experiment is tested in two classifiers, namely IBk and J48 tree. It shows that the use of previous label, Word2Vec (Skip-Gram), and resampling improves performance. After doing all the experiments in the 10-fold cross-validation, the highest performance of F-measure is achieved 84.97% by combining Word2Vec (Skip-Gram), all features, and resampling.

Finding research literature pertaining to a task at hand is one of the essential tasks that scientists face on daily basis. Standard information retrieval techniques allow to quickly obtain a vast number of potentially relevant documents. Unfortunately, the search results then require significant effort for manual inspection, where we would rather select relevant publications based on more fine-grained, semantically rich queries involving a publication's contributions, methods, or application domains. We argue that a novel combination of three distinct methods can significantly advance this vision: (i) Natural Language Processing (NLP) for Rhetorical Entity (RE) detection; (ii) Named Entity (NE) recognition based on the Linked Open Data (LOD) cloud; and (iii) automatic generation of RDF triples for both NEs and REs using semantic web ontologies to interconnect them. Combined in a single workflow, these techniques allow us to automatically construct a knowledge base that facilitates numerous advanced use cases for managing scientific documents.

The word ‘significant’ is widely used in scientific papers. A PubMed search for ‘significant’ in June 2014 yielded more than two million results! But the word ‘significant’ has several meanings, and so is often misunderstood. There is a simple way to avoid this confusion: Avoid the word ‘significant’ and use alternative wording. (Published: 25 September 2014) Citation: Advances in Regenerative Biology 2014, 1 : 25155 - http://dx.doi.org/10.3402/arb.v1.25155

The key objective of image steganography is to hide the secret data in different embedding medium called as carriers. These carriers can be images, video files, audio files and text files. This paper discussed recent steganographic techniques for image files. Imperceptibility, embedding capacity and robustness are the key issues of image steganography. These recent techniques are analysed the key issues of steganography and discussed the merits and demerits of these techniques, to identify the suitability for data hiding method which will provide guidance to the people working in this field. Steganalysis was also discussed separately.

BackgroundWith the development of biomedicine, the number of biomedical documents has increased rapidly bringing a great challenge for researchers trying to retrieve the information they need. Information retrieval aims to meet this challenge by searching relevant documents from abundant documents based on the given query. However, sometimes the relevance of search results needs to be evaluated from multiple aspects in specific retrieval tasks, thereby increasing the difficulty of biomedical information retrieval.ObjectiveThis study aimed to find a more systematic method for retrieving relevant scientific literature for a given patient.MethodsIn the initial retrieval stage, we supplemented query terms through query expansion strategies and applied query boosting to obtain an initial ranking list of relevant documents. In the re-ranking phase, we employed a text classification model and relevance matching model to evaluate documents from different dimensions and then combined the outputs through logistic regression to re-rank all the documents from the initial ranking list.ResultsThe proposed ensemble method contributed to the improvement of biomedical retrieval performance. Compared with the existing deep learning–based methods, experimental results showed that our method achieved state-of-the-art performance on the data collection provided by the Text Retrieval Conference 2019 Precision Medicine Track.ConclusionsIn this paper, we proposed a novel ensemble method based on deep learning. As shown in the experiments, the strategies we used in the initial retrieval phase such as query expansion and query boosting are effective. The application of the text classification model and relevance matching model better captured semantic context information and improved retrieval performance.