Go Eun Heo

PKDE4J: Entity and relation extraction for public knowledge discovery

Due to an enormous number of scientific publications that cannot be handled manually, there is a rising interest in text-mining techniques for automated information extraction, especially in the biomedical field. Such techniques provide effective means of information search, knowledge discovery, and hypothesis generation. Most previous studies have primarily focused on the design and performance improvement of either named entity recognition or relation extraction. In this paper, we present PKDE4J, a comprehensive text-mining system that integrates dictionary-based entity extraction and rule-based relation extraction in a highly flexible and extensible framework. Starting with the Stanford CoreNLP, we developed the system to cope with multiple types of entities and relations. The system also has fairly good performance in terms of accuracy as well as the ability to configure text-processing components. We demonstrate its competitive performance by evaluating it on many corpora and found that it surpasses existing systems with average F-measures of 85% for entity extraction and 81% for relation extraction.

Topic-Network based Topic Shift Detection on Twitter

This study identified topic shifts and patterns over time by analyzing an enormous amount of Twitter data whose characteristics are high accessibility and briefness. First, we extracted keywords for a certain product and used them for representing the topic network allows for intuitive understanding of keywords associated with topics by nodes and edges by co-word analysis. We conducted temporal analysis of term co-occurrence as well as topic modeling to examine the results of network analysis. In addition, the results of comparing topic shifts on Twitter with the corresponding retrieval results from newspapers confirm that Twitter makes immediate responses to news media and spreads the negative issues out quickly. Our findings may suggest that companies utilize the proposed technique to identify public`s negative opinions as quickly as possible and to apply for the timely decision making and effective responses to their customers.

Analyzing topic evolution in bioinformatics: investigation of dynamics of the field with conference data in DBLP

In this paper we analyze topic evolution over time within bioinformatics to uncover the underlying dynamics of that field, focusing on the recent developments in the 2000s. We select 33 bioinformatics related conferences indexed in DBLP from 2000 to 2011. The major reason for choosing DBLP as the data source instead of PubMed is that DBLP retains most bioinformatics related conferences, and to study dynamics of the field, conference papers are more suitable than journal papers. We divide a period of a dozen years into four periods: period 1 (2000–2002), period 2 (2003–2005), period 3 (2006–2008) and period 4 (2009–2011). To conduct topic evolution analysis, we employ three major procedures, and for each procedure, we develop the following novel technique: the Markov Random Field-based topic clustering, automatic cluster labeling, and topic similarity based on Within-Period Cluster Similarity and Between-Period Cluster Similarity. The experimental results show that there are distinct topic transition patterns between different time periods. From period 1 to period 3, new topics seem to have emerged and expanded, whereas from period 3 to period 4, topics are merged and display more rigorous interaction with each other. This trend is confirmed by the collaboration pattern over time.

Identifying the landscape of Alzheimer's disease research with network and content analysis

Alzheimer’s disease (AD) is one of degenerative brain diseases, whose cause is hard to be diagnosed accurately. As the number of AD patients has increased, researchers have strived to understand the disease and develop its treatment, such as medical experiments and literature analysis. In the area of literature analysis, several traditional studies analyzed the literature at the macro level like author, journal, and institution. However, analysis of the literature both at the macro level and micro level will allow for better recognizing the AD research field. Therefore, in this study we adopt a more comprehensive approach to analyze the AD literature, which consists of productivity analysis (year, journal/proceeding, author, and Medical Subject Heading terms), network analysis (co-occurrence frequency, centrality, and community) and content analysis. To this end, we collect metadata of 96,081 articles retrieved from PubMed. We specifically perform the concept graph-based network analysis applying the five centrality measures after mapping the semantic relationship between the UMLS concepts from the AD literature. We also analyze the time-series topical trend using the Dirichlet multinomial regression topic modeling technique. The results indicate that the year 2013 is the most productive year and Journal of Alzheimer’s Disease the most productive journal. In discovery of the core biological entities and their relationships resided in the AD related PubMed literature, the relationship with glycogen storage disease is founded most frequently mentioned. In addition, we analyze 16 main topics of the AD literature and find a noticeable increasing trend in the topic of transgenic mouse.

Establishing a baseline for literature mining human genetic variants and their relationships to disease cohorts

BackgroundThe Variome corpus, a small collection of published articles about inherited colorectal cancer, includes annotations of 11 entity types and 13 relation types related to the curation of the relationship between genetic variation and disease. Due to the richness of these annotations, the corpus provides a good testbed for evaluation of biomedical literature information extraction systems.MethodsIn this paper, we focus on assessing performance on extracting the relations in the corpus, using gold standard entities as a starting point, to establish a baseline for extraction of relations important for extraction of genetic variant information from the literature. We test the application of the Public Knowledge Discovery Engine for Java (PKDE4J) system, a natural language processing system designed for information extraction of entities and relations in text, on the relation extraction task using this corpus.ResultsFor the relations which are attested at least 100 times in the Variome corpus, we realise a performance ranging from 0.78–0.84 Precision-weighted F-score, depending on the relation. We find that the PKDE4J system adapted straightforwardly to the range of relation types represented in the corpus; some extensions to the original methodology were required to adapt to the multi-relational classification context. The results are competitive with state-of-the-art relation extraction performance on more heavily studied corpora, although the analysis shows that the Recall of a co-occurrence baseline outweighs the benefit of improved Precision for many relations, indicating the value of simple semantic constraints on relations.ConclusionsThis work represents the first attempt to apply relation extraction methods to the Variome corpus. The results demonstrate that automated methods have good potential to structure the information expressed in the published literature related to genetic variants, connecting mutations to genes, diseases, and patient cohorts. Further development of such approaches will facilitate more efficient biocuration of genetic variant information into structured databases, leveraging the knowledge embedded in the vast publication literature.

Grounded Feature Selection for Biomedical Relation Extraction by the Combinative Approach

Relation extraction is an important task in biomedical areas such as protein-protein interaction, gene-disease interactions, and drug-disease interactions. In recent years, it has been widely researched to automatically extract biomedical relations in a vest amount of biomedical text data. In this paper, we propose a hybrid approach to extracting relations based on a rule-based approach feature set. We then use different classification algorithms such as SVM, Naïve Bayes, and Decision Tree classifiers for relation classification. The rationale for adopting shallow parsing and other NLP techniques to extract relations is two-folds: simplicity and robustness. We select seven features with the rule-based shallow parsing technique and evaluate the performance with four different PPI public corpora. Our experimental results show the stable performance in F-measure even with the relatively fewer features.

SemPathFinder: Semantic path analysis for discovering publicly unknown knowledge

The enormous amount of biomedicines natural-language texts creates a daunting challenge to discover novel and interesting patterns embedded in the text corpora that help biomedical professionals find new drugs and treatments. These patterns constitute entities such as genes, compounds, treatments, and side effects and their associations that spread across publications in different biomedical specialties. This paper proposes SemPathFinder to discover previously unknown relations in biomedical text. SemPathFinder overcomes the problems of Swansons ABC model by using semantic path analysis to tell a story about plausible connections between biological terms. Storytelling-based semantic path analysis can be viewed as relation navigation for bio-entities that are semantically close to each other, and reveals insight into how a series of entity pairs is organized, and how it can be harnessed to explain seemingly unrelated connections. We apply SemPathFinder for two well-known use cases of Swansons ABC model, and the experimental results show that SemPathFinder detects all intermediate terms except for one and also infers several interesting new hypotheses.

Examining the Intellectual Structure of a Medical Informatics Journal with Author Co-citation Analysis and Co-word Analysis

ABSTRACT Due to the development of science and technology, the convergen ce of various disciplines has been fostered. Accordingly, interdisciplinary studies have increasingly been expanded by integrating knowledge and methodology from different discipl ines. The primary focus of biblimetric methods is on investigating the intellectual str ucture a field, and analysis of the characterization of interdisciplinary studies is overlooked. In this study, we aim to identify the intellectual structure of the field of medical informatics through author co-citation analysis and co-word analysis by the representative journal “IEEE ENG MED BIOL.” In addition, we examine authors and MeSH Terms of top three representative journals for further analysis of the field. We examine the intellectual structure of the medical informatics field by author and word clusters to identify the n etwork structure of medical informatics disciplines.키워드: 의료정보학, 저자동시인용 분석, 동시출현단어 분석, 네트워크 분석 medical informatics, author co-citation analysis, co-word analy sis, network analysis

Trajectory analysis of drug-research trends in pancreatic cancer on PubMed and ClinicalTrials.gov

Increasing interest in developing treatments for pancreatic cancer has led to a surge in publications in the field. Analyses of drug-research trends are needed to minimize risk in anti-cancer drug development. Here, we analyzed publications on anti-cancer drugs extracted from PubMed records and ClinicalTrials datasets. We conducted a drug cluster analysis by proposing the entity Dirichlet Multinomial Regression (eDMR) technique and in-depth network analysis of drug cluster and target proteins. The results show two distinct research clusters in both the ClinicalTrials dataset and the PubMed records. Specifically, various targets associated with anti-cancer drugs are investigated in new drug testing while the diverse chemicals are studied together with a standard therapeutic agent in the academic literature. In addition, our study confirms that drug research published in PubMed is preceded by clinical trials. Although we only evaluate drugs for pancreatic cancer in the present study, our method can be applied to drug-research trends of other diseases.

Analyzing the field of bioinformatics with the multi-faceted topic modeling technique

BackgroundBioinformatics is an interdisciplinary field at the intersection of molecular biology and computing technology. To characterize the field as convergent domain, researchers have used bibliometrics, augmented with text-mining techniques for content analysis. In previous studies, Latent Dirichlet Allocation (LDA) was the most representative topic modeling technique for identifying topic structure of subject areas. However, as opposed to revealing the topic structure in relation to metadata such as authors, publication date, and journals, LDA only displays the simple topic structure.MethodsIn this paper, we adopt the Tang et al.’s Author-Conference-Topic (ACT) model to study the field of bioinformatics from the perspective of keyphrases, authors, and journals. The ACT model is capable of incorporating the paper, author, and conference into the topic distribution simultaneously. To obtain more meaningful results, we use journals and keyphrases instead of conferences and bag-of-words.. For analysis, we use PubMed to collected forty-six bioinformatics journals from the MEDLINE database. We conducted time series topic analysis over four periods from 1996 to 2015 to further examine the interdisciplinary nature of bioinformatics.ResultsWe analyze the ACT Model results in each period. Additionally, for further integrated analysis, we conduct a time series analysis among the top-ranked keyphrases, journals, and authors according to their frequency. We also examine the patterns in the top journals by simultaneously identifying the topical probability in each period, as well as the top authors and keyphrases. The results indicate that in recent years diversified topics have become more prevalent and convergent topics have become more clearly represented.ConclusionThe results of our analysis implies that overtime the field of bioinformatics becomes more interdisciplinary where there is a steady increase in peripheral fields such as conceptual, mathematical, and system biology. These results are confirmed by integrated analysis of topic distribution as well as top ranked keyphrases, authors, and journals.