A Text Mining Protocol for Predicting Drug–Drug Interaction and Adverse Drug Reactions from PubMed Articles

Shukkoor, Mohamed Saleem Abdul; Raja, Kalpana; Baharuldin, Mohamad Taufik Hidayat

doi:10.1007/978-1-0716-2305-3_13

Cited by 4 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering the abundance of scientific articles that are published every day ( Uddin, Khan & Baur, 2015 ), keeping up with the latest research is becoming a significant challenge for researchers in many fields. This is at least partially due to the fact that we are still holding on to an archaic paradigm of scientific publishing: the canonical way to publish scientific results is by writing them up in long English texts called articles, which are in the best case easy to read by human experts but remain mostly inaccessible to automated approaches (except on a very superficial level with text mining approaches) ( Bhargava et al, 2017 ; Westergaard et al, 2018 ; Shukkoor, Raja & Baharuldin, 2022 ). These articles then undergo peer reviewing, which is typically done in a way that is secretive and not standardized, with the effect that the reviewing process may lack transparency and that valuable comments from the reviewers cannot be reused or built upon.…”

Section: Introductionmentioning

confidence: 99%

Nanopublication-based semantic publishing and reviewing: a field study with formalization papers

Bucur

Kuhn

Ceolin

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

With the rapidly increasing amount of scientific literature, it is getting continuously more difficult for researchers in different disciplines to keep up-to-date with the recent findings in their field of study. Processing scientific articles in an automated fashion has been proposed as a solution to this problem, but the accuracy of such processing remains very poor for extraction tasks beyond the most basic ones (like locating and identifying entities and simple classification based on predefined categories). Few approaches have tried to change how we publish scientific results in the first place, such as by making articles machine-interpretable by expressing them with formal semantics from the start. In the work presented here, we propose a first step in this direction by setting out to demonstrate that we can formally publish high-level scientific claims in formal logic, and publish the results in a special issue of an existing journal. We use the concept and technology of nanopublications for this endeavor, and represent not just the submissions and final papers in this RDF-based format, but also the whole process in between, including reviews, responses, and decisions. We do this by performing a field study with what we call formalization papers, which contribute a novel formalization of a previously published claim. We received 15 submissions from 18 authors, who then went through the whole publication process leading to the publication of their contributions in the special issue. Our evaluation shows the technical and practical feasibility of our approach. The participating authors mostly showed high levels of interest and confidence, and mostly experienced the process as not very difficult, despite the technical nature of the current user interfaces. We believe that these results indicate that it is possible to publish scientific results from different fields with machine-interpretable semantics from the start, which in turn opens countless possibilities to radically improve in the future the effectiveness and efficiency of the scientific endeavor as a whole.

show abstract

Section: Introductionmentioning

confidence: 99%

Nanopublication-based semantic publishing and reviewing: a field study with formalization papers

Bucur

Kuhn

Ceolin

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Considering the abundance of scientific articles that are published every day [157], keeping up with the latest research is becoming a significant challenge for researchers in many fields. This is at least partially due to the fact that we are still holding on to an archaic paradigm of scientific publishing: the canonical way to publish scientific results is by writing them up in long English texts called articles, which are in the best case easy to read by human experts but remain mostly inaccessible to automated approaches (except on a very superficial level with text mining approaches) [15,143,164]. These articles then undergo peer reviewing, which is typically done in a way that is secretive and not standardized, with the effect that the reviewing process may lack transparency and that valuable comments from the reviewers cannot be reused or built upon.…”

Section: Introductionmentioning

confidence: 99%

Linkflows: Towards Genuine Semantic Publishing in Science

Bucur

View full text Add to dashboard Cite

Scientific publishing is the main way in which novel research ideas,evidence,data,and scientific results are communicated,shared and assessed.Currently this process of scientific publishing lacks, most of the time,transparency and machine-interpretable representations.As such,scientific publication is still done in scientific articles,basically long coarse-grained text with complicated structures in natural language that are optimized for human readers and not for automated systems.Moreover, peer reviewing continues to be the main method of quality assessment in science, despite the many and serious flaws emphasized by the scientific community like the lack of transparency,accuracy and efficiency of such a practice.And,as science is rapidly changing and moving more towards a digital environment,with scientific contributions increasing in volume and complexity each day,scientific publishing that still follows this old paradigm of publishing seems to be at odds with the current scientific progress. In this thesis,we propose and evaluate models that tackle how scientific articles and their findings together with their reviews are written and published and how these models can be used in concrete applications that can assist the publishing process for humans and machines alike.First,we introduce a novel fine-grained semantically rich model for reviews in which reviews consist of review comments that contain formal links and semantics from the start and as such are able to make the reviewing process better organized and more accurate.Next,we use this fine-grained reviewing model in conjunction with semantic representations of all the elements of publications and their structure together with all the involved processes,actors and provenance to represent in a unified semantic model the scientific publishing process. Then,we focus on finding a semantic pattern that is able to formally represent the content of high-level scientific findings in a way that can be automatically interpreted.Afterwards,we test these models and applications on a multitude of scientific publication roles such as reviewers, editors,authors and eventually readers in a field study where we prove that not only the scientific publication process (with submissions and final articles) can be represented with formal semantics from the start,but also the whole process in between,including reviews,responses and decisions. Altogether,our research shows that it is possible to bring scientific publishing closer to automated systems and make the elements of publications and their assessments,along with their corresponding processes more machine-interpretable,especially if semantics are captured from the start.The complementary models proposed can be used in the publishing practice and can be adequately used in tools that support the publishing activities and workflows.And,regardless of the challenges encountered along these processes,after reasonable guidance,users are able to use these tools and models with ease despite their novelty, demonstrating that overall our ideas work.This can be a first step in the direction of involving machines more in the research community,where a new way of publishing that considers the Semantic Web principles can help in creating a society where computers can assist the Open Science research community become more efficient and make scientific contributions FAIR: easier to find,more accessible,interoperable and reusable.

show abstract

Multiple sampling schemes and deep learning improve active learning performance in drug-drug interaction information retrieval analysis from the literature

et al. 2023

View full text Add to dashboard Cite

Background Drug-drug interaction (DDI) information retrieval (IR) is an important natural language process (NLP) task from the PubMed literature. For the first time, active learning (AL) is studied in DDI IR analysis. DDI IR analysis from PubMed abstracts faces the challenges of relatively small positive DDI samples among overwhelmingly large negative samples. Random negative sampling and positive sampling are purposely designed to improve the efficiency of AL analysis. The consistency of random negative sampling and positive sampling is shown in the paper. Results PubMed abstracts are divided into two pools. Screened pool contains all abstracts that pass the DDI keywords query in PubMed, while unscreened pool includes all the other abstracts. At a prespecified recall rate of 0.95, DDI IR analysis precision is evaluated and compared. In screened pool IR analysis using supporting vector machine (SVM), similarity sampling plus uncertainty sampling improves the precision over uncertainty sampling, from 0.89 to 0.92 respectively. In the unscreened pool IR analysis, the integrated random negative sampling, positive sampling, and similarity sampling improve the precision over uncertainty sampling along, from 0.72 to 0.81 respectively. When we change the SVM to a deep learning method, all sampling schemes consistently improve DDI AL analysis in both screened pool and unscreened pool. Deep learning has significant improvement of precision over SVM, 0.96 vs. 0.92 in screened pool, and 0.90 vs. 0.81 in the unscreened pool, respectively. Conclusions By integrating various sampling schemes and deep learning algorithms into AL, the DDI IR analysis from literature is significantly improved. The random negative sampling and positive sampling are highly effective methods in improving AL analysis where the positive and negative samples are extremely imbalanced.

show abstract

A Text Mining Protocol for Predicting Drug–Drug Interaction and Adverse Drug Reactions from PubMed Articles

Cited by 4 publications

References 27 publications

Nanopublication-based semantic publishing and reviewing: a field study with formalization papers

Nanopublication-based semantic publishing and reviewing: a field study with formalization papers

Linkflows: Towards Genuine Semantic Publishing in Science

Multiple sampling schemes and deep learning improve active learning performance in drug-drug interaction information retrieval analysis from the literature

Contact Info

Product

Resources

About